RF Toolbox™

User's Guide

R2024b
How to Contact MathWorks

Latest news: www.mathworks.com

Sales and services: www.mathworks.com/sales_and_services

User community: www.mathworks.com/matlabcentral

Technical support: www.mathworks.com/support/contact_us

Phone: 508-647-7000

The MathWorks, Inc.

1 Apple Hill Drive
Natick, MA 01760-2098
RF Toolbox™ User's Guide
© COPYRIGHT 2004–2024 by The MathWorks, Inc.
The software described in this document is furnished under a license agreement. The software may be used or copied
only under the terms of the license agreement. No part of this manual may be photocopied or reproduced in any form
without prior written consent from The MathWorks, Inc.
FEDERAL ACQUISITION: This provision applies to all acquisitions of the Program and Documentation by, for, or through
the federal government of the United States. By accepting delivery of the Program or Documentation, the government
hereby agrees that this software or documentation qualifies as commercial computer software or commercial computer
software documentation as such terms are used or defined in FAR 12.212, DFARS Part 227.72, and DFARS 252.227-7014.
Accordingly, the terms and conditions of this Agreement and only those rights specified in this Agreement, shall pertain
to and govern the use, modification, reproduction, release, performance, display, and disclosure of the Program and
Documentation by the federal government (or other entity acquiring for or through the federal government) and shall
supersede any conflicting contractual terms or conditions. If this License fails to meet the government's needs or is
inconsistent in any respect with federal procurement law, the government agrees to return the Program and
Documentation, unused, to The MathWorks, Inc.
Trademarks
MATLAB and Simulink are registered trademarks of The MathWorks, Inc. See
www.mathworks.com/trademarks for a list of additional trademarks. Other product or brand names may be
trademarks or registered trademarks of their respective holders.
Patents
MathWorks products are protected by one or more U.S. patents. Please see www.mathworks.com/patents for
more information.
Revision History
June 2004 Online only New for Version 1.0 (Release 14)
August 2004 Online only Revised for Version 1.0.1 (Release 14+)
March 2005 Online only Revised for Version 1.1 (Release 14SP2)
September 2005 Online only Revised for Version 1.2 (Release 14SP3)
March 2006 Online only Revised for Version 1.3 (Release 2006a)
September 2006 Online only Revised for Version 2.0 (Release 2006b)
March 2007 Online only Revised for Version 2.1 (Release 2007a)
September 2007 Online only Revised for Version 2.2 (Release 2007b)
March 2008 Online only Revised for Version 2.3 (Release 2008a)
October 2008 Online only Revised for Version 2.4 (Release 2008b)
March 2009 Online only Revised for Version 2.5 (Release 2009a)
September 2009 Online only Revised for Version 2.6 (Release 2009b)
March 2010 Online only Revised for Version 2.7 (Release 2010a)
September 2010 Online only Revised for Version 2.8 (Release 2010b)
April 2011 Online only Revised for Version 2.8.1 (Release 2011a)
September 2011 Online only Revised for Version 2.9 (Release 2011b)
March 2012 Online only Revised for Version 2.10 (Release 2012a)
September 2012 Online only Revised for Version 2.11 (Release 2012b)
March 2013 Online only Revised for Version 2.12 (Release 2013a)
September 2013 Online only Revised for Version 2.13 (Release 2013b)
March 2014 Online only Revised for Version 2.14 (Release 2014a)
October 2014 Online only Revised for Version 2.15 (Release 2014b)
March 2015 Online only Revised for Version 2.16 (Release 2015a)
September 2015 Online only Revised for Version 2.17 (Release 2015b)
March 2016 Online only Revised for Version 3.0 (Release 2016a)
September 2016 Online only Revised for Version 3.1 (Release 2016b)
March 2017 Online only Revised for Version 3.2 (Release 2017a)
September 2017 Online only Revised for Version 3.3 (Release 2017b)
March 2018 Online only Revised for Version 3.4 (Release 2018a)
September 2018 Online only Revised for Version 3.5 (Release 2018b)
March 2019 Online only Revised for Version 3.6 (Release 2019a)
September 2019 Online only Revised for Version 3.7 (Release 2019b)
March 2020 Online only Revised for Version 3.8 (Release 2020a)
September 2020 Online only Revised for Version 4.0 (Release 2020b)
March 2021 Online only Revised for Version 4.1 (Release 2021a)
September 2021 Online only Revised for Version 4.2 (Release 2021b)
March 2022 Online only Revised for Version 4.3 (Release 2022a)
September 2022 Online only Revised for Version 4.4 (Release 2022b)
March 2023 Online only Revised for Version 4.5 (Release 2023a)
September 2023 Online only Revised for Version 23.2 (R2023b)
March 2024 Online only Revised for Version 24.1 (R2024a)
September 2024 Online only Revised for Version 24.2 (R2024b)
 Contents

RF Objects
1
RF Data Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
Types of Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
Available Data Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
Data Object Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3

RF Circuit Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4

Overview of RF Circuit Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
Components Versus Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
Available Components and Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5
Circuit Object Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-6

RF Model Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8

Overview of RF Model Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8
Available Model Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8
Model Object Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8

RF Network Parameter Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-10

Overview of Network Parameter Objects . . . . . . . . . . . . . . . . . . . . . . . . . 1-10
Available Network Parameter Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-10
Network Parameter Object Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-10

Model an RF Component
2
Create RF Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
Construct a New Object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
Copy an Existing Object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3

Specify or Import Component Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4

RF Object Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
Set Property Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
Import Property Values from Data Files . . . . . . . . . . . . . . . . . . . . . . . . . . 2-6
Use Data Objects to Specify Circuit Properties . . . . . . . . . . . . . . . . . . . . . 2-8
Retrieve Property Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-9
Reference Properties Directly Using Dot Notation . . . . . . . . . . . . . . . . . 2-11

Specify Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-12

Available Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-12
Set Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-12
Display Available Operating Condition Values . . . . . . . . . . . . . . . . . . . . . 2-12

v
 Process File Data for Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14
Convert Single-Ended S-Parameters to Mixed-Mode S-Parameters . . . . . 2-14
Extract M-Port S-Parameters from N-Port S-Parameters . . . . . . . . . . . . . 2-15
Cascade N-Port S-Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-16

Analyze and Plot RF Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-18

Analyze Networks in Frequency Domain . . . . . . . . . . . . . . . . . . . . . . . . . 2-18
Visualize Component and Network Data . . . . . . . . . . . . . . . . . . . . . . . . . 2-18
Compute and Plot Time-Domain Specifications . . . . . . . . . . . . . . . . . . . . 2-23

Export Component Data to File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-26

Available Export Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-26
How to Export Object Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-26
Export Object Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-27

Basic Operations with RF Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-28

Export Verilog-A Models

3
Model RF Objects Using Verilog-A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
Behavioral Modeling Using Verilog-A . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
Supported Verilog-A Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2

Export Verilog-A Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4

Represent Circuit Object with Model Object . . . . . . . . . . . . . . . . . . . . . . . 3-4
Write Verilog-A Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5

AMP File Format

4
AMP File Data Sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
Denoting Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
Data Sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
S, Y, or Z Network Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
Noise Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4
Noise Figure Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5
Power Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-6
IP3 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8
Inconsistent Data Sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-9

vi Contents
 How Tos, Definitions, Algorithms
5
Determining Parameter Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
Primary and Secondary Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
Determining Formats for One Parameter . . . . . . . . . . . . . . . . . . . . . . . . . 5-3
Determining Formats for Multiple Parameters . . . . . . . . . . . . . . . . . . . . . 5-3

RF and Microwave Filter Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4

Design Workflows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4

Designing Matching Networks in RF Systems . . . . . . . . . . . . . . . . . . . . . . 5-6

Available Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-6
Design Workflows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-6

RF Toolbox Examples
6
Superheterodyne Receiver Using RF Budget Analyzer App . . . . . . . . . . . . 6-2

Visualizing RF Budget Analysis over Bandwidth . . . . . . . . . . . . . . . . . . . 6-14

Bandpass Filter Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-21

MOS Interconnect and Crosstalk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-27

Bandpass Filter Response Using RFCKT Objects . . . . . . . . . . . . . . . . . . . 6-33

MOS Interconnect and Crosstalk Using RFCKT Objects . . . . . . . . . . . . . 6-39

Model and Analyze High-Speed Backplanes . . . . . . . . . . . . . . . . . . . . . . . 6-47

Use S-parameters with Port Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-49

Fit S-Parameters with a Rational Function . . . . . . . . . . . . . . . . . . . . . . . 6-53

Compute Differential TDR and TDT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-60

Build Simulink Model from Rational Function . . . . . . . . . . . . . . . . . . . . . 6-63

Export Verilog-A module from Rational Function . . . . . . . . . . . . . . . . . . . 6-67

Using 'NPoles' Parameter With rationalfit . . . . . . . . . . . . . . . . . . . . . . . . 6-71

Using 'Weight' Parameter With rationalfit . . . . . . . . . . . . . . . . . . . . . . . . 6-75

Improve Quality of Fit Using 'DelayFactor' Parameter . . . . . . . . . . . . . . . 6-81

Data Analysis on S-Parameters of RF Data Files . . . . . . . . . . . . . . . . . . . 6-85

vii
 Write S2P Touchstone Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-92

Visualize Mixer Spurs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-95

Finding Free IF Bandwidths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-101

De-Embedding S-Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-110

Bisect S-Parameters of Cascaded Probes . . . . . . . . . . . . . . . . . . . . . . . . 6-114

Designing Matching Networks for Low Noise Amplifiers . . . . . . . . . . . . 6-119

Designing Matching Networks Using Single Stub Transmission Lines 6-129

Design Broadband Matching Networks for Antennas . . . . . . . . . . . . . . 6-139

Design Broadband Matching Networks for Amplifier . . . . . . . . . . . . . . 6-148

Impedance Matching of Small Monopole Antenna . . . . . . . . . . . . . . . . . 6-166

Operations with RF Circuit Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-173

Operations with RF Data Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-179

Design IF Butterworth Bandpass Filter . . . . . . . . . . . . . . . . . . . . . . . . . . 6-183

Test, Visualize, and Enforce Passivity of Rational Output . . . . . . . . . . . 6-187

Design, Visualize and Explore Inverse Chebyshev Filter - I . . . . . . . . . . 6-193

Design, Visualize and Explore Inverse Chebyshev Filter - II . . . . . . . . . 6-198

Design Matching Networks for Passive Multiport Network . . . . . . . . . . 6-204

Frequency Sweep in RF Budget Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 6-212

Rational Fit S-Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-214

Design Two-Stage Low Noise Amplifier Using Microstrip Transmission

Line Matching Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-218

RF Budget Harmonic Balance Analysis of Low-IF Receiver, IP2 and NF

........................................................ 6-225

Analysis of Coplanar Waveguide Transmission Line in X-Band Application

........................................................ 6-231

Extract S-Parameters from Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-236

Extract S-Parameters from Mutual Inductor . . . . . . . . . . . . . . . . . . . . . 6-240

Lossy Multiconductor Transmission Line Circuit . . . . . . . . . . . . . . . . . . 6-245

Richards-Kuroda Workflow for RF Filter Circuit . . . . . . . . . . . . . . . . . . 6-252

viii Contents
Design RF Chain Using RF Antenna Object . . . . . . . . . . . . . . . . . . . . . . 6-261

Circuit Envelope Simulation at MATLAB Command Line . . . . . . . . . . . 6-273

Design Matching Network Using Lumped Components from Modelithics

Library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-276

Perform Frequency Planning to Find Spur-Free IF Bandwidths . . . . . . 6-288

Characterize High-Frequency Data and Interconnects Using IEEE P370

Standard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-298

Design Input Matching Network for Dipole Antenna . . . . . . . . . . . . . . . 6-321

Model RF Impairments in 5G NR Downlink Waveform . . . . . . . . . . . . . 6-325

ix
 1

RF Objects

• “RF Data Objects” on page 1-2

• “RF Circuit Objects” on page 1-4
• “RF Model Objects” on page 1-8
• “RF Network Parameter Objects” on page 1-10
1 RF Objects

RF Data Objects
In this section...
“Overview” on page 1-2
“Types of Data” on page 1-2
“Available Data Objects” on page 1-2
“Data Object Methods” on page 1-3

Overview
RF Toolbox software uses data (rfdata) objects to store:

• Component data created from files or from information that you specify in the MATLAB®
workspace.
• Analyzed data from a frequency-domain simulation of a circuit object.

You can perform basic tasks, such as plotting and network parameter conversion, on the data stored
in these objects. However, data objects are primarily used to store data for use by other RF objects.

Types of Data
The toolbox uses RF data objects to store one or more of the following types of data:

• Network parameters
• Spot noise
• Noise figure
• Third-order intercept point (IP3)
• Power out versus power in

Available Data Objects

The following table lists the available rfdata object constructors and describes the data the
corresponding objects represent. For more information on a particular object, follow the link in the
table to the reference page for that object.

Constructor Description
rfdata.data Data object containing network parameter data
rfdata.ip3 Data object containing IP3 information
rfdata.mixerspur Data object containing mixer spur information from an
intermodulation table
rfdata.network Data object containing network parameter information
rfdata.nf Data object containing noise figure information
rfdata.noise Data object containing noise information
rfdata.power Data object containing power and phase information

1-2
 RF Data Objects

Data Object Methods

The following table lists the methods of the data objects, the types of objects on which each can act,
and the purpose of each method.

Method Types of Objects Purpose

extract rfdata.data, Extract specified network parameters from a
rfdata.network circuit or data object and return the result in an
array
read rfdata.data Read RF data parameters from a file to a new or
existing data object.
write rfdata.data Write RF data from a data object to a file.

See Also

More About
• “RF Analysis”
• “RF Circuit Objects” on page 1-4
• “RF Model Objects” on page 1-8
• “RF Network Parameter Objects” on page 1-10

1-3
1 RF Objects

RF Circuit Objects

In this section...
“Overview of RF Circuit Objects” on page 1-4
“Components Versus Networks” on page 1-4
“Available Components and Networks” on page 1-5
“Circuit Object Methods” on page 1-6

Overview of RF Circuit Objects

RF Toolbox software uses circuit (rfckt) objects to represent the following components:

• Circuit components such as amplifiers, transmission lines, and ladder filters

• RLC network components
• Networks of RF components

The toolbox represents each type of component and network with a different object. You use these
objects to analyze components and networks in the frequency domain.

Components Versus Networks

You define component behavior using network parameters and physical properties.

To specify an individual RF component:

1 Construct a circuit object to represent the component.

2 Specify or import component data.

You define network behavior by specifying the components that make up the network. These
components can be either individual components (such as amplifiers and transmission lines) or other
networks.

To specify an RF network:

1 Build circuit objects to represent the network components.

2 Construct a circuit object to represent the network.

Note This object defines how to connect the network components. However, the network is
empty until you specify the components that it contains.
3 Specify, as the Ckts property of the object that represents the network, a list of components that
make up the network.

These procedures are illustrated by example in “Model Cascaded Network”.

1-4
 RF Circuit Objects

Available Components and Networks

To create circuit objects that represent components, you use constructors whose names describe the
components. To create circuit objects that represent networks, you use constructors whose names
describe how the components are connected together.

The following table lists the available rfckt object constructors and describes the components or
networks the corresponding objects represent. For more information on a particular object, follow the
link in the table to the reference page for that object.

Constructor Description
rfckt.amplifier Amplifier, described by an rfdata object
rfckt.cascade Cascaded network, described by the list of components and
networks that comprise it
rfckt.coaxial Coaxial transmission line, described by dimensions and
electrical characteristics
rfckt.cpw Coplanar waveguide transmission line, described by dimensions
and electrical characteristics
rfckt.datafile General circuit, described by a data file
rfckt.delay Delay line, described by loss and delay
rfckt.hybrid Hybrid connected network, described by the list of components
and networks that comprise it
rfckt.hybridg Inverse hybrid connected network, described by the list of
components and networks that comprise it
rfckt.lcbandpasspi LC bandpass pi network, described by LC values
rfckt.lcbandpasstee LC bandpass tee network, described by LC values
rfckt.lcbandstoppi LC bandstop pi network, described by LC values
rfckt.lcbandstoptee LC bandstop tee network, described by LC values
rfckt.lchighpasspi LC highpass pi network, described by LC values
rfckt.lchighpasstee LC highpass tee network, described by LC values
rfckt.lclowpasspi LC lowpass pi network, described by LC values
rfckt.lclowpasstee LC lowpass tee network, described by LC values
rfckt.microstrip Microstrip transmission line, described by dimensions and
electrical characteristics
rfckt.mixer Mixer, described by an rfdata object
rfckt.parallel Parallel connected network, described by the list of components
and networks that comprise it
rfckt.parallelplate Parallel-plate transmission line, described by dimensions and
electrical characteristics
rfckt.passive Passive component, described by network parameters
rfckt.rlcgline RLCG transmission line, described by RLCG values
rfckt.series Series connected network, described by the list of components
and networks that comprise it

1-5
1 RF Objects

Constructor Description
rfckt.seriesrlc Series RLC network, described by RLC values
rfckt.shuntrlc Shunt RLC network, described by RLC values
rfckt.twowire Two-wire transmission line, described by dimensions and
electrical characteristics
rfckt.txline General transmission line, described by dimensions and
electrical characteristics

Circuit Object Methods

The following table lists the methods of the circuit objects, the types of objects on which each can act,
and the purpose of each method.

Method Types of Objects Purpose

analyze All circuit objects Analyze a circuit object in the frequency
domain.
calculate All circuit objects Calculate specified parameters for a circuit
object.
copy All circuit objects Copy a circuit or data object.
extract All circuit objects Extract specified network parameters from a
circuit or data object, and return the result in
an array.
getdata All circuit objects Get data object containing analyzed result of a
specified circuit object.
getz0 rfckt.txline, Get characteristic impedance of a
rfckt.rlcgline, transmission line.
rfckt.twowire,
rfckt.parallelplate,
rfckt.coaxial,
rfdata.microstrip,
rfckt.cpw
listformat All circuit objects List valid formats for a specified circuit object
parameter.
listparam All circuit objects List valid parameters for a specified circuit
object.
loglog All circuit objects Plot specified circuit object parameters using
a log-log scale.
plot All circuit objects Plot the specified circuit object parameters on
an X-Y plane.
plotyy All circuit objects Plot the specified object parameters with y-
axes on both the left and right sides.
polar All circuit objects Plot the specified circuit object parameters on
polar coordinates.

1-6
 RF Circuit Objects

Method Types of Objects Purpose

read rfckt.datafile, Read RF data from a file to a new or existing
rfckt.passive, circuit object.
rfckt.amplifier,
rfckt.mixer
restore rfckt.datafile, Restore data to original frequencies of
rfckt.passive, NetworkData for plotting.
rfckt.amplifier,
rfckt.mixer
semilogx All circuit objects Plot the specified circuit object parameters
using a log scale for the X-axis
semilogy All circuit objects Plot the specified circuit object parameters
using a log scale for the Y-axis
smith All circuit objects Plot the specified circuit object parameters on
a Smith chart.
write All circuit objects Write RF data from a circuit object to a file.
smithplot All circuit objects Plot measurement data on Smith chart

See Also

More About
• “RF Model Objects” on page 1-8
• “RF Analysis”
• “RF Data Objects” on page 1-2
• “RF Network Parameter Objects” on page 1-10

1-7
1 RF Objects

RF Model Objects
In this section...
“Overview of RF Model Objects” on page 1-8
“Available Model Objects” on page 1-8
“Model Object Methods” on page 1-8

Overview of RF Model Objects

RF Toolbox software uses model (rfmodel) objects to represent components and measured data
mathematically for computing information such as time-domain response. Each type of model object
uses a different mathematical model to represent the component.

RF model objects provide a high-level component representation for use after you perform detailed
analysis using RF circuit objects. Use RF model objects to:

• Compute time-domain figures of merit for RF components

• Export Verilog-A models of RF components

Available Model Objects

The following table lists the available rfmodel object constructors and describes the model the
corresponding objects use. For more information on a particular object, follow the link in the table to
the reference page for that object.

Constructor Description
rfmodel.rational Rational function model

Model Object Methods

The following table lists the methods of the model objects, the types of objects on which each can act,
and the purpose of each method.

Method Types of Objects Purpose

freqresp All model objects Compute the frequency response of a model
object.
timeresp All model objects Compute the time response of a model object.
write All model objects Write data from a model object to a file.

See Also

More About
• “RF Analysis”
• “RF Data Objects” on page 1-2
• “RF Circuit Objects” on page 1-4

1-8
 RF Model Objects

• “RF Network Parameter Objects” on page 1-10

1-9
1 RF Objects

RF Network Parameter Objects

In this section...
“Overview of Network Parameter Objects” on page 1-10
“Available Network Parameter Objects” on page 1-10
“Network Parameter Object Functions” on page 1-10

Overview of Network Parameter Objects

RF Toolbox software offers network parameter objects for:

• Importing network parameter data from a Touchstone file.

• Converting network parameters.
• Analyzing network parameter data.

Unlike circuit, model, and data objects, you can use existing RF Toolbox functions to operate directly
on network parameter objects.

Available Network Parameter Objects

The following table lists the available network parameter objects and the functions that are used to
construct them. For more information on a particular object, follow the link in the table to the
reference page for that functions.

Network Parameter Object Type Network Parameter Object Function

ABCD Parameter object abcdparameters
Hybrid-g parameter object gparameters
Hybrid parameter object hparameters
S-parameter object sparameters
Y-parameter object yparameters
Z-parameter object zparameters

Network Parameter Object Functions

The following table lists the functions that accept network parameter objects as inputs, the types of
objects on which each can act, and the purpose of each function.

Function Types of Objects Purpose

abcdparameters All network parameter objects Convert any network
parameters to ABCD
parameters
gparameters All network parameter objects Convert any network
parameters to hybrid-g
parameters

1-10
 RF Network Parameter Objects

Function Types of Objects Purpose

hparameters All network parameter objects Convert any network
parameters to hybrid
parameters
sparameters All network parameter objects Convert any network
parameters to S-parameters
yparameters All network parameter objects Convert any network
parameters to Y-parameters
zparameters All network parameter objects Convert any network
parameters to Z-parameters
cascadesparams S-parameter objects Cascade S-parameters
deembedsparams S-parameter objects De-embed S-parameters
gammain S-parameter objects Calculate input reflection
coefficient
gammaml S-parameter objects Calculate load reflection
coefficient
gammams S-parameter objects Calculate source reflection
coefficient
gammaout S-parameter objects Calculate output reflection
coefficient
ispassive S-parameter objects Check S-parameter data
passivity
makepassive S-parameter objects Make S-parameter data passive
newref S-parameter objects Change reference impedance
powergain S-parameter objects Calculate power gain
rfplot S-parameter objects Plot network parameters
rfinterp1 All network parameter objects Interpolate network parameters
at new frequencies
rfparam All network parameter objects Extract vector of network
parameters
s2tf S-parameter objects Create transfer function from S-
parameters
stabilityk S-parameter objects Calculate stability factor K of 2-
port network
stabilitymu S-parameter objects Calculate stability factor μ of 2-
port network
smith All network parameter objects Plot network parameter data on
a Smith® Chart
smithplot All network parameter objects Plot measurement data on
Smith chart

1-11
1 RF Objects

See Also

More About
• “RF Data Objects” on page 1-2
• “RF Circuit Objects” on page 1-4
• “RF Model Objects” on page 1-8
• “S-Parameter Notation”

1-12
 2

Model an RF Component

• “Create RF Objects” on page 2-2

• “Specify or Import Component Data” on page 2-4
• “Specify Operating Conditions” on page 2-12
• “Process File Data for Analysis” on page 2-14
• “Analyze and Plot RF Components” on page 2-18
• “Export Component Data to File” on page 2-26
• “Basic Operations with RF Objects” on page 2-28
2 Model an RF Component

Create RF Objects
In this section...
“Construct a New Object” on page 2-2
“Copy an Existing Object” on page 2-3

Construct a New Object

You can create any rfdata, rfckt or rfmodel object by calling the object constructor. You can
create an rfmodel object by fitting a rational function to passive component data.

This section contains the following topics:

• “Call the Object Constructor” on page 2-2

• “Fit a Rational Function to Passive Component Data” on page 2-3

Call the Object Constructor

To create a new RF object with default property values, you call the object constructor without any
arguments:

h = objecttype.objectname

where:

• h is the handle to the new object.

• objecttype is the object type (rfdata, rfckt, or rfmodel).
• objectname is the object name.

For example, to create an RLCG transmission line object, type:

h = rfckt.rlcgline

because the RLCG transmission line object is a circuit (rfckt) object named rlcgline.

The following code illustrates how to call the object constructor to create a microstrip transmission
line object with default property values. The output t1 is the handle of the newly created
transmission line object.

t1 = rfckt.microstrip

RF Toolbox software lists the properties of the transmission line you created along with the
associated default property values.

t1 =
Name: 'Microstrip Transmission Line'
nPort: 2
AnalyzedResult: []
LineLength: 0.0100
StubMode: 'NotAStub'
Termination: 'NotApplicable'
Width: 6.0000e-004
Height: 6.3500e-004

2-2
 Create RF Objects

Thickness: 5.0000e-006
EpsilonR: 9.8000
SigmaCond: Inf
LossTangent: 0

The reference page describes these properties in detail, rfckt.microstrip.

Fit a Rational Function to Passive Component Data

You can create a model object by fitting a rational function to passive component data. You use this
approach to create a model object that represents one of the following using a rational function:

• A circuit object that you created and analyzed.

• Data that you imported from a file.

For more information, see “Fit Model Object to Circuit Object Data” on page 2-24.

Copy an Existing Object

You can create a new object with the same property values as an existing object by using the copy
function to copy the existing object. This function is useful if you have an object that is similar to one
you want to create.

For example,

t2 = copy(t1);

creates a new object, t2, which has the same property values as the microstrip transmission line
object, t1.

You can later change specific property values for this copy. For information on modifying object
properties, see “Specify or Import Component Data” on page 2-4.

Note The syntax t2 = t1 copies only the object handle and does not create a new object.

See Also

More About
• “Process File Data for Analysis” on page 2-14

2-3
2 Model an RF Component

Specify or Import Component Data

In this section...
“RF Object Properties” on page 2-4
“Set Property Values” on page 2-4
“Import Property Values from Data Files” on page 2-6
“Use Data Objects to Specify Circuit Properties” on page 2-8
“Retrieve Property Values” on page 2-9
“Reference Properties Directly Using Dot Notation” on page 2-11

RF Object Properties
Object properties specify the behavior of an object. You can specify object properties, or you can
import them from a data file. To learn about properties that are specific to a particular type of circuit,
data, or model object, see the reference page for that type of object.

Note The “RF Circuit Objects” on page 1-4, “RF Data Objects” on page 1-2,“RF Model Objects” on
page 1-8 sections list the available types of objects and provide links to their reference pages.

Set Property Values

You can specify object property values when you construct an object or you can modify the property
values of an existing object.

This section contains the following topics:

• “Specify Property Values at Construction” on page 2-4

• “Change Property Values of an Existing Object” on page 2-5

Specify Property Values at Construction

To set a property when you construct an object, include a comma-separated list of one or more
property/value pairs in the argument list of the object construction command. A property/value pair
consists of the arguments 'PropertyName',PropertyValue, where:

• PropertyName is a character vector specifying the property name. The name is case-insensitive.
In addition, you need only type enough letters to uniquely identify the property name. For
example, 'st' is sufficient to refer to the StubMode property.

Note You must use single quotation marks around the property name.
• PropertyValue is the value to assign to the property.

Include as many property names in the argument list as there are properties you want to set. Any
property values that you do not set retain their default values. The circuit and data object reference
pages list the valid values as well as the default value for each property.

This section contains examples of how to perform the following tasks:

2-4
 Specify or Import Component Data

• “Construct Components with Specified Properties” on page 2-5

• “Construct Networks of Specified Components” on page 2-5
Construct Components with Specified Properties

The following code creates a coaxial transmission line circuit object to represent a coaxial
transmission line that is 0.05 meters long. Notice that the toolbox lists the available properties and
their values.
t1 = rfckt.coaxial('LineLength',0.05)

t1 =

Name: 'Coaxial Transmission Line'

nPort: 2
AnalyzedResult: []
LineLength: 0.0500
StubMode: 'NotAStub'
Termination: 'NotApplicable'
OuterRadius: 0.0026
InnerRadius: 7.2500e-004
MuR: 1
EpsilonR: 2.3000
LossTangent: 0
SigmaCond: Inf

Construct Networks of Specified Components

To combine a set of RF components and existing networks to form an RF network, you create a
network object with the Ckts property set to an array containing the handles of all the circuit objects
in the network.

Suppose you have the following RF components:

t1 = rfckt.coaxial('LineLength',0.05);
a1 = rfckt.amplifier;
t2 = rfckt.coaxial('LineLength',0.1);

The following code creates a cascaded network of these components:

casc_network = rfckt.cascade('Ckts',{t1,a1,t2});

Change Property Values of an Existing Object

There are two ways to change the properties of an existing object:

• Using the set command

• Using structure-like assignments called dot notation

This section discusses the first option. For details on the second option, see “Reference Properties
Directly Using Dot Notation” on page 2-11.

To modify the properties of an existing object, use the set command with one or more property/value
pairs in the argument list. The general syntax of the command is
set(h,Property1',value1,'Property2',value2,...)

where

2-5
2 Model an RF Component

• h is the handle of the object.

• 'Property1',value1,'Property2',value2,... is the list of property/value pairs.

For example, the following code creates a default coaxial transmission line object and changes it to a
series stub with open termination.

t1 = rfckt.coaxial;
set(t1,'StubMode','series','Termination','open')

Note You can use the set command without specifying any property/value pairs to display a list of
all properties you can set for a specific object. This example lists the properties you can set for the
coaxial transmission line t1:

set(t1)

ans =
LineLength: {}
StubMode: {}
Termination: {}
OuterRadius: {}
InnerRadius: {}
MuR: {}
EpsilonR: {}
LossTangent: {}
SigmaCond: {}

Import Property Values from Data Files

RF Toolbox software lets you import industry-standard data files, MathWorks® AMP files, and
Agilent® P2D and S2D files into specific objects. This import capability lets you simulate the behavior
of measured components.

You can import the following file formats:

• Industry-standard file formats — Touchstone SNP, YNP, ZNP, HNP, and GNP formats specify the
network parameters and noise information for measured and simulated data.

For more information on Touchstone files, see https://ibis.org/connector/

touchstone_spec11.pdf.
• Agilent P2D file format — Specifies amplifier and mixer large-signal, power-dependent network
parameters, noise data, and intermodulation tables for several operating conditions, such as
temperature and bias values.

The P2D file format lets you import system-level verification models of amplifiers and mixers.
• Agilent S2D file format — Specifies amplifier and mixer network parameters with gain
compression, power-dependent S21 parameters, noise data, and intermodulation tables for several
operating conditions.

The S2D file format lets you import system-level verification models of amplifiers and mixers.
• MathWorks amplifier (AMP) file format — Specifies amplifier network parameters, output power
versus input power, noise data and third-order intercept point.

2-6
 Specify or Import Component Data

For more information about .amp files, see “AMP File Data Sections” on page 4-2.

This section contains the following topics:

• “Objects Used to Import Data from a File” on page 2-7

• “How to Import Data Files” on page 2-7

Objects Used to Import Data from a File

One data object and three circuit objects accept data from a file. The following table lists the objects
and any corresponding data format each supports.

Object Description Supported Format(s)

rfdata.data Data object containing network Touchstone, AMP, P2D, S2D
parameter data, noise figure,
and third-order intercept point
rfckt.amplifier Amplifier Touchstone, AMP, P2D, S2D
rfckt.mixer Mixer Touchstone, AMP, P2D, S2D
rfckt.passive Generic passive component Touchstone

How to Import Data Files

To import file data into a circuit or data object at construction, use a read command of the form:

obj = read(obj_type,'filename');

where

• obj is the handle of the circuit or data object.

• obj_type is the type of object in which to store the data, from the list of objects that accept file
data shown in “Objects Used to Import Data from a File” on page 2-7.
• filename is the name of the file that contains the data.

For example,

ckt_obj=read(rfckt.amplifier, 'default.amp');

imports data from the file default.amp into an rfckt.amplifier object.

You can also import file data into an existing circuit object. The following commands are equivalent to
the previous command:

ckt_obj=rfckt.amplifier;
read(ckt_obj, 'default.amp');

Note When you import component data from a .p2d or .s2d file, properties are defined for several
operating conditions. You must select an operating condition to specify the object behavior, as
described in “Specify Operating Conditions” on page 2-12.

2-7
2 Model an RF Component

Use Data Objects to Specify Circuit Properties

To specify a circuit object property using a data object, use the set command with the name of the
data object as the value in the property/value pair.

For example, suppose you have the following rfckt.amplifier and rfdata.nf objects:

amp = rfckt.amplifier
f = 2.0e9;
nf = 13.3244;
nfdata = rfdata.nf('Freq',f,'Data',nf)

The following command uses the rfdata.nf data object to specify the rfckt.amplifier
NoiseData property:

set(amp,'NoiseData',nfdata)

Set Circuit Object Properties Using Data Objects

In this example, you create a circuit object. Then, you create three data objects and use them to
update the properties of the circuit object.

1 Create an amplifier object. This circuit object, rfckt.amplifier, has a network parameter,
noise data, and nonlinear data properties. These properties control the frequency response of the
amplifier, which is stored in the AnalyzedResult property. By default, all amplifier properties
contain values from the default.amp file. The NetworkData property is an rfdata.network
object that contains 50-ohm S-parameters. The NoiseData property is an rfdata.noise object
that contains frequency-dependent spot noise data. The NonlinearData property is an
rfdata.power object that contains output power and phase information.

amp = rfckt.amplifier

The toolbox displays the following output:

amp =

Name: 'Amplifier'
nPort: 2
AnalyzedResult: [1x1 rfdata.data]
IntpType: 'Linear'
NetworkData: [1x1 rfdata.network]
NoiseData: [1x1 rfdata.noise]
NonlinearData: [1x1 rfdata.power]
2 Create a data object that stores network data. Type the following set of commands at the
MATLAB prompt to create an rfdata.network object that stores the 2-port Y-parameters at
2.08 GHz, 2.10 GHz, and 2.15 GHz. Later in this example, you use this data object to update the
NetworkData property of the rfckt.amplifier object.

f = [2.08 2.10 2.15]*1.0e9;

y(:,:,1) = [-.0090-.0104i, .0013+.0018i; ...
-.2947+.2961i, .0252+.0075i];
y(:,:,2) = [-.0086-.0047i, .0014+.0019i; ...
-.3047+.3083i, .0251+.0086i];
y(:,:,3) = [-.0051+.0130i, .0017+.0020i; ...
-.3335+.3861i, .0282+.0110i];

2-8
 Specify or Import Component Data

netdata = rfdata.network('Type','Y_PARAMETERS',...
'Freq',f,'Data',y)

The toolbox displays the following output:

netdata =

Name: 'Network parameters'

Type: 'Y_PARAMETERS'
Freq: [3x1 double]
Data: [2x2x3 double]
Z0: 50
3 Create a data object that stores noise figure values. Type the following set of commands at
the MATLAB prompt to create a rfdata.nf object that contains noise figure values, in dB, at
seven different frequencies. Later in this example, you use this data object to update the
NoiseData property of the rfckt.amplifier object.

f = [1.93 2.06 2.08 2.10 2.15 2.30 2.40]*1.0e9;

nf=[12.4521 13.2466 13.6853 14.0612 13.4111 12.9499 13.3244];

nfdata = rfdata.nf('Freq',f,'Data',nf)

The toolbox displays the following output:

nfdata =

Name: 'Noise figure'

Freq: [7x1 double]
Data: [7x1 double]
4 Create a data object that stores output third-order intercept points. Type the following
command at the MATLAB prompt to create a rfdata.ip3 object that contains an output third-
order intercept point of 8.45 watts, at 2.1 GHz. Later in this example, you use this data object to
update the NonlinearData property of the rfckt.amplifier object.

ip3data = rfdata.ip3('Type','OIP3','Freq',2.1e9,'Data',8.45)

The toolbox displays the following output:

ip3data =

Name: '3rd order intercept'

Type: 'OIP3'
Freq: 2.1000e+009
Data: 8.4500
5 Update the properties of the amplifier object. Type the following set of commands at the
MATLAB prompt to update the NetworkData, NoiseData, and NonlinearData properties of
the amplifier object with the data objects you created in the previous steps:

amp.NetworkData = netdata;
amp.NoiseData = nfdata;
amp.NonlinearData = ip3data;

Retrieve Property Values

You can retrieve one or more property values of an existing object using the get command.

2-9
2 Model an RF Component

This section contains the following topics:

• “Retrieve Specified Property Values” on page 2-10

• “Retrieve All Property Values” on page 2-10

Retrieve Specified Property Values

To retrieve specific property values for an object, use the get command with the following syntax:
PropertyValue = get(h,PropertyName)

where

• PropertyValue is the value assigned to the property.

• h is the handle of the object.
• PropertyName is a character vector specifying the property name.

For example, suppose you have the following coaxial transmission line:
h2 = rfckt.coaxial;

The following code retrieves the value of the inner radius and outer radius for the coaxial
transmission line:
ir = get(h2,'InnerRadius')
or = get(h2,'OuterRadius')

ir =
7.2500e-004

or =
0.0026

Retrieve All Property Values

To display a list of properties associated with a specific object as well as their current values, use the
get command without specifying a property name.

For example:
get(h2)
Name: 'Coaxial Transmission Line'
nPort: 2
AnalyzedResult: []
LineLength: 0.0100
StubMode: 'NotAStub'
Termination: 'NotApplicable'
OuterRadius: 0.0026
InnerRadius: 7.2500e-004
MuR: 1
EpsilonR: 2.3000
LossTangent: 0
SigmaCond: Inf

Note This list includes read-only properties that do not appear when you type set(h2). For a
coaxial transmission line object, the read-only properties are Name, nPort, and AnalyzedResult.

2-10
 Specify or Import Component Data

The Name and nPort properties are fixed by the toolbox. The AnalyzedResult property value is
calculated and set by the toolbox when you analyze the component at specified frequencies.

Reference Properties Directly Using Dot Notation

An alternative way to query for or modify property values is by structure-like referencing. The field
names for RF objects are the property names, so you can retrieve or modify property values with the
structure-like syntax.

• PropertyValue = rfobj.PropertyName stores the value of the PropertyName property of

the rfobj object in the PropertyValue variable. This command is equivalent to
PropertyValue = get(rfobj,'PropertyName').
• rfobj.PropertyName = PropertyValue sets the value of the PropertyName property to
PropertyValue for the rfobj object. This command is equivalent to
set(rfobj,'PropertyName',PropertyValue).

For example, typing

ckt = rfckt.amplifier('IntpType','cubic');
ckt.IntpType

gives the value of the property IntpType for the circuit object ckt.

ans =
Cubic

Similarly,

ckt.IntpType = 'linear';

resets the interpolation method to linear.

You do not need to type the entire field name or use uppercase characters. You only need to type the
minimum number of characters sufficient to identify the property name uniquely. Thus entering the
commands

ckt = rfckt.amplifier('IntpType','cubic');
ckt.in

also produces

ans =
Cubic

See Also
“AMP File Data Sections” on page 4-2

2-11
2 Model an RF Component

Specify Operating Conditions

In this section...
“Available Operating Conditions” on page 2-12
“Set Operating Conditions” on page 2-12
“Display Available Operating Condition Values” on page 2-12

Available Operating Conditions

Agilent P2D and S2D files contain simulation results at one or more operating conditions. Operating
conditions define the independent parameter settings that are used when creating the file data. The
specified conditions differ from file to file.

When you import component data from a .p2d or .s2d file, the object contains property values for
several operating conditions. The available conditions depend on the data in the file. By default, RF
Toolbox software defines the object behavior using the property values that correspond to the
operating conditions that appear first in the file. To use other property values, you must select a
different operating condition.

Set Operating Conditions

To set the operating conditions of a circuit or data object, use a setop command of the form:

setop(,'Condition1',value1,...,'ConditionN',valueN,...)

where

• is the handle of the circuit or data object.

• Condition1,value1,...,ConditionN,valueN are the condition/value pairs that specify the
operating condition.

For example,

setop(myp2d, 'BiasL', 2, 'BiasU', 6.3)

specifies an operating condition of BiasL = 2 and BiasU = 6.3 for myp2d.

Display Available Operating Condition Values

To display a list of available operating condition values for a circuit or data object, use the setop
method.

setop(obj)

displays the available values for all operating conditions of the object obj.

setop(obj,'Condition1')

displays the available values for Condition1.

2-12
 Specify Operating Conditions

See Also

More About
• “S-Parameter Notation”
• “AMP File Data Sections” on page 4-2
• “Determining Parameter Formats” on page 5-2

2-13
2 Model an RF Component

Process File Data for Analysis

In this section...
“Convert Single-Ended S-Parameters to Mixed-Mode S-Parameters” on page 2-14
“Extract M-Port S-Parameters from N-Port S-Parameters” on page 2-15
“Cascade N-Port S-Parameters” on page 2-16

This topic shows you how to process RF data for analysis. After you import your data, you can process
it using one of the following methods:

• Convert single-ended S-parameters to mixed-mode S-parameters.

• Extract M-port S-parameters from N-Port S-parameters.
• Cascade N-port S-parameters

Convert Single-Ended S-Parameters to Mixed-Mode S-Parameters

This section contains the following topics:

• “Functions for Converting S-Parameters” on page 2-14

• “Convert S-Parameters” on page 2-14

Functions for Converting S-Parameters

To convert between 4-port single-ended S-parameter data and 2-port differential-, common-, and
cross-mode S-parameters, use one of these functions:

• s2scc — Convert 4-port, single-ended S-parameters to 2-port, common-mode S-parameters (Scc).

• s2scd — Convert 4-port, single-ended S-parameters to 2-port, cross-mode S-parameters (Scd).
• s2sdc — Convert 4-port, single-ended S-parameters to cross-mode S-parameters (Sdc).
• s2sdd — Convert 4-port, single-ended S-parameters to 2-port, differential-mode S-parameters
(Sdd).

To perform the above conversions all at once, or to convert larger data sets, use one of these
functions:

• s2smm — Convert 4N-port, single-ended S-parameters to 2N-port, mixed-mode S-parameters.

• smm2s — Convert 2N-port, mixed-mode S-parameters to 4N-port, single-ended S-parameters.

Conversion functions support a variety of port orderings. For more information on these functions,
see the corresponding reference pages.

Convert S-Parameters

In this example, use the toolbox to import 4-port single-ended S-parameter data from a file, convert
the data to 2-port differential S-parameter data, and create a new circuit object to store the
converted data for analysis.

At the MATLAB prompt:

1 Type this command to import data from the file default.s4p:

2-14
 Process File Data for Analysis

singleEnded4PortSparam = sparameters('default.s4p');
2 Type these commands to convert 4-port single-ended S-parameters to 2-port mixed-mode S-
parameters:

differential2PortData = s2sdd(singleEnded4PortSparam.Parameters);
differential2PortSparam = sparameters(differential2PortData, ...
singleEnded4PortSparam.Frequencies,2*singleEnded4PortSparam.Impedance);
3 Type this command to create an nport object that stores the 2-port differential S-parameters for
addition to a circuit or rfbudget object:

differential2Port = nport(differential2PortSparam);

Extract M-Port S-Parameters from N-Port S-Parameters

After you import file data (as described in “Import Property Values from Data Files” on page 2-6), you
can extract a set of data with a smaller number of ports by terminating one or more ports with a
specified impedance.

This section contains the following topics:

• “Extract S-Parameters” on page 2-15

• “Extract S-Parameters from Imported File Data” on page 2-15

Extract S-Parameters

To extract M-port S-parameters from N-port S-parameters, use the snp2smp function with the
following syntax: To extract M-port S-parameter data from N-port S-parameter data, use the
snp2smp.

The following figure illustrates how to specify the ports for the output data and the termination of the
remaining ports.

For more details about the arguments to this function, see the snp2smp reference page.

Extract S-Parameters from Imported File Data

In this example, use the toolbox to import 16-port S-parameter data from a file, convert the data to 4-
port S-parameter data by terminating the remaining ports, and create a new sparameters object to
store the extracted data for analysis.

2-15
2 Model an RF Component

At the MATLAB prompt:

1 Type this command to import data from the file default.s16p into an sparameters object,

singleEnded16PortData:singleEnded16PortSparam = sparameters('default.s16p');
2 Type these commands to convert 16-port S-parameters to 4-port S-parameters by using ports 1,
16, 2, and 15 as the first, second, third, and fourth ports, and terminating the remaining 12 ports
with an impedance of 50 ohms:
N2M_index = [1 16 2 15];
fourPortData = snp2smp(singleEnded16PortSparam.Parameters, ...
singleEnded16PortSparam.Impedance, N2M_index, 50);
fourPortSparam = sparameters(fourPortData, ...
singleEnded16PortSparam.Frequencies,singleEnded16PortSparam.Impedance);

Alternatively, apply the snp2smp function to the sparameters object:

fourPortSparam = snp2smp(singleEnded16PortSparam, N2M_index, 50);

Cascade N-Port S-Parameters

Cascade two or more networks of N-port S-parameters using the cascadesparams function.

Import and Cascade N-Port S-Parameters

In this example, use the toolbox to import 16-port and 4-port S-parameter file data and cascade the
two S-parameter networks by connecting the last three ports of the 16-port network to the first three
ports of the 4-port network. Then, create a new sparameters object to store the resulting network
for analysis.

At the MATLAB prompt:

1 Type these commands to import data from the files default.s16p and default.s4p, and create
the 16- and 4-port networks of S-parameters:

S_16Port = sparameters('default.s16p');
S_4Port = sparameters('default.s4p');
freq = [2e9 2.1e9];
sparams_16p = rfinterp1(S_16Port,freq);
sparams_4p = rfinterp1(S_4Port,freq);
2 Type this command to cascade 16-port S-parameters and 4-port S-parameters by connecting
ports 14, 15, and 16 of the 16-port network to ports 1, 2, and 3 of the 4-port network:

sparams_cascaded = ...
cascadesparams(sparams_16p.Parameters, sparams_4p.Parameters,3);

cascadesparams creates a 14-port network. Ports 1–13 are the first 13 ports of the 16-port
network. Port 14 is the fourth port of the 4-port network.
3 Type this command to create an sparameter object that stores the 14-port S-parameters for
simulation:

Ckt14 = sparameters(sparams_cascaded,freq);

2-16
 Process File Data for Analysis

See Also

More About
• “S-Parameter Notation”
• “AMP File Data Sections” on page 4-2
• “Determining Parameter Formats” on page 5-2

2-17
2 Model an RF Component

Analyze and Plot RF Components

In this section...
“Analyze Networks in Frequency Domain” on page 2-18
“Visualize Component and Network Data” on page 2-18
“Compute and Plot Time-Domain Specifications” on page 2-23

Analyze Networks in Frequency Domain

RF Toolbox lets you analyze RF components and networks in the frequency domain. You use the
analyze function to analyze a circuit object over a specified set of frequencies.

For example, to analyze a coaxial transmission line from 1 GHz to 2.9 GHz in increments of 10 MHz:

ckt = rfckt.coaxial;
f = [1.0e9:1e7:2.9e9];
analyze(ckt,f);

Note For all circuits objects except those that contain data from a file, you must perform a
frequency-domain analysis with the analyze method before visualizing component and network
data. For circuits that contain data from a file, the toolbox performs a frequency-domain analysis
when you use the read method to import the data.

When you analyze a circuit object, the toolbox computes the circuit network parameters, noise figure
values, and output third-order intercept point (OIP3) values at the specified frequencies and stores
the result of the analysis in the object's AnalyzedResult property.

For more information, see the analyze function page.

Visualize Component and Network Data

The RF Toolbox lets you validate the behavior of circuit objects that represent RF components and
networks by plotting the following data:

• Large- and small-signal S-parameters

• Noise figure
• Output third-order intercept point
• Power data
• Phase noise
• Voltage standing-wave ratio
• Power gain
• Group delay
• Reflection coefficients
• Stability data
• Transfer function

2-18
 Analyze and Plot RF Components

This table summarizes the available plots and charts, along with the functions you can use to create
each one and a description of its contents.

Plot Type Functions Plot Contents

“Rectangular Plot” on page 2-20 plot Parameters as a function of frequency or,
where applicable, operating condition.
plotyy The available parameters include:

loglog • S-parameters
• Noise figure
semilogx
• Voltage standing-wave ratio (VSWR)
semilogy • OIP3
“Budget Plot” on page 2-20 plot Parameters as a function of frequency for
each component in a cascade, where the
curve for a given component represents
the cumulative contribution of each RF
component up to and including the
parameter value of that component.
“Mixer Spur Plot” on page 2-21 plot Mixer spur power as a function of
frequency for an rfckt.mixer object or
an rfckt.cascade object that contains
a mixer.
“Polar Plots and Smith Charts” on polar Polar plot: Magnitude and phase of S-
page 2-22 parameters as a function of frequency.
smithplot
Smith plot: Real and imaginary parts of
S-parameters as a function of frequency,
used for analyzing the reflections caused
by impedance mismatch.

For each plot you create, you choose a parameter to plot and, optionally, a format in which to plot
that parameter. The plot format defines how the RF Toolbox displays the data on the plot. The
available formats vary with the data you select to plot. The data you can plot depends on the type of
plot you create.

Note You can use the listparam function to list the parameters of a specified circuit object that are
available for plotting. You can use the listformat function to list the available formats for a
specified circuit object parameter.

The following topics describe the available plots:

• “Rectangular Plot” on page 2-20

• “Budget Plot” on page 2-20
• “Mixer Spur Plot” on page 2-21
• “Polar Plots and Smith Charts” on page 2-22

2-19
2 Model an RF Component

Rectangular Plot

You can plot any parameters that are relevant to your object on a rectangular plot. You can plot
parameters as a function of frequency for any object. When you import object data from a .p2d
or .s2d file, you can also plot parameters as a function of any operating condition from the file that
has numeric values, such as bias. In addition, when you import object data from a .p2d file, you can
plot large-signal S-parameters as a function of input power or as a function of frequency. These
parameters are denoted LS11, LS12, LS21, and LS22.

This table summarizes the methods that are available in the toolbox for creating rectangular plots
and describes the uses of each one. For more information on a particular type of plot, follow the link
in the table to the documentation for that method.

Method Description
plot Plot of one or more object parameters
plotyy Plot of one or more object parameters with y-axes on both the
left and right sides
semilogx Plot of one or more object parameters using a log scale for the X-
axis
semilogy Plot of one or more object parameters using a log scale for the Y-
axis
loglog Plot of one or more object parameters using a log-log scale

Budget Plot

You use the link budget or budget plot to understand the individual contribution of each component
to a plotted parameter value in a cascaded network with multiple components. The budget plot shows
one or more curves of parameter values as a function of frequency, ordered by the circuit index of the
cascaded network.

Consider the following cascaded network:

casc = rfckt.cascade('Ckts',...
{rfckt.amplifier,rfckt.lcbandpasspi,rfckt.txline})

This figure shows how the circuit index is assigned to each component in the cascade, based on its
sequential position in the network.

You create a 3-D budget plot for this cascade using the plot method with the second argument set to
'budget', as shown in the following command:
analyze(casc,linspace(1e9,3e9,100));
plot(casc,'budget','s21')

Note that you have to analyze your circuit before plotting the budget plot and by default the budget
plot is a 2-D plot. If you specify the array of frequencies in the analyze function you can visualize the
budget results in 3-D. A curve on the budget plot for each circuit index represents the contributions
to the parameter value of the RF components up to that index. This figure shows the budget plot.

2-20
 Analyze and Plot RF Components

If you specify two or more parameters, the RF Toolbox puts the parameters in a single plot. You can
only specify a single format for all the parameters.

Mixer Spur Plot

You use the mixer spur plot to understand how mixer nonlinearities affect output power at the desired
mixer output frequency and at the intermodulation products that occur at the following frequencies:

f out = N ∗ f in + M ∗ f LO

where

• f in is the input frequency.

• f LO is the local oscillator frequency.
• N and M are integers.

The RF toolbox calculates the output power from the mixer intermodulation table (IMT). These tables
are described in detail in the “Visualize Mixer Spurs” on page 6-95 example.

The mixer spur plot shows power as a function of frequency for an rfckt.mixer object or an
rfckt.cascade object that contains a mixer. By default, the plot is three-dimensional and shows a
stem plot of power as a function of frequency, ordered by the circuit index of the object. You can
create a two-dimensional stem plot of power as a function of frequency for a single circuit index by
specifying the index in the mixer spur plot command.

Consider the following cascaded network:

FirstCkt = rfckt.amplifier('NetworkData', ...

rfdata.network('Type', 'S', 'Freq', 2.1e9, ...
'Data', [0,0;10,0]), 'NoiseData', 0, 'NonlinearData', inf);
SecondCkt = read(rfckt.mixer, 'samplespur1.s2d');
ThirdCkt = rfckt.lcbandpasstee('L', [97.21 3.66 97.21]*1e-9, ...
'C', [1.63 43.25 1.63]*1.0e-12);
CascadedCkt = rfckt.cascade('Ckts', ...
{FirstCkt, SecondCkt, ThirdCkt});

This shows how the circuit index is assigned to the components in the cascade, based on its
sequential position in the network.

2-21
2 Model an RF Component

• Circuit index 0 corresponds to the cascade input.

• Circuit index 1 corresponds to the LNA output.
• Circuit index 2 corresponds to the mixer output.
• Circuit index 3 corresponds to the filter output.

You create a spur plot for this cascade using the plot method with the second argument set to
'mixerspur', as shown in the following command:

plot(CascadedCkt,'mixerspur')

Within the three dimensional plot, the stem plot for each circuit index represents the power at that
circuit index. This figure shows the mixer spur plot.

For more information on mixer spur plots, see the plot reference page.

Polar Plots and Smith Charts

You can use the RF toolbox to generate Polar plots and Smith charts. If you specify two or more
parameters, the RF toolbox puts the parameters in a single plot.

The following table describes the Polar plot and Smith charts options, as well as the available
parameters.

Note LS11, LS12, LS21, and LS22 are large-signal S-parameters. You can plot these parameters as a
function of input power or as a function of frequency.

2-22
 Analyze and Plot RF Components

Plot Type Method Parameter

Polar plane polar S11, S12, S21, S22

LS11, LS12, LS21, LS22

(Objects with data from a P2D
file only)
Z Smith chart smithplot with type S11, S22
argument set to 'z'
LS11, LS22 (Objects with data
from a P2D file only)
Y Smith chart smithplot with type S11, S22
argument set to 'y'
LS11, LS22 (Objects with data
from a P2D file only)
ZY Smith chart smithplot with type S11, S22
argument set to 'zy'
LS11, LS22 (Objects with data
from a P2D file only)

By default, the RF toolbox plots the parameter as a function of frequency. When you import block data
from a .p2d or .s2d file, you can also plot parameters as a function of any operating condition from
the file that has numeric values, such as bias.

Note The circle method lets you place circles on a Smith Chart to depict stability regions and
display constant gain, noise figure, reflection and immittance circles. For more information about this
function, see the circle reference page or “Designing Matching Networks for Low Noise Amplifiers”
on page 6-119 example about designing matching networks.

Compute and Plot Time-Domain Specifications

The RF toolbox lets you compute and plot time-domain characteristics for RF components.

This section contains the following topics:

• “Compute Network Transfer Function” on page 2-23

• “Fit Model Object to Circuit Object Data” on page 2-24
• “Compute and Plot Time-Domain Response” on page 2-24

Compute Network Transfer Function

You use the s2tf function to convert 2-port S-parameters to a transfer function. The function returns
a vector of transfer function values that represent the normalized voltage gain of a 2-port network.

The following code illustrates how to read a file data into a passive circuit object, extract the 2-port S-
parameters from the object, and compute the transfer function of the data at the frequencies for
which the data is specified. Here z0 is the reference impedance of the S-parameters, zs is the source
impedance, and zl is the load impedance. See the s2tf reference page for more information on how
these impedances are used to define the gain.
PassiveCkt = rfckt.passive('File','passive.s2p')
z0=50; zs=50; zl=50;

2-23
2 Model an RF Component

[SParams, Freq] = extract(PassiveCkt, 'S Parameters', z0);

TransFunc = s2tf(SParams, z0, zs, zl);

Fit Model Object to Circuit Object Data

You use the rationalfit function to fit a rational function to the transfer function of a passive
component. The rationalfit function returns an rfmodel object that represents the transfer
function analytically.

The following code illustrates how to use the rationalfit function to create an
rfmodel.rational object that contains a rational function model of the transfer function that you
created in the previous example.
RationalFunc = rationalfit(Freq, TransFunc)

To find out how many poles the RF toolbox used to represent the data, look at the length of the A
vector of the RationalFunc model object.
nPoles = length(RationalFunc.A)

Note The number of poles is important if you plan to use the RF model object to create a model for
use in another simulator, because a large number of poles can increase simulation time. For
information on how to represent a component accurately using a minimum number of poles, see
“Represent Circuit Object with Model Object” on page 3-4.

Use the freqresp function to compute the frequency response of the fitted data. To validate the
model fit, plot the transfer function of the original data and the frequency response of the fitted data.
Resp = freqresp(RationalFunc, Freq);
plot(Freq, 20*log10(abs(TransFunc)), 'r', ...
Freq, 20*log10(abs(Resp)), 'b--');
ylabel('Magnitude of H(s) (decibels)');
xlabel('Frequency (Hz)');
legend('Original', 'Fitting result');
title(['Rational fitting with ', int2str(nPoles), ' poles']);

Compute and Plot Time-Domain Response

You use the timeresp function to compute the time-domain response of the transfer function that
RationalFunc represents. This code illustrates how to create a random input signal, compute the
time-domain response of RationalFunc to the input signal, and plot the results.
SampleTime=1e-11;
NumberOfSamples=4750;
OverSamplingFactor = 25;
InputTime = double((1:NumberOfSamples)')*SampleTime;
InputSignal = ...
sign(randn(1, ceil(NumberOfSamples/OverSamplingFactor)));
InputSignal = repmat(InputSignal, [OverSamplingFactor, 1]);
InputSignal = InputSignal(:);

[tresp,t]=timeresp(RationalFunc,InputSignal,SampleTime);
plot(t*1e9,tresp);
title('Fitting Time-Domain Response', 'fonts', 12);
ylabel('Response to Random Input Signal');
xlabel('Time (ns)');

2-24
 Analyze and Plot RF Components

For more information about computing the time response of a model object, see the timeresp
function.

See Also

More About
• “RF Analysis”
• “Export Component Data to File” on page 2-26

2-25
2 Model an RF Component

Export Component Data to File

In this section...
“Available Export Formats” on page 2-26
“How to Export Object Data” on page 2-26
“Export Object Data” on page 2-27

Available Export Formats

RF Toolbox software lets you export data from any rfckt object or from an rfdata.data object to
industry-standard data files and MathWorks AMP files. This export capability lets you store data for
use in other simulations.

Note The toolbox also lets you export data from an rfmodel object to a Verilog-A file. For
information on how to do this, see “Export Verilog-A Model” on page 3-4.

You can export data to the following file formats:

• Industry-standard file formats — Touchstone SNP, YNP, ZNP, HNP, and GNP formats specify the
network parameters and noise information for measured and simulated data.

For more information about Touchstone files, see https://ibis.org/connector/

touchstone_spec11.pdf.
• MathWorks amplifier (AMP) file format — Specifies amplifier network parameters, output power
versus input power, noise data and third-order intercept point.

For more information about .amp files, see “AMP File Data Sections” on page 4-2.

How to Export Object Data

To export data from a circuit or data object, use a write command of the form

status = write(obj,'filename');

where

• status is a return value that indicates whether the write operation was successful.
• obj is the handle of the circuit or rfdata.data object.
• filename is the name of the file that contains the data.

For example,

status = write(rfckt.amplifier, 'myamp.amp');

exports data from an rfckt.amplifier object to the file myamp.amp.

2-26
 Export Component Data to File

Export Object Data

In this example, use the toolbox to create a vector of S-parameter data, store it in an rfdata.data
object, and export it to a Touchstone file.

At the MATLAB prompt:

1 Type the following to create a vector, s_vec, of S-parameter values at three frequency values:

s_vec(:,:,1) = ...
[-0.724725-0.481324i, -0.685727+1.782660i; ...
0.000000+0.000000i, -0.074122-0.321568i];
s_vec(:,:,2) = ...
[-0.731774-0.471453i, -0.655990+1.798041i; ...
0.001399+0.000463i, -0.076091-0.319025i];
s_vec(:,:,3) = ...
[-0.738760-0.461585i, -0.626185+1.813092i; ...
0.002733+0.000887i, -0.077999-0.316488i];
2 Type the following to create an rfdata.data object called txdata with the default property
values:

txdata = rfdata.data;
3 Type the following to set the S-parameter values of txdata to the values you specified in s_vec:

txdata.S_Parameters = s_vec;
4 Type the following to set the frequency values of txdata to [1e9 2e9 3e9]:

txdata.Freq=1e9*[1 2 3];
5 Type the following to export the data in txdata to a Touchstone file called test.s2p:

write(txdata,'test')

See Also

More About
• “Export Verilog-A Model” on page 3-4

2-27
2 Model an RF Component

Basic Operations with RF Objects

This example shows how to read, analyze, and de-embed RF data from a Touchstone data file.

Video Walkthrough

For a walkthrough of the example, play the video.

Read and Analyze RF Data from Touchstone Data File

In this example, you create an sparameters object by reading the S-Parameters of a two-port
passive network stored in the Touchstone format data file, passive.s2p.

Read S-Parameter data from a data file. Use RF Toolbox™ sparameters command to read the
Touchstone data file, passive.s2p. This file contains 50-ohm S-Parameters at frequencies ranging from
315 kHz to 6 GHz. This operation creates an sparameters object, S_50, and stores data from the
file in the object's properties.

S_50 = sparameters('passive.s2p');

Use sparameters to convert the 50-ohm S-Parameters in the sparameters object, to 75-ohm S-
Parameters and save them in the variable S_75. You can easily convert between parameters, for
example, for Y-Parameters from the sparameters object use yparameters and save them in the
variable Y.

Znew = 75;
S_75 = sparameters(S_50, Znew);
Y = yparameters(S_75);

2-28
 Basic Operations with RF Objects

Plot the S11 parameters. Use the smithplot command to plot the 75-ohm S11 parameters on a
Smith® Chart:

smithplot(S_75,1,1)

View the 75-ohm S-Parameters and Y-Parameters at 6 GHz. Type the following set of commands at the
MATLAB® prompt to display the 2-port 75-ohm S-Parameter values and the 2-port Y-Parameter
values at 6 GHz.

freq = S_50.Frequencies;
f = freq(end)

f =
6.0000e+09

s_6GHz = S_75.Parameters(:,:,end)

s_6GHz = 2×2 complex

-0.0764 - 0.5401i 0.6087 - 0.3018i

0.6094 - 0.3020i -0.1211 - 0.5223i

y_6GHz = Y.Parameters(:,:,end)

y_6GHz = 2×2 complex

0.0210 + 0.0252i -0.0215 - 0.0184i

2-29
2 Model an RF Component

-0.0215 - 0.0185i 0.0224 + 0.0266i

For more information, see the sparameters, yparameters, smithplot reference pages.

De-Embed S-Parameters

The Touchstone data file samplebjt2.s2p contains S-Parameter data collected from a bipolar transistor
in a test fixture. The input of the fixture has a bond wire connected to a bond pad. The output of the
fixture has a bond pad connected to a bond wire.

The configuration of the bipolar transistor, which is the device under test (DUT), and the fixture is
shown in the following figure.

In this example, you remove the effects of the fixture and extract the S-parameters of the DUT.

Create RF circuit objects

Create a sparameters object for the measured S-Parameters by reading the Touchstone data file
samplebjt2.s2p. Then, create two more circuit objects, one each for the input pad and output pad.
measured_data = sparameters('samplebjt2.s2p');

L_left = inductor(1e-9);
C_left = capacitor(100e-15);
input_pad = circuit('inputpad');
add(input_pad,[1 2],L_left)
add(input_pad,[2 0],C_left)
setports(input_pad,[1 0],[2 0])

L_right = inductor(1e-9);
C_right = capacitor(100e-15);
output_pad = circuit('outputpad');
add(output_pad,[3 0],C_right)
add(output_pad,[3 4],L_right)
setports(output_pad,[3 0],[4 0])

Analyze the input pad and output pad circuit objects. Analyze the circuit objects at the frequencies at
which the S-Parameters are measured.
freq = measured_data.Frequencies;
input_pad_sparams = sparameters(input_pad,freq);
output_pad_sparams = sparameters(output_pad,freq);

De-embed the S-parameters.

Extract the S-Parameters of the DUT from the measured S-Parameters by removing the effects of the
input and output pads.

2-30
 Basic Operations with RF Objects

de_embedded_sparams = deembedsparams(measured_data,...
input_pad_sparams, output_pad_sparams);

Plot the measured and de-embedded S11 parameters. Type the following set of commands at the
MATLAB® prompt to plot both the measured and the de-embedded S11 parameters on a Z Smith®
Chart:

figure;
smithplot(measured_data,1,1);
hold on
h = smithplot(de_embedded_sparams,1,1);
h.LineStyle = {'-';'--'};
h.ColorOrder = [1 0 0;0 0 1];
h.LegendLabels = {'Measured S11', 'De-embedded S11'};

Plot the measured and de-embedded S22 parameters. Type the following set of commands at the
MATLAB® prompt to plot the measured and the de-embedded S22 parameters on a Z Smith® Chart:

figure;
smithplot(measured_data,2,2);
hold on
h = smithplot(de_embedded_sparams,2,2);
h.LineStyle = {'-';':'};
h.ColorOrder = [1 0 0;0 0 1];
h.LegendLabels = {'Measured S22', 'De-embedded S22'};

2-31
2 Model an RF Component

Plot the measured and de-embedded S21 parameters. Type the following set of commands at the
MATLAB® prompt to plot the measured and the de-embedded S21 parameters, in decibels, on an X-Y
plane:

figure
rfplot(measured_data,2,1,'db','r');
hold on
rfplot(de_embedded_sparams,2,1,'db',':b');
legend('Measured S_{21}', 'De-embedded S_{21}');

2-32
 Basic Operations with RF Objects

See Also

More About
• “RF Data Objects” on page 1-2
• “RF Circuit Objects” on page 1-4
• “RF Model Objects” on page 1-8
• “RF Network Parameter Objects” on page 1-10

2-33
 3

Export Verilog-A Models

• “Model RF Objects Using Verilog-A” on page 3-2

• “Export Verilog-A Model” on page 3-4
3 Export Verilog-A Models

Model RF Objects Using Verilog-A

In this section...
“Overview” on page 3-2
“Behavioral Modeling Using Verilog-A” on page 3-2
“Supported Verilog-A Models” on page 3-2

Overview
Verilog-A is a language for modeling the high-level behavior of analog components and networks.
Verilog-A describes components mathematically, for fast and accurate simulation.

RF Toolbox software lets you export a Verilog-A description of your circuit. You can create a Verilog-A
model of any passive RF component or network and use it as a behavioral model for transient analysis
in a third-party circuit simulator. This capability is useful in signal integrity engineering. For example,
you can import the measured four-port S-parameters of a backplane into the toolbox, export a
Verilog-A model of the backplane to a circuit simulator, and use the model to determine the
performance of your driver and receiver circuitry when they are communicating across the
backplane.

Behavioral Modeling Using Verilog-A

The Verilog-A language is a high-level language that uses modules to describe the structure and
behavior of analog systems and their components. A module is a programming building block that
forms an executable specification of the system.

Verilog-A uses modules to capture high-level analog behavior of components and systems. Modules
describe circuit behavior in terms of

• Input and output nets characterized by predefined Verilog-A disciplines that describe the
attributes of the nets.
• Equations and module parameters that define the relationship between the input and output nets
mathematically.

When you create a Verilog-A model of your circuit, the toolbox writes a Verilog-A module that
specifies circuit's input and output nets and the mathematical equations that describe how the circuit
operates on the input to produce the output.

Supported Verilog-A Models

RF Toolbox software lets you export a Verilog-A model of an rfmodel object. The toolbox provides
one rfmodel object, rfmodel.rational, that you can use to represent any RF component or
network for export to Verilog-A.

The rfmodel.rational object represents components as rational functions in pole-residue form, as

described in the rfmodel.rational reference page. This representation can include complex poles
and residues, which occur in complex-conjugate pairs.

The toolbox implements each rfmodel.rational object as a series of Laplace Transform S-domain
filters in Verilog-A using the numerator-denominator form of the Laplace transform filter:

3-2
 Model RF Objects Using Verilog-A

M
∑ nksk
k=0
H(s) = N
∑ dksk
k=0

where

• M is the order of the numerator polynomial.

• N is the order of the denominator polynomial.
• nk is the coefficient of the kth power of s in the numerator.
• dk is the coefficient of the kth power of s in the denominator.

The number of poles in the rational function is related to the number of Laplace transform filters in
the Verilog-A module. However, there is not a one-to-one correspondence between the two. The
difference arises because the toolbox combines each pair of complex-conjugate poles and the
corresponding residues in the rational function to form a Laplace transform numerator and
denominator with real coefficients. the toolbox converts the real poles of the rational function directly
to a Laplace transform filter in numerator-denominator form.

See Also

More About
• “Export Verilog-A Model” on page 3-4

3-3
3 Export Verilog-A Models

Export Verilog-A Model

In this section...
“Represent Circuit Object with Model Object” on page 3-4
“Write Verilog-A Module” on page 3-5

Represent Circuit Object with Model Object

Before you can write a Verilog-A model of an RF circuit object, you need to create an
rfmodel.rational object to represent the component.

There are two ways to create an RF model object:

• You can fit a rational function model to the component data using the rationalfit function.
• You can use the rfmodel.rational constructor to specify the pole-residue representation of the
component directly.

This section discusses using a rational function model. For more information on using the constructor,
see the rfmodel.rational reference page.

When you use the rationalfit function to create an rfmodel.rational object that represents an
RF component, the arguments you specify affect how quickly the resulting Verilog-A model runs in a
circuit simulator.

You can use the rationalfit function with only the two required arguments. The syntax is:
model_obj = rationalfit(freq,data)

where

• model_obj is a handle to the rational function model object.

• freq is a vector of frequency values that correspond to the data values.
• data is a vector that contains the data to fit.

For faster simulation, create a model object with the smallest number of poles required to accurately
represent the component. To control the number of poles, use the syntax:
model_obj = rationalfit(freq,data,tol,weight,delayfactor)

where

• tol — the relative error-fitting tolerance, in decibels. Specify the largest acceptable tolerance for
your application. Using tighter tolerance values may force the rationalfit function to add more
poles to the model to achieve a better fit.
• weight — a vector that specifies the weighting of the fit at each frequency.
• delayfactor — a value that controls the amount of delay used to fit the data. Delay introduces a
phase shift in the frequency domain that may require a large number of poles to fit using a
rational function model. When you specify the delay factor, the rationalfit function represents
the delay as an exponential phase shift. This phase shift allows the function to fit the data using
fewer poles.

For more information, see rationalfit.

3-4
 Export Verilog-A Model

Note You can also specify the number of poles directly using the npoles argument. The model
accuracy is not guaranteed with approach, so you should not specify npoles when accuracy is
critical. For more information on the npoles argument, see the rationalfit reference page.

If you plan to integrate the Verilog-A module into a large design for simulation using detailed models,
such as transistor-level circuit models, the simulation time consumed by a Verilog-A module may have
a trivial impact on the overall simulation time. In this case, there is no reason to take the time to
optimize the rational function model of the component.

For more information on the rationalfit function arguments, see the rationalfit reference
page.

Write Verilog-A Module

You use the writeva method to create a Verilog-A module that describes the RF model object. This
method writes the module to a specified file. Use the syntax:

status = writeva(model_obj,'obj1',{'inp','inn'},{'outp','outn'})

to write a Verilog-A module for the model object model_obj to the file obj1.va. The module has
differential input nets, inp and inn, and differential output nets, outp and outn. The method
returns status, a logical value of true if the operation is successful and false otherwise.

The write reference page describes the method arguments in detail.

An example of exporting a Verilog-A module appears in the RF Toolbox example, “Export Verilog-A
module from Rational Function” on page 6-67.

See Also

More About
• “Model RF Objects Using Verilog-A” on page 3-2
• “Export Component Data to File” on page 2-26

3-5
 4

AMP File Format

4 AMP File Format

AMP File Data Sections

In this section...
“Overview” on page 4-2
“Denoting Comments” on page 4-2
“Data Sections” on page 4-3
“S, Y, or Z Network Parameters” on page 4-3
“Noise Parameters” on page 4-4
“Noise Figure Data” on page 4-5
“Power Data” on page 4-6
“IP3 Data” on page 4-8
“Inconsistent Data Sections” on page 4-9

Overview
The AMP data file describes a single nonlinear device. Its format can contain the following types of
data:

• S, Y, or Z network parameters
• Noise parameters
• Noise figure data
• Power data
• IP3 data

An AMP file must contain either power data or network parameter data to be valid. To accommodate
analysis at more than one frequency, the file can contain more than one section of power data. Noise
data, noise figure data, and IP3 data are optional.

Note If the file contains both network parameter data and power data, RF Toolbox software checks
the data for consistency. If the amplifier gain computed from the network parameters is not
consistent with the gain computed from the power data, a warning appears.

Two AMP files, samplepa1.amp and default.amp, ship with the toolbox to show the AMP format.
They describe a nonlinear 2-port amplifier with noise. See “Model Cascaded Network” for an example
that shows how to use an AMP file.

Denoting Comments
An asterisk (*) or an exclamation point (!) precedes a comment that appears on a separate line.

A semicolon (;) precedes a comment that appears following data on the same line.

4-2
 AMP File Data Sections

Data Sections
Each kind of data resides in its own section. Each section consists of a two-line header followed by
lines of numeric data. Numeric values can be in any valid MATLAB format.

A new header indicates the end of the previous section. The data sections can appear in any order in
the file.

Note In the data section descriptions, brackets ([]) indicate optional data or characters. All values
are case insensitive.

S, Y, or Z Network Parameters
Header Line 1

The first line of the header has the format

Keyword [Parameter] [R[REF][=]value]

Keyword indicates the type of network parameter. Its value can be S[PARAMETERS],
Y[PARAMETERS], or Z[PARAMETERS]. Parameter indicates the form of the data. Its value can be
MA, DB, or RI. The default for S-parameters is MA. The default for Y- and Z-parameters is RI. R[REF]
[=]value is the reference impedance. The default reference impedance is 50 ohms.

Note R[REF][=]value must be a positive real scalar or vector. If R[REF][=]value is a vector,

then the vector must be equal to the number of network parameter data points or frequency vector.

The following table explains the meaning of the allowable Parameter values.

Parameter Description
MA Data is given in (magnitude, angle) pairs with angle in degrees
(default for S-parameters).
DB Data is given in (dB-magnitude, angle) pairs with angle in degrees.
RI Data is given in (real, imaginary) pairs (default for Y- and Z-
parameters).

This example of a first line indicates that the section contains S-parameter data given in (real,
imaginary) pairs, and that the reference impedance is 50 ohms.
S RI R 50

Header Line 2

The second line of the header has the format

Independent_variable Units

The data in a section is a function of the Independent_variable. Currently, for S-, Y-, and Z-
parameters, the value of Independent_variable is always F[REQ]. Units indicates the default units of
the frequency data. It can be GHz, MHz, or KHz. You must specify Units, but you can override this
default on any given line of data.

4-3
4 AMP File Format

This example of a second line indicates that the default units for frequency data is GHz.
FREQ GHZ

Data

The data that follows the header typically consists of nine columns.

The first column contains the frequency points where network parameters are measured. They can
appear in any order. If the frequency is given in units other than those you specified as the default,
you must follow the value with the appropriate units; there should be no intervening spaces. For
example,
FREQ GHZ
1000MHZ ...
2000MHZ ...
3000MHZ ...

Columns two though nine contain 2-port network parameters in the order N11, N21, N12, N22.
Similar to the Touchstone format, each Nnn corresponds to two consecutive columns of data in the
chosen form: MA, DB, or RI. The data can be in any valid MATLAB format.

This example is derived from the file default.amp. A comment line explains the column
arrangement of the data where re indicates real and im indicates imaginary.
S RI R 50
FREQ GHZ
* FREQ reS11 imS11 reS21 imS21 reS12 imS12 reS22 imS22
1.00 -0.724725 -0.481324 -0.685727 1.782660 0.000000 0.000000 -0.074122 -0.321568
1.01 -0.731774 -0.471453 -0.655990 1.798041 0.001399 0.000463 -0.076091 -0.319025
1.02 -0.738760 -0.461585 -0.626185 1.813092 0.002733 0.000887 -0.077999 -0.316488

Noise Parameters
Header Line 1

The first line of the header has the format

Keyword

Keyword must be NOI[SE].

Header Line 2

The second line of the header has the format

Variable Units

Variable must be F[REQ]. Units indicates the default units of the frequency data. It can be GHz,
MHz, or KHz. You can override this default on any given line of data. This example of a second line
indicates that frequency data is assumed to be in GHz, unless other units are specified.
FREQ GHz

Data

The data that follows the header must consist of five columns.

The first column contains the frequency points at which noise parameters were measured. The
frequency points can appear in any order. If the frequency is given in units other than those you

4-4
 AMP File Data Sections

specified as the default, you must follow the value with the appropriate units; there should be no
intervening spaces. For example,
NOI
FREQ GHZ
1000MHZ ...
2000MHZ ...
3 ...
4 ...
5 ...

Columns two through five contain, in order,

• Minimum noise figure in decibels

• Magnitude of the source reflection coefficient to realize minimum noise figure
• Phase in degrees of the source reflection coefficient
• Effective noise resistance normalized to the reference impedance of the network parameters

This example is taken from the file default.amp. A comment line explains the column arrangement
of the data.
NOI RN
FREQ GHz
* Freq Fmin(dB) GammmaOpt(MA:Mag) GammmaOpt(MA:Ang) RN/Zo
1.90 10.200000 1.234000 -78.400000 0.240000
1.93 12.300000 1.235000 -68.600000 0.340000
2.06 13.100000 1.254000 -56.700000 0.440000
2.08 13.500000 1.534000 -52.800000 0.540000
2.10 13.900000 1.263000 -44.400000 0.640000

Noise Figure Data

The AMP file format supports the use of frequency-dependent noise figure (NF) data.

Header Line 1

The first line of the header has the format

Keyword [Units]

For noise figure data, Keyword must be NF. The optional Units field indicates the default units of the
NF data. Its value must be dB, i.e., data must be given in decibels.

This example of a first line indicates that the section contains NF data, which is assumed to be in
decibels.
NF

Header Line 2

The second line of the header has the format

Variable Units

Variable must be F[REQ]. Units indicates the default units of the frequency data. It can be GHz,
MHz, or KHz. This example of a second line indicates that frequency data is assumed to be in GHz.

4-5
4 AMP File Format

FREQ GHz

Data

The data that follows the header typically consists of two columns.

The first column contains the frequency points at which the NF data are measured. Frequency points
can appear in any order. For example,

NF
FREQ MHz
2090 ...
2180 ...
2270 ...

Column two contains the corresponding NF data in decibels.

This example is derived from the file samplepa1.amp.

NF dB
FREQ GHz
1.900 10.3963213
2.000 12.8797965
2.100 14.0611765
2.200 13.2556751
2.300 12.9498642
2.400 13.3244309
2.500 12.7545104

Note If your noise figure data consists of a single scalar value with no associated frequency, that
same value is used for all frequencies. Enter the value in column 1 of the line following header line 2.
You must include the second line of the header, but it is ignored.

Power Data
An AMP file describes power data as input power-dependent output power.

Header Line 1

The first line of the header has the format

Keyword [Units]

For power data, Keyword must be POUT, indicating that this section contains power data. Because
output power is complex, Units indicates the default units of the magnitude of the output power
data. It can be dBW, dBm, mW, or W. The default is W. You can override this default on any given line of
data.

The following table explains the meaning of the allowable Units values.

4-6
 AMP File Data Sections

Allowable Power Data Units

Units Description
dBW Decibels referenced to one watt
dBm Decibels referenced to one milliwatt
mW Milliwatts
W Watts

This example of a first line indicates that the section contains output power data whose magnitude is
assumed to be in decibels referenced to one milliwatt, unless other units are specified.

POUT dBm

Header Line 2

The second line of the header has the format

Keyword [Units] FREQ[=]value

Keyword must be PIN. Units indicates the default units of the input power data. The default is W.
You can override this default on any given line of data. FREQ[=]value is the frequency point at
which the power is measured. The units of the frequency point must be specified explicitly using the
abbreviations GHz, MHz, kHz, or Hz.

This example of a second line indicates that the section contains input power data that is assumed to
be in decibels referenced to one milliwatt, unless other units are specified. It also indicates that the
power data was measured at a frequency of 2.1E+009 Hz.

PIN dBm FREQ=2.1E+009Hz

Data

The data that follows the header typically consists of three columns:

• The first column contains input power data. The data can appear in any order.
• The second column contains the corresponding output power magnitude.
• The third column contains the output phase shift in degrees.

Note RF Toolbox software does not use the phase data directly. RF Blockset™ blocks use this data
in conjunction with RF Toolbox software to create the AM/PM conversion table for the Equivalent
Baseband library General Amplifier and General Mixer blocks.

If all phases are zero, you can omit the third column. If all phases are zero or omitted, the toolbox
assumes that the small signal phase from the network parameter section of the file
(180*angle(S21(f))/pi) is the phase for all power levels.

In contrast, if one or more phases in the power data section are nonzero, the toolbox interpolates and
extrapolates the data to determine the phase at all power levels. The small signal phase
(180*angle(S21(f))/pi) from the network parameter section is ignored.

Inconsistency between the power data and network parameter sections of the file may cause
incorrect results. To avoid this outcome, verify that the following criteria must is met:

4-7
4 AMP File Format

• The lowest input power value for which power data exists falls in the small signal (linear) region.
• In the power table for each frequency point f, the power gain and phase at the lowest input power
value are equal to 20*log10(abs(S21(f))) and 180*angle(S21(f))/pi, respectively, in the
network parameter section.

If the power is given in units other than those you specified as the default, you must follow the value
with the appropriate units. There should be no intervening spaces.

This example is derived from the file default.amp. A comment line explains the column
arrangement of the data.

POUT dbm
PIN dBm FREQ = 2.10GHz
* Pin Pout Phase(degrees)
0.0 19.28 0.0
1.0 20.27 0.0
2.0 21.26 0.0

Note The file can contain more than one section of power data, with each section corresponding to a
different frequency value. When you analyze data from a file with multiple power data sections,
power data is taken from the frequency point that is closest to the analysis frequency.

IP3 Data
An AMP file can include frequency-dependent, third-order input (IIP3) or output (OIP3) intercept
points.

Header Line 1

The first line of the header has the format

Keyword [Units]

For IP3 data, Keyword can be either IIP3 or OIP3, indicating that this section contains input IP3
data or output IP3 data. Units indicates the default units of the IP3 data. Valid values are dBW, dBm,
mW, and W. The default is W.

This example of a first line indicates that the section contains input IP3 data which is assumed to be
in decibels referenced to one milliwatt.

IIP3 dBm

Header Line 2

The second line of the header has the format

Variable Units

Variable must be FREQ. Units indicates the default units of the frequency data. Valid values are
GHz, MHz, and KHz. This example of a second line indicates that frequency data is assumed to be in
GHz.

FREQ GHz

4-8
 AMP File Data Sections

Data

The data that follows the header typically consists of two columns.

The first column contains the frequency points at which the IP3 parameters are measured. Frequency
points can appear in any order.

OIP3
FREQ GHz
2.010 ...
2.020 ...
2.030 ...

Column two contains the corresponding IP3 data.

This example is derived from the file samplepa1.amp.

OIP3 dBm
FREQ GHz
2.100 38.8730377

Note If your IP3 data consists of a single scalar value with no associated frequency, then that same
value is used for all frequencies. Enter the value in column 1 of the line following header line 2. You
must include the second line of the header, but the application ignores it.

Inconsistent Data Sections

If an AMP file contains both network parameter data and power data, RF Toolbox software checks the
data for consistency.

The toolbox compares the small-signal amplifier gain defined by the network parameters, S21, and by
the power data, Pout – Pin. The discrepancy between the two is computed in dBm using the following
equation:

ΔP = S21(f P) − Pout(f P) + Pin(f P)

wherefP is the lowest frequency for which power data is specified.

The discrepancy is shown in the following graph.

4-9
4 AMP File Format

If ΔP is more than 0.4 dB, a warning appears. Large discrepancies may indicate measurement errors
that require resolution.

See Also

More About
• “S-Parameter Notation”
• “Determining Parameter Formats” on page 5-2

4-10
 5

How Tos, Definitions, Algorithms

• “Determining Parameter Formats” on page 5-2

• “RF and Microwave Filter Modeling” on page 5-4
• “Designing Matching Networks in RF Systems” on page 5-6
5 How Tos, Definitions, Algorithms

Determining Parameter Formats

In this section...
“Primary and Secondary Formats” on page 5-2
“Determining Formats for One Parameter” on page 5-3
“Determining Formats for Multiple Parameters” on page 5-3

When you call plotyy without specifying the formats for the specified parameter, plotyy
determines the formats from the primary and secondary formats.

Primary and Secondary Formats

The following table shows the primary and secondary formats for the parameters for all circuit and
data objects. Use the listparam method to list the valid parameters for a particular object. Use the
listformat method to list valid formats.

Parameter Primary Format Secondary Format

S11, S12, S21, S22 Magnitude(decibels) Angle(Degrees)
LS11, LS12, LS21, LS22 Magnitude(decibels) Angle(Degrees)
NF Magnitude(decibels) none
OIP3 dBm W
Pout dBm W
Phase Angle(Degrees) none
AM/AM Magnitude(decibels) none
AM/PM Angle(Degrees) none
GammaIn, GammaOut Magnitude(decibels) Angle(Degrees)
Gt, Ga, Gp, Gmag, Gmsg Magnitude(decibels) none
Delta Magnitude(decibels) Angle(Degrees)
TF1, TF2 Magnitude(decibels) Angle(Degrees)
GammaMS, GammaML Magnitude(decibels) Angle(Degrees)
VSWRIn, VSWROut Magnitude(decibels) none
GroupDelay ns none
Fmin Magnitude(decibels) none
GammaOPT Magnitude(decibels) Angle(Degrees)
K, Mu, MuPrime none none
RN none none
PhaseNoise dBc/Hz none
NTemp K none
NFactor none none

5-2
 Determining Parameter Formats

Determining Formats for One Parameter

When you specify only one parameter for plotting, plotyy creates the plot as follows:

• The predefined primary format is the format for the left Y-axis.
• The predefined secondary format is the format for the right Y-axis.

If the specified parameter does not have the predefined secondary format, plotyy behaves the same
way as plot, and does not add a second y-axis to the plot.

Determining Formats for Multiple Parameters

To plot multiple parameters on two Y-axes, plotyy tries to find two formats from the predefined
primary and secondary formats for the specified parameters. To be used in the plot, the formats must
meet the following criteria:

• Each format must be a valid format for at least one parameter.

• Each parameter must be plotted at least on one Y-axis.

If cannot meet these criteria,plotyy it issues an error message.

The function uses the following algorithm to determine the two parameters:

1 Look up the primary and secondary formats for the specified parameters.
2 If one or more pairs of primary-secondary formats meets the preceding criteria for all
parameters:

• Select the pair that applies to the most parameters.

• Use these formats to create the plot.

Otherwise, proceed to the next step.

3 If no pairs of primary-secondary formats meet the criteria for all parameters, try to find one or
more pairs of primary-primary format that meets the criteria. If one or more pairs of primary-
primary formats meets the preceding criteria for all parameters:

• Select the pair that applies to the most parameters.

• Use these formats to create the plot.

Otherwise, proceed to the next step.

4 If the preceding steps fail to produce a plot, try to find one format from the predefined primary
formats. If a primary format is valid for all parameters, use this format to create the plot with the
MATLAB plot function.
5 If all the preceding steps are not successful, issue an error message.

See Also

More About
• “S-Parameter Notation”
• “AMP File Data Sections” on page 4-2

5-3
5 How Tos, Definitions, Algorithms

RF and Microwave Filter Modeling

Receivers, transmitters, and frequency synthesizers use RF and microwave filters to select or reject
signals with a particular band of frequency. Filtering a signal modifies its phase and magnitude
components. RF receiver systems use filters such as the RF preselector filter, image rejection filter,
and IF filter. You can design filters suited to your application using the rffilter object from RF
Toolbox or the Filter blocks from RF Blockset. For example, you can design an RF preselector filter in
Chebyshev configurations using an rffilter object or the Circuit Envelope Filter block and filter
undesired frequency bands causing spurious emissions and intermodulation distortions in mixers [1].

You can also use the rffilter object to create a Butterworth or inverse Chebyshev filter. The
rffilter object also supports implementing a transfer function. For more information, see “Design
Data for Transfer Function Implementation”. In addition to the three configurations, you can also
design an ideal filter in Simulink® environment using the Circuit Envelope Filter block. Ideal filters
perfectly allow frequencies in the passband and completely reject frequencies in the stopband.

Microstrip filters play an important role in microwave applications. Almost all communication
systems contain an RF front end that performs signal processing using RF and microwave filters. You
can design coupled-line, hairpin, and stepped-impedance lowpass filters in the microstrip form. These
filters have very low insertion loss and are easy to fabricate in compact sizes. Design these filters in a
printed circuit board (PCB) using RF PCB Toolbox™ filter objects.

Design Workflows
Design RF and microwave filters using these workflows:

• “Design, Visualize and Explore Inverse Chebyshev Filter - I” on page 6-193 — This workflow shows
how to determine the transfer function for a fifth-order inverse Chebyshev lowpass filter with 1 dB
passband attenuation, cutoff frequency of 1 rad/sec, and a minimum attenuation of 50 dB in the
stopband.
• “Frequency Response of RF Transmit/Receive Duplex Filter” (RF Blockset) — This workflow shows
how to use RF Blockset Circuit Envelope blocks to simulate a transmit/receive duplex filter and
calculate the frequency response curves from a broadband white-noise input.
• “Stepped Impedance Maximally Flat Lowpass Filter for Microwave Applications” (RF PCB
Toolbox) — This workflow shows you how to design a stepped-impedance lowpass filter for X-band
applications.
• “Model RF Filter Using Circuit Envelope” (RF Blockset) — This workflow shows how to model an
RF filter using the Circuit Envelope library and compare the input and output signal amplitudes to
study signal attenuation.
• “Design and Analyze HighPass Filter Using pcbComponent” (RF PCB Toolbox) — This example
shows how to design and analyze a microstrip highpass filter. The filter design is based on a three-
pole Chebyshev highpass prototype with 0.1 dB passband ripple and a cutoff frequency of 1.5
GHz.

References
[1] Besser, Les, and Rowan Gilmore. Practical RF Circuit Design for Modern Wireless Systems. 1:
Passive Circuits and Systems. Boston, Mass.: Artech House, 2003.

5-4
 RF and Microwave Filter Modeling

See Also

Related Examples
• “Designing Matching Networks in RF Systems” on page 5-6

5-5
5 How Tos, Definitions, Algorithms

Designing Matching Networks in RF Systems

Matching networks in RF system enable maximum power transfer from the source to the load. You
can design a set of circuits that match the impedance of a source to the impedance of a load at a
specific center frequency using the matchingnetwork object or the Matching Network Designer
app. You can also use the object and the app to visualize, and compare matching networks for the
one-port loads.

Using the object and the app, you can also:

• Design two- and three-component lumped-element matching networks at desired frequencies and
unloaded-Q factors.
• Provide source and load impedance as a one-port Touchstone file, scalar impedance, RF circuit
object, RF network parameter object, Antenna Toolbox™ object, or as an anonymous function.
• Sort matching networks using constraints such as operating frequency range and power wave S-
parameters.
• Plot power wave S-parameters [1] of the matching network on a Smith chart and Cartesian plot.
• Plot voltage standing wave ratio (VSWR) and impedance transformation plots.
• Plot magnitude, phase, real, and imaginary parts of power wave S-parameters of the matching
network.
• Export selected networks as circuit objects or power wave S-parameters as sparameters
objects.

Available Configuration
You can design matching networks in these network configurations:

• Pi Topology
• T Topology
• L Topology
• Three-Components

Design Workflows
You can design matching networks for RF systems using these workflows:

• “Design Matching Networks for Passive Multiport Network” on page 6-204 — This workflow
shows how to design matching networks for 16-port passive networks at 39 GHz for 5G mm Wave
systems. You design a matching network for each port that functions between two 1-port
terminations.
• “Design Broadband Matching Networks for Antennas” on page 6-139 — This workflow shows how
to design a broadband matching network between a resistive source and inductive load using
optimization with direct search methods.
• “Designing Matching Networks for Low Noise Amplifiers” on page 6-119 — This workflow shows
how to verify the design of input and output matching networks for a low noise amplifier (LNA)
using a gain and noise figure plot.
• “Design Broadband Matching Networks for Amplifier” on page 6-148 — This workflow shows how
to design broadband matching networks for a LNA.

5-6
 Designing Matching Networks in RF Systems

• “Design of Quarter-Wave Transformer for Impedance Matching Applications” (RF PCB Toolbox) —
This workflow shows how to design a quarter-wave transformer for impedance-matching
applications using pcbComponent, microstripLine, and traceRectangular objects.

References
[1] Kurokawa, K. “Power Waves and the Scattering Matrix.” IEEE Transactions on Microwave Theory
and Techniques 13, no. 2 (March 1965): 194–202. https://doi.org/10.1109/
TMTT.1965.1125964.

[2] Ludwig, Reinhold, and Gene Bogdanov. RF Circuit Design: Theory and Applications. Upper Saddle
River, NJ: Prentice-Hall, 2009.

See Also

Related Examples
• “RF and Microwave Filter Modeling” on page 5-4

5-7
 6

RF Toolbox Examples
6 RF Toolbox Examples

Superheterodyne Receiver Using RF Budget Analyzer App

This example shows how to:

• Build a superheterodyne receiver.

• Analyze the receiver's RF budget for gain, noise figure, and IP3 using the RF Budget Analyzer app.
• Verify your analysis using RF Blockset circuit envelope simulation.

You can either build all the superheterodyne receiver components using the MATLAB® command line
or using the RF Budget Analyzer app. This example shows you how to design the receiver
components in the MATLAB command line and view the analysis using the RF Budget Analyzer app.
The receiver bandwidth is between 5.825 GHz and 5.845 GHz and the receiver is a part of a
transmitter-receiver system described in the IEEE conference papers, [1] and [2].

Introduction

RF system designers begin the design process with a budget specification for how much gain, noise
figure (NF), and nonlinearity (IP3) the entire system must satisfy. To assure the feasibility of an
architecture modeled as a simple cascade of RF elements, designers calculate both the per-stage and
cascade values of gain, noise figure, and third-order intercept point (IP3).

Using the RF Budget Analyzer app, you can:

• Build a cascade of RF elements.

• Calculate the per-stage and cascade output power, gain, noise figure, SNR, and IP3 of the system.
• Export the per-stage and cascade values to the MATLAB workspace.
• Export the system design to RF Blockset for simulation.
• Export the system design to an RF Blockset measurement testbench as a device-under-test (DUT)
subsystem and verify the results using the App.

For more information, see RF Budget Analyzer app.

Video Walkthrough

For a walkthrough of the example, play the video.

6-2
 Superheterodyne Receiver Using RF Budget Analyzer App

Build Superheterodyne Receiver

The first components in the superheterodyne receiver system architecture are the antenna and TR
switch. You can replace the antenna element with the effective power reaching the switch.

1. The receiver uses the TR switch to switch between the transmitter and the receiver. The switch
adds a loss of 1.3 dB to the system. Use the rfelement object to create a TR switch with a gain of –
1.3 dB, and output third-order intercept (OIP3) of 37 dBm. To match the RF budget results from
reference [1], assume that the noise figure is 2.3 dB.

elements(1) = rfelement('Name','TRSwitch','Gain',-1.3,'NF',2.3,'OIP3',37);

2. Next element in the chain is the RF bandpass filter. Use an rffilter object to model an RF
bandpass filter. For more information about the design specification of the RF Bandpass filter used in
this example, see “Design IF Butterworth Bandpass Filter” on page 6-183.

Fcenter = 5.8e9;
Bwpass = 20e6;

6-3
6 RF Toolbox Examples

elements(2) = rffilter('ResponseType','Bandpass', ...

'FilterType','Butterworth','FilterOrder',6, ...
'PassbandAttenuation',10*log10(2), ...
'Implementation','Transfer function', ...
'PassbandFrequency',[Fcenter-Bwpass/2 Fcenter+Bwpass/2], ...
'Name','RF_Filter');

3. The S-parameters for the element(2) filter are not ideal. Therefore, use a 1 dB attenuator to
model the insertion loss of the filter into the system.

elements(3) = attenuator("Attenuation",1);

4. Use the amplifier object to model a Low Noise Amplifier block with a gain of 15 dB, noise figure
of 1.5 dB, and OIP3 of 26 dBm.

elements(4) = amplifier('Name','LNA','Gain',15,'NF',1.5,'OIP3',26);

5. Model a Gain block with a gain of 10.5 dB, noise figure of 3.5 dB, and OIP3 of 23 dBm.

elements(5) = amplifier('Name','Gain','Gain',10.5,'NF',3.5,'OIP3',23);

6. The receiver downconverts the RF frequency to an IF frequency of 400 MHz. Use the modulator
object to create a Demodulator block with local oscillator (LO) frequency of 5.4 GHz, gain of –7 dB,
noise figure of 7 dB, and OIP3 of 15 dBm.

elements(6) = modulator('Name','Demod','Gain',-7,'NF',7,'OIP3',15, ...

'LO',5.4e9, 'ConverterType','Down');

7. Model an IF bandpass filter using rffilter object.

Fcenter = 400e6;
Bwpass = 5e6;
elements(7) = rffilter('ResponseType','Bandpass', ...
'FilterType','Chebyshev','FilterOrder',4, ...
'PassbandAttenuation',0.1, ...
'Implementation','Transfer function', ...
'PassbandFrequency',[Fcenter-Bwpass/2 Fcenter+Bwpass/2], ...
'Name','IF_Filter');

8. The S-parameters for the element(7) filter are not ideal. Therefore, use a 1 dB attenuator to
model the insertion loss of the filter into the system.

elements(8) = attenuator("Attenuation",1);

9. Model an IF Amplifier block with a gain of 40 dB and a noise figure of 2.5 dB.

elements(9) = amplifier('Name','IFAmp','Gain',40,'NF',2.5);

10. As seen in the references, the receiver uses an Automatic Gain Control (AGC) block where the
gain varies with the available input power. For an input power of –80 dB, the AGC gain is at a
maximum of 17.5 dB. Use an Amplifier block to model an AGC. Model an AGC block with a gain of
17.5 dB, noise figure of 4.3 dB, and OIP3 of 36 dBm.

elements(10) = amplifier('Name','AGC','Gain',17.5,'NF',4.3,'OIP3',36);

Cascade all the elements to construct an RF chain. Use the rfbudget object to calculate the RF
budget of the superheterodyne receiver. Set Input frequency to 5.8 GHz and signal bandwidth to 20

6-4
 Superheterodyne Receiver Using RF Budget Analyzer App

MHz. Replace the antenna element with the effective available input power which is estimated to be –
66 dB when reaching the TR switch.
superhet = rfbudget('Elements',elements,'InputFrequency',5.8e9, 'AvailableInputPower',-66,'Signa

superhet =
rfbudget with properties:

Elements: [1x10 rf.internal.rfbudget.Element]

InputFrequency: 5.8 GHz
AvailableInputPower: -66 dBm
SignalBandwidth: 20 MHz
Solver: Friis
AutoUpdate: true

Analysis Results
OutputFrequency: (GHz) [ 5.8 5.8 5.8 5.8 5.8 0.4 0.4 0.4 0
OutputPower: (dBm) [-67.3 -67.3 -68.3 -53.3 -42.8 -49.8 -49.9 -50.9 -10
TransducerGain: (dB) [ -1.3 -1.3 -2.3 12.7 23.2 16.2 16.1 15.1 55
NF: (dB) [ 2.3 2.3 3.112 4.39 4.494 4.524 4.524 4.534 4.
IIP2: (dBm) []
OIP2: (dBm) []
IIP3: (dBm) [ 38.3 38.3 38.3 13.29 -0.3904 -3.824 -3.824 -3.824 -3.8
OIP3: (dBm) [ 37 37 36 25.99 22.81 12.38 12.28 11.28 51.
SNR: (dB) [32.66 32.66 31.85 30.57 30.47 30.44 30.44 30.43 30.

View the analysis in the RF Budget Analyser app by typing this command at the command line. The
app displays the cascade values such as output frequency of the receiver, output power, gain, noise
figure, OIP3, and Signal-to- Noise-Ratio (SNR). The RF Budget Analyzer app saves the model in a
MAT-file.
show(superhet);

6-5
6 RF Toolbox Examples

Plot Cascade Transducer Gain and Cascade Noise Figure

Plot the cascade transducer gain of the receiver using the rfplot function. You can also use the Plot
button on the RF Budget Analyzer app to plot the different output values.

rfplot(superhet,'GainT')
view(90,0)

2. Plot the cascade noise figure of the receiver.

rfplot(superhet,'NF')
view(90,0)

6-6
 Superheterodyne Receiver Using RF Budget Analyzer App

Export to MATLAB Script

Export the model to a MATLAB script. You can use this script to edit your RF chain or share your
receiver system with others.

h = exportScript(superhet);

The script opens automatically in a MATLAB Editor window.

h.closeNoPrompt

Verify Output Power and Transducer Gain Using RF Blockset Simulation

Use the Export button to export the receiver to RF Blockset to perform circuit envelope simulation.
You can use circuit envelope simulation to verify the analysis results that you generated in the
app.You can also use the exportRFBlockset function to export the chain to RF Blockset.

exportRFBlockset(superhet)

• Run the RF Blockset model to calculate the output power (dBm) and transducer gain (dB) of the
receiver. Note that the results match the values of the Pout (dBm) and the GainT (dB)
properties of the receiver obtained using the RF Budget Analyzer app.

6-7
6 RF Toolbox Examples

Look under the mask of the Demodulator block. This block consists of an ideal filter and a channel
select filter and an LO for frequency up or down conversion. The stop time for the simulation is zero.
To simulate time-varying results, you must change the stop time.

Export to RF Blockset Testbench

Use the Export button to export the receiver to an RF Blockset measurement testbench. Measuring
properties such as gain, noise figure, and intercept points (IPs) in RF systems is crucial to the system
design and analysis processes. Such measurements are accomplished by placing a DUT in a
testbench. Use the testbenches blocks to measure the RF properties and verify the response of the
DUT.

The RF Blockset testbench consists of two subsystems: RF Measurement Unit and Device Under
Test.

6-8
 Superheterodyne Receiver Using RF Budget Analyzer App

The Device Under Test subsystem contains the superheterodyne receiver you exported from the
RF Budget Analyzer app. Double-click the Device Under Test subsystem to look inside.

Double-click the RF Measurement Unit subsystem to see the system parameters. By default, the
RF Blockset testbench verifies the gain.

6-9
6 RF Toolbox Examples

Verify Gain, Noise Figure, and IP3 Using RF Blockset Testbench

You can verify the gain, noise figure, and IP3 measurements using the RF Blockset testbench.

By default, the model verifies the gain of the DUT. Run the model to check the gain. The simulated
gain value matches the cascade transducer gain value from the app. The scope shows an output
power of approximately 6.7 dB at 400 MHz that matches the output power value in the RF Budget
Analyzer app.

6-10
 Superheterodyne Receiver Using RF Budget Analyzer App

The RF Blockset testbench calculates the spot noise figure. The calculation assumes a frequency-
independent system within a given bandwidth. To simulate a frequency-independent system and
calculate the correct noise figure value, you must reduce the broad bandwidth of 20 MHz to a narrow
bandwidth.

To do so, first, stop all simulations. Double-click the RF Measurement Unit subsystem. This action
opens the RF measurement unit parameters. In the Measured Quantity parameter drop-down list,
select NF (noise figure). In the Parameters tab, set the Baseband bandwidth (Hz) parameter to
2000 Hz. Click Apply. To learn more about how to manipulate noise figure verification, click the
Instructions tab.

Run the model again to check the noise figure value. The testbench noise figure value matches the
cascade noise figure value from the RF Budget Analyzer app.

IP3 measurements rely on the creation and measurement of intermodulation tones that are usually
small in amplitude and may be below the DUT's noise floor. For accurate IP3 measurements, clear the
Simulate noise check box.

To verify OIP3, stop all simulations. Open the dialog box of the RF Measurement Unit block. Clear
the Simulate noise (both stimulus and DUT internal) check box. Change the Measured

6-11
6 RF Toolbox Examples

Quantity parameter to IP3. Keep the IP Type as Output referred. To learn more about how to
manipulate OIP3 verification, click the Instructions tab. Click Apply.

Run the model. The testbench OIP3 value matches the cascade OIP3 value from the app.

To verify IIP3, stop all simulations. Open dialog box of the RF Measurement Unit block. Clear the
Simulate noise (both stimulus and DUT internal) check box. Change the Measured Quantity
parameter to IP3. Change the IP Type to Input referred. To learn more about how to manipulate
IIP3 verification, click the Instructions tab. Click Apply.

Run the model again to check the IIP3 value.

To sum up, in this example, you have seen how to build, analyze, and verify a superheterodyne
receiver. Similarly, you can build and analyze an RF chain programmatically in the MATLAB command
line or interactively using the RF Budget Analyzer app, and generate circuit envelope models for
system-level simulation.

References

[1] Hongbao Zhou, Bin Luo. " Design and budget analysis of RF receiver of 5.8GHz ETC reader"
Published at Communication Technology (ICCT), 2010 12th IEEE International Conference, Nanjing,
China, November 2010.

6-12
 Superheterodyne Receiver Using RF Budget Analyzer App

[2] Bin Luo, Peng Li. "Budget Analysis of RF Transceiver Used in 5.8GHz RFID Reader Based on the
ETC-DSRC National Specifications of China" Published at Wireless Communications, Networking and
Mobile Computing, WiCom '09. 5th International Conference, Beijing, China, September 2009.

See Also
RF Budget Analyzer

More About
• “Visualizing RF Budget Analysis over Bandwidth” on page 6-14
• “Bandpass Filter Response” on page 6-21

6-13
6 RF Toolbox Examples

Visualizing RF Budget Analysis over Bandwidth

This example shows how to programmatically perform an RF budget analysis of an RF receiver

system and visualize the computed budget results across the bandwidth of the input signal.

First, use amplifier, modulator, rfelement, and nport objects to specify the 2-port RF elements
in a design. Then compute RF budget results by cascading the elements together into an RF system
with rfbudget.

The rfbudget object enables design exploration and visualization at the MATLAB® command-line or
graphically in the RF Budget Analyzer app. It also enables automatic RF Blockset™ model and
measurement testbench generation.

Introduction

RF system designers typically begin a design process with budget specifications for the gain, noise
figure (NF), and nonlinearity (IP3) of the entire system.

MATLAB functionality supporting RF budget analysis makes it easy to visualize gain, NF and IP3
results at multiple frequencies throughout the bandwidth of the signal. You can:

• Programmatically build an rfbudget object out of 2-port RF elements.

• Use the Command Line display of the rfbudget object to view single-frequency budget results.
• Vectorize the input frequency of the rfbudget object and use MATLAB plot to visualize RF
budget results across the bandwidth of the input signal.

In addition, with an rfbudget object you can:

• Use export methods to generate MATLAB scripts, RF Blockset models, or measurement

testbenches in Simulink®.
• Use show command to copy an rfbudget object into the RF Budget Analyzer app.

Building Elements of RF Receiver

A basic RF receiver consists of an RF filter, an RF amplifier, a demodulator, an IF filter, and an IF

amplifier.

First build and parameterize each of the 2-port RF elements. Then use rfbudget to cascade the
elements with input frequency 2.1 GHz, input power -30 dBm, and input bandwidth 45 MHz.
f1 = nport('RFBudget_RF.s2p','RFBandpassFilter');

a1 = amplifier('Name','RFAmplifier', ...
'Gain',11.53, ...
'NF',1.53, ...
'OIP3',35);

d = modulator('Name','Demodulator', ...
'Gain',-6, ...
'NF',4, ...
'OIP3',50, ...
'LO',2.03e9, ...
'ConverterType','Down');

6-14
 Visualizing RF Budget Analysis over Bandwidth

f2 = nport('RFBudget_IF.s2p','IFBandpassFilter');

a2 = amplifier('Name','IFAmplifier', ...
'Gain',30, ...
'NF',8, ...
'OIP3',37);

b = rfbudget('Elements',[f1 a1 d f2 a2], ...

'InputFrequency',2.1e9, ...
'AvailableInputPower',-30, ...
'SignalBandwidth',45e6);

Visualize RF Budget Results in MATLAB

Scalar frequency results can be viewed simply by using MATLAB disp to see the results at the
Command Line. Each column of the budget shows the results of cascading only the elements of the
previous columns. Note that final column shows the RF budget results of the entire cascade.

disp(b)

rfbudget with properties:

Elements: [1x5 rf.internal.rfbudget.Element]

InputFrequency: 2.1 GHz
AvailableInputPower: -30 dBm
SignalBandwidth: 45 MHz
Solver: Friis
AutoUpdate: true

Analysis Results
OutputFrequency: (GHz) [ 2.1 2.1 0.07 0.07 0.07]
OutputPower: (dBm) [-31.53 -20 -26 -27.15 2.847]
TransducerGain: (dB) [-1.534 9.996 3.996 2.847 32.85]
NF: (dB) [ 1.533 3.064 3.377 3.611 7.036]
IIP2: (dBm) []
OIP2: (dBm) []
IIP3: (dBm) [ Inf 25 24.97 24.97 4.116]
OIP3: (dBm) [ Inf 35 28.97 27.82 36.96]
SNR: (dB) [ 65.91 64.38 64.07 63.83 60.41]

Plot RF Budget Results Versus Input Frequency

Use the budget's rfplot function to produce report-ready plots of cumulative RF budget results
versus a range of cascade input frequencies. Cumulative (i.e. terminated sub-cascade) results are
automatically computed to show the variation of the RF budget result through the entire design. Use
the data cursor of the figure window to interactively explore values at different frequencies at
different stages.

rfplot(b,'Pout')

6-15
6 RF Toolbox Examples

rfplot(b,'GainT')

6-16
 Visualizing RF Budget Analysis over Bandwidth

Plot RF Budget Network Parameter Results Versus Input Frequency

Use the RF budget smithplot/polar function to produce plots of cumulative RF budget sparameter
results versus a range of cascade input frequencies. Use smithplot function to view reflection
coefficients and polar to view reflection and transmission coefficients.

smithplot(b,1,1)

6-17
6 RF Toolbox Examples

polar(b,2,1)

6-18
 Visualizing RF Budget Analysis over Bandwidth

Easily Export to RF Blockset and Simulink

The rfbudget object has other useful MATLAB methods:

• exportScript - generate a MATLAB script that builds the current design

• exportRFBlockset - generate an RF Blockset model for simulation
• exportTestbench - generate a Simulink measurement testbench

Visualize RF Budget Results in App

Use the show command to copy a single-frequency rfbudget object into the RF Budget Analyzer
app. The Plot, Smith, and Polar button in the app, with its pull-down options, calls rfplot,
smithplot, and polar respectively.

In the app, the Export button copies the current design to an rfbudget object in the MATLAB
workspace. All of the other export methods of the RF budget object are available through the
pulldown options of the Export button.

show(b)

6-19
6 RF Toolbox Examples

Automatically Create Reports From MATLAB Files

If you have written a 'myfile.m' script that builds your design and visualizes it with rfplot
commands, try the publish('myfile.m') function at the command line (or click the Publish
button in the MATLAB editor). This automatically generates all figures and produces a report for your
colleagues, saved as an html file.

To save your design, first undock using the commands shown below and then use the Figure Toolbar
to pulldown the File Menu and save using File -> Save As and select the Save as type to png or pdf.
To redock the figure window into the app you can click the Dock affordance on the upper right corner
of the figure window.

h = findall(0,'type','figure','name','untitled');
set(h,'WindowStyle','normal')
set(h,'MenuBar','figure')
set(h,'ToolBar','auto')

See Also

More About
• “Superheterodyne Receiver Using RF Budget Analyzer App” on page 6-2
• “Bandpass Filter Response” on page 6-21

6-20
 Bandpass Filter Response

Bandpass Filter Response

This example shows how to compute the time-domain response of a simple bandpass filter. The eight
steps involved in computing the time-domain response of a simple bandpass filter are,

1 Use the classic image parameter design to assign inductance and capacitance values to the
bandpass filter.
2 Use the circuit, capacitor, and inductor objects with the add function to programmatically
construct a Butterworth circuit.
3 Use the setports function to define the circuit as a 2-port network.
4 Use the sparameters function to extract the S-parameters of the 2-port network over a wide
frequency range.
5 Use the s2tf function to compute the voltage transfer function from the input to the output.
6 Use the rational object to generate rational fits that capture the ideal RC circuit to a very high
degree of accuracy.
7 Use the randn function to create noise in order to create a noisy input voltage waveform.
8 Use the timeresp function to compute the transient response to a noisy input voltage
waveform.

Design Bandpass Filter Using Image Parameters

The image parameter design is a framework for analytically computing the values of the series and
parallel components in the passive filters. For more information on Image Parameters, see "Complete
Wireless Design" by Cotter W. Sayre, McGraw-Hill 2008 p. 331.

Figure 1: A Butterworth bandpass filter built out of two half-sections.

Generate the component values for a bandpass filter with a lower 3 dB cutoff frequency of 2.4 GHz
and an upper 3 dB cutoff frequency of 2.5 GHz.

Ro = 50;
f1C = 2400e6;
f2C = 2500e6;
Ls = (Ro / (pi*(f2C - f1C)))/2; % Ls1 and Ls2
Cs = 2*(f2C - f1C)/(4*pi*Ro*f2C*f1C); % Cs1 and Cs2

6-21
6 RF Toolbox Examples

Lp = 2*Ro*(f2C - f1C)/(4*pi*f2C*f1C); % Lp1 and Lp2

Cp = (1/(pi*Ro*(f2C - f1C)))/2; % Cp1 and Cp2

Programmatically Construct Circuit

Before building the circuit using the inductor and capacitor objects, nodes in the circuit are
numbered. This is shown in figure 1.

Figure 2: Node numbers added to the Butterworth bandpass filter.

Create a circuit object and populate it with the inductor and the capacitor objects using the
add function.
ckt = circuit('butterworthBPF');

add(ckt,[3 2],inductor(Ls)); % Ls1

add(ckt,[4 3],capacitor(Cs)); % Cs1
add(ckt,[5 4],capacitor(Cs)); % Cs2
add(ckt,[6 5],inductor(Ls)); % Ls2

add(ckt,[4 1],capacitor(Cp)); % Cp1

add(ckt,[4 1],inductor(Lp)); % Lp1
add(ckt,[4 1],inductor(Lp)); % Lp2
add(ckt,[4 1],capacitor(Cp)); % Cp2

Extract S-Parameters From 2-Port Network

To extract S-parameters from the circuit object, first use the setports function to define the circuit
as a 2-port network.
freq = linspace(2e9,3e9,101);

Use the sparameters function to extract the S-parameters at the frequencies of interest.
setports(ckt,[2 1],[6 1])
S = sparameters(ckt,freq);

Fit Transfer Function of Circuit to Rational Function

Use the s2tf function to generate a transfer function from the S-parameter object.
tfS = s2tf(S);

Use the rational object to fit the transfer function data to a rational function.

6-22
 Bandpass Filter Response

fit = rational(freq,tfS);

Verify Rational Fit Approximation

Use the freqresp function to verify that the rational fit approximation has reasonable behavior
outside both sides of the fitted frequency range.

widerFreqs = linspace(2e8,5e9,1001);
resp = freqresp(fit,widerFreqs);

Plot to visualize rational fit approximation. The rational fit behaves well outside the fitted frequency
range.

figure
semilogy(freq,abs(tfS),widerFreqs,abs(resp),'--','LineWidth',2)
xlabel('Frequency (Hz)');
ylabel('Magnitude');
legend('data','fit');
title('Rational Fit Approximation');

Construct Input Signal to Test Bandpass Filter

To test the bandpass filter, designed by the Image Parameter technique, a sinusoidal signal at 2.45
GHz is recovered from the noisy input signal. The noise input signal is generated by the inclusion of
zero-mean random noise and a blocker at 2.35 GHz to the input signal.

Construct an input and a noisy input signal with 8192 samples.

6-23
6 RF Toolbox Examples

fCenter = 2.45e9;
fBlocker = 2.35e9;
period = 1/fCenter;
sampleTime = period/16;
signalLen = 8192;
t = (0:signalLen-1)'*sampleTime; % 256 periods
input = sin(2*pi*fCenter*t); % Clean input signal
rng('default')
noise = randn(size(t)) + sin(2*pi*fBlocker*t);
noisyInput = input + noise; % Noisy input signal

Compute Transient Response to Input Signal

Use the timeresp function to compute the analytic solutions of the state-space.

output = timeresp(fit,noisyInput,sampleTime);

View Input Signal and Filter Response in Time Domain

Plot the input signal, noisy input signal, and the band pass filter output in a figure window.

xmax = t(end)/8;
figure
subplot(3,1,1)
plot(t,input)
axis([0 xmax -1.5 1.5])
title('Input')

subplot(3,1,2)
plot(t,noisyInput)
axis([0 xmax floor(min(noisyInput)) ceil(max(noisyInput))]);
title('Noisy Input');
ylabel('Amplitude (volts)');

subplot(3,1,3)
plot(t,output)
axis([0 xmax -1.5 1.5]);
title('Filter Output');
xlabel('Time (sec)');

6-24
 Bandpass Filter Response

View Input Signal and Filter Response in Frequency Domain

Overlaying the noisy input and the filter response in the frequency domain explains why the filtering
operation is successful. Both the blocker signal at 2.35 GHz and much of the noise are significantly
attenuated.

NFFT = 2^nextpow2(signalLen); % Next power of 2 from length of y

Y = fft(noisyInput,NFFT)/signalLen;
samplingFreq = 1/sampleTime;
f = samplingFreq/2*linspace(0,1,NFFT/2+1)';
O = fft(output,NFFT)/signalLen;

figure
subplot(2,1,1)
plot(freq,abs(tfS),'b','LineWidth',2)
axis([freq(1) freq(end) 0 1.1]);
legend('filter transfer function');
title('Transfer function of Bandpass filter');
ylabel('Magnitude');

subplot(2,1,2)
plot(f,2*abs(Y(1:NFFT/2+1)),'g',f,2*abs(O(1:NFFT/2+1)),'r','LineWidth',2)
axis([freq(1) freq(end) 0 1.1]);
legend('input+noise','output');
title('Filter characteristic and noisy input spectrum.');
xlabel('Frequency (Hz)');
ylabel('Magnitude (Volts)');

6-25
6 RF Toolbox Examples

To compute and display this bandpass filter response using RFCKT objects, see “Bandpass Filter
Response Using RFCKT Objects” on page 6-33.

See Also

More About
• “Design IF Butterworth Bandpass Filter” on page 6-183
• “Superheterodyne Receiver Using RF Budget Analyzer App” on page 6-2
• “Visualizing RF Budget Analysis over Bandwidth” on page 6-14

6-26
 MOS Interconnect and Crosstalk

MOS Interconnect and Crosstalk

This example shows how to build and simulate an RC tree circuit using the RF Toolbox™.

In "Asymptotic Waveform Evaluation for Timing Analysis" (IEEE Transactions on Computer-Aided

Design, Vol., 9, No. 4, April 1990), Pillage and Rohrer presented and simulated an RC tree circuit that
models signal integrity and crosstalk in low- to mid-frequency MOS circuit interconnect. This example
confirms their simulations using RF Toolbox software.

Their circuit, reproduced in the following figure, consists of 11 resistors and 12 capacitors. In the
paper, Pillage and Rohrer:

• Apply a ramp voltage input

• Compute transient responses
• Plot the output voltages across two different capacitors, C7 and C12.

Figure 1: An RC tree model of MOS interconnect with crosstalk.

With RF Toolbox, you can programmatically construct this circuit in MATLAB and perform signal
integrity simulations.

This example shows:

1 How to use circuit, resistor, and capacitor with the add function to programmatically
construct the circuit.
2 How to use clone, setports, and sparameters objects to calculate S-parameters for each
desired output over a wide frequency range.
3 How to use s2tf with Zsource = 0 and Zload = Inf to compute the voltage transfer function
from input to each desired output.

6-27
6 RF Toolbox Examples

4 How to use rationalfit function to produce rational-function approximations that capture the
ideal RC-circuit behavior to a very high degree of accuracy.
5 How to use timeresp function to compute the transient response to the input voltage waveform.

Insert Node Numbers Into Circuit Diagram

Before building the circuit using resistor and capacitor objects, we must number the nodes of
the circuit shown in figure 1.

Figure 2: The circuit drawn with node numbers

Programmatically Construct Circuit

Create a circuit and use the add function to populate the circuit with named resistor and
capacitor objects.

ckt = circuit('crosstalk');

add(ckt,[2 1],resistor(10,'R1'))
add(ckt,[2 0],capacitor(0.114e-12,'C1'))
add(ckt,[3 2],resistor(72,'R2'))
add(ckt,[3 0],capacitor(1.238e-12,'C2'))
add(ckt,[4 3],resistor(34,'R3'))
add(ckt,[4 0],capacitor(0.021e-12,'C3'))
add(ckt,[5 4],resistor(96,'R4'))
add(ckt,[5 0],capacitor(0.028e-12,'C4'))
add(ckt,[6 5],resistor(72,'R5'))
add(ckt,[6 0],capacitor(0.007e-12,'C5'))
add(ckt,[7 6],resistor(10,'R6'))
add(ckt,[7 0],capacitor(1.048e-12,'C6'))
add(ckt,[8 7],resistor(120,'R7'))
add(ckt,[8 0],capacitor(0.47e-12,'C7'))

6-28
 MOS Interconnect and Crosstalk

add(ckt,[12 8],resistor(24,'R8'))
add(ckt,[12 0],capacitor(0.2e-12,'C8'))

add(ckt,[10 2],resistor(48,'R9'))
add(ckt,[10 0],capacitor(0.007e-12,'C9'))
add(ckt,[11 10],resistor(24,'R10'))
add(ckt,[11 0],capacitor(0.2e-12,'C10'))

add(ckt,[9 8],capacitor(0.1e-12,'C11'))
add(ckt,[9 0],resistor(1000,'R11'))
add(ckt,[9 0],capacitor(1e-12,'C12'))

Simulation Setup

The input signal used by Pillage and Rohrer is a voltage ramp from 0 to 5 volts with a rise time of one
nanosecond and a duration of ten nanoseconds. The following MATLAB code models this signal with
1000 timepoints with a sampleTime of 0.01 nanoseconds.

The following MATLAB code also uses the logspace function to generate a vector of 101
logarithmically spaced analysis frequencies between 1 Hz and 100 GHz. Specifying a wide set of
frequency points improves simulation accuracy.
sampleTime = 1e-11;
t = (0:1000)'*sampleTime;
input = [(0:100)'*(5/100); (101:1000)'*0+5];
freq = logspace(0,11,101)';

Calculate S-Parameters For Each 2-Port Network

To calculate the response across both the C7 and C12 capacitors, two separate S-parameter
calculations must be made: first, assuming the C7 capacitor represents the output port, and second,
assuming the C12 capacitor represents the output port. To calculate the S-parameters for each setup:
1 Copy the original circuit ckt using the clone function.
2 Define the input and output ports of the network using the setports function.
3 Calculate the S-parameters using the sparameters object.

Calculate S-parameters with C7 capacitor represents the output port.

cktC7 = clone(ckt);
setports(cktC7,[1 0],[8 0])
S_C7 = sparameters(cktC7,freq);

Calculate S-parameters with C12 capacitor represents the output port.

cktC12 = clone(ckt);
setports(cktC12,[1 0],[9 0])
S_C12 = sparameters(cktC12,freq);

Simulate Each 2-Port Network

To simulate each network:

1 The s2tf function, with option = 2, computes the gain from the source voltage to the output
voltage. It allows arbitrary source and load impedances, in this case Zsource = 0 and Zload =
Inf. The resulting transfer functions tfC7 and tfC12 are frequency-dependent data vectors that
can be fit with rational-function approximation.

6-29
6 RF Toolbox Examples

2 The rationalfit function generates high-accuracy rational-function approximations. The

resulting approximations match the networks to machine accuracy.
3 The timeresp function computes the analytic solution to the state-space equations defined by a
rational-function approximation. This methodology is fast enough to enable one to push a million
bits through a channel.

Simulate cktC7 circuit.

tfC7 = s2tf(S_C7,0,Inf,2);
fitC7 = rationalfit(freq,tfC7);
outputC7 = timeresp(fitC7,input,sampleTime);

Simulate cktC12 circuit.

tfC12 = s2tf(S_C12,0,Inf,2);
fitC12 = rationalfit(freq,tfC12);
outputC12 = timeresp(fitC12,input,sampleTime);

Plot Transient Responses

The outputs match Figures 23 and 24 of the Pillage and Rohrer paper. Plot ramp response of low- to
mid-frequency MOS circuit interconnect with crosstalk.

figure
plot(t,input,t,outputC7,'LineWidth',2)
axis([0 2.5e-9 0 5.5]);
title('Ramp Response of Low- to Mid-Frequency MOS Circuit Interconnect with Crosstalk');
xlabel('Time (sec)');
ylabel('Voltage (volts)');
legend('Vinput','V(C7)','Location','NorthWest');

6-30
 MOS Interconnect and Crosstalk

Plot crosstalk in low- to mid-frequency MOS circuit interconnect with ramp input.

figure
plot(t,input,t,outputC12,'LineWidth',2)
axis([0 5e-9 0 .5])
title('Crosstalk in Low- to Mid-Frequency MOS Circuit Interconnect with Ramp Input')
xlabel('Time (sec)')
ylabel('Voltage (volts)')
legend('Vinput','V(C12)','Location','NorthEast')

6-31
6 RF Toolbox Examples

Verify Rational Fit Outside Fit Range

Though not shown in this example, you can also use the freqresp function to check the behavior of
rationalfit function well outside the specified frequency range. The fit outside the specified range
can sometimes cause surprising behavior, especially if frequency data near 0 Hz (DC) is not provided.

To perform this check for the rational-function approximation in this example, uncomment and run
the following MATLAB code.

% widerFreqs = logspace(0,12,1001);
% respC7 = freqresp(fitC7,widerFreqs);
% figure
% loglog(freq,abs(tfC7),'+',widerFreqs,abs(respC7))
% respC12 = freqresp(fitC12,widerFreqs);
% figure
% loglog(freq,abs(tfC12),'+',widerFreqs,abs(respC12))

For example on how to build and simulate this RC tree circuit using RFCKT objects, see “MOS
Interconnect and Crosstalk Using RFCKT Objects” on page 6-39.

See Also

More About
• “MOS Interconnect and Crosstalk Using RFCKT Objects” on page 6-39

6-32
 Bandpass Filter Response Using RFCKT Objects

Bandpass Filter Response Using RFCKT Objects

This example shows how to compute the time-domain response of a simple bandpass filter:

1 Choose inductance and capacitance values using the classic image parameter design method.
2 Use rfckt.seriesrlc, rfckt.shuntrlc, and rfckt.cascade objects to programmatically
construct a Butterworth circuit as a 2-port network.
3 Use analyze to extract the S-parameters of the 2-port network over a wide frequency range.
4 Use s2tf function to compute the voltage transfer function from the input to the output.
5 Use rationalfit function to generate rational fits that capture the ideal RC circuit to a very
high degree of accuracy.
6 Create a noisy input voltage waveform.
7 Use timeresp function to compute the transient response to a noisy input voltage waveform.

Design Bandpass Filter by Image Parameters

The image parameter design method is a framework for analytically computing the values of the
series and parallel components in passive filters. For more information on this method, see "Complete
Wireless Design" by Cotter W. Sayre, McGraw-Hill 2008 p. 331.

Figure 1: A Butterworth bandpass filter built out of two half-sections.

The following MATLAB® code generates component values for a bandpass filter with a lower 3-dB
cutoff frequency of 2.4 GHz and an upper 3 dB cutoff frequency of 2.5 GHz.

Ro = 50;
f1C = 2400e6;
f2C = 2500e6;

Ls = (Ro / (pi*(f2C - f1C)))/2;

Cs = 2*(f2C - f1C)/(4*pi*Ro*f2C*f1C);

Lp = 2*Ro*(f2C - f1C)/(4*pi*f2C*f1C);
Cp = (1/(pi*Ro*(f2C - f1C)))/2;

6-33
6 RF Toolbox Examples

Programmatically Construct Circuit as 2-Port Network

The L and C building blocks are formed by selecting appropriate values with the rfckt.shuntrlc
object shown in Figure 2 or the rfckt.seriesrlc object shown in Figure 3. The building blocks are
then connected together with rfckt.cascade as shown in Figure 4.

Figure 2: The 2-port network created by the rfckt.shuntrlc object

Figure 3: The 2-port network created by the rfckt.seriesrlc object

Figure 4: Connecting 2-port networks with the rfckt.cascade object

Seg1 = rfckt.seriesrlc('L',Ls,'C',Cs);
Seg2 = rfckt.shuntrlc('L',Lp,'C',Cp);
Seg3 = rfckt.shuntrlc('L',Lp,'C',Cp);
Seg4 = rfckt.seriesrlc('L',Ls,'C',Cs);

cktBPF = rfckt.cascade('Ckts',{Seg1,Seg2,Seg3,Seg4});

Extract S-Parameters From 2-Port Network

The analyze function extracts the S-parameters from a circuit over a specified vector of frequencies.
This example provides a set of frequencies that spans the passband of the filter and analyzes with the
default 50-Ohm reference, source impedance, and load impedances. Next, the s2tf function

6-34
 Bandpass Filter Response Using RFCKT Objects

computes the voltage transfer function across the S-parameter model of the circuit. Finally, we
generate a high-accuracy rational approximation using the rationalfit function. The resulting
approximation matches the network to machine accuracy.

freq = linspace(2e9,3e9,101);
analyze(cktBPF,freq);
sparams = cktBPF.AnalyzedResult.S_Parameters;
tf = s2tf(sparams);
fit = rationalfit(freq,tf);

Verify that Rational Fit Tends to Zero

Use the freqresp function to verify that the rational fit approximation has reasonable behavior
outside both sides of the fitted frequency range.

widerFreqs = linspace(2e8,5e9,1001);
resp = freqresp(fit,widerFreqs);

figure
semilogy(freq,abs(tf),widerFreqs,abs(resp),'--','LineWidth',2)
xlabel('Frequency (Hz)')
ylabel('Magnitude')
legend('data','fit')
title('The rational fit behaves well outside the fitted frequency range.')

6-35
6 RF Toolbox Examples

Construct Input Signal to Test Band Pass Filter

This bandpass filter should be able to recover a sinusoidal signal at 2.45 GHz that is made noisy by
the inclusion of zero-mean random noise and a blocker at 2.35 GHz. The following MATLAB code
constructs such a signal from 4096 samples.

fCenter = 2.45e9;
fBlocker = 2.35e9;
period = 1/fCenter;
sampleTime = period/16;
signalLen = 8192;
t = (0:signalLen-1)'*sampleTime; % 256 periods

input = sin(2*pi*fCenter*t); % Clean input signal

rng('default')
noise = randn(size(t)) + sin(2*pi*fBlocker*t);
noisyInput = input + noise; % Noisy input signal

Compute Transient Response to Input Signal

The timeresp function computes the analytic solution to the state-space equations defined by the
rational fit and the input signal.

output = timeresp(fit,noisyInput,sampleTime);

View Input Signal and Filter Response in Time Domain

Plot the input signal, noisy input signal, and the band pass filter output in a figure window.

xmax = t(end)/8;
figure
subplot(3,1,1)
plot(t,input)
axis([0 xmax -1.5 1.5])
title('Input')

subplot(3,1,2)
plot(t,noisyInput)
axis([0 xmax floor(min(noisyInput)) ceil(max(noisyInput))])
title('Noisy Input')
ylabel('Amplitude (volts)')

subplot(3,1,3)
plot(t,output)
axis([0 xmax -1.5 1.5])
title('Filter Output')
xlabel('Time (sec)')

6-36
 Bandpass Filter Response Using RFCKT Objects

View Input Signal and Filter Response in Frequency Domain

Overlaying the noisy input and the filter response in the frequency domain explains why the filtering
operation is successful. Both the blocker signal at 2.35 GHz and much of the noise is significantly
attenuated.

NFFT = 2^nextpow2(signalLen); % Next power of 2 from length of y

Y = fft(noisyInput,NFFT)/signalLen;
samplingFreq = 1/sampleTime;
f = samplingFreq/2*linspace(0,1,NFFT/2+1)';
O = fft(output,NFFT)/signalLen;

figure
subplot(2,1,1)
plot(freq,abs(tf),'b','LineWidth',2)
axis([freq(1) freq(end) 0 1.1])
legend('filter transfer function')
ylabel('Magnitude')

subplot(2,1,2)
plot(f,2*abs(Y(1:NFFT/2+1)),'g',f,2*abs(O(1:NFFT/2+1)),'r','LineWidth',2)
axis([freq(1) freq(end) 0 1.1])
legend('input+noise','output')
title('Filter characteristic and noisy input spectrum.')
xlabel('Frequency (Hz)')
ylabel('Magnitude (Volts)')

6-37
6 RF Toolbox Examples

See Also

More About
• “Bandpass Filter Response” on page 6-21
• “Operations with RF Circuit Objects” on page 6-173

6-38
 MOS Interconnect and Crosstalk Using RFCKT Objects

MOS Interconnect and Crosstalk Using RFCKT Objects

This example shows how to build and simulate an RC tree circuit using the RF Toolbox™.

In "Asymptotic Waveform Evaluation for Timing Analysis" (IEEE Transactions on Computer-Aided

Design, Vol., 9, No. 4, April 1990), Pillage and Rohrer presented and simulated an RC tree circuit that
models signal integrity and crosstalk in low- to mid-frequency MOS circuit interconnect. This example
confirms their simulations using RF Toolbox software.

Their circuit, reproduced in the following figure, consists of 11 resistors and 12 capacitors. In the
paper, Pillage and Rohrer:

• Apply a ramp voltage input

• Compute transient responses
• Plot the output voltages across two different capacitors, C7 and C12.

Figure 1: RC tree model of MOS interconnect with crosstalk.

With RF Toolbox software, you can programmatically construct this circuit in MATLAB and perform
signal integrity simulations.

This example shows:

1 How to use rfckt.seriesrlc, rfckt.shuntrlc, rfckt.series, and rfckt.cascade object
to programmatically construct the circuit as two different networks, depending on the desired
output.
2 How to use analyze function to extract the S-parameters for each 2-port network over a wide
frequency range.
3 How to use s2tf function with Zsource = 0 and Zload = Inf to compute the voltage transfer
function from input to each desired output.

6-39
6 RF Toolbox Examples

4 How to use rationalfit function to produce rational-function approximations that capture the
ideal RC-circuit behavior to a very high degree of accuracy.
5 How to use timeresp function to compute the transient response to the input voltage waveform.

Redraw Circuit as Distinct 2-Port Networks

To duplicate both output plots, RF Toolbox calculates the output voltage across C7 and C12. To that
end, the circuit must be expressed as two distinct 2-port networks, each with the appropriate
capacitor at the output. Figure 2 shows the 2-port configuration for computing the voltage across C7.
Figure 3 shows the configuration for C12. Both 2-port networks retain the original circuit topology,
and share much of the same structure.

Figure 2: The circuit drawn as a 2-port network with output across C7.

Figure 3: The circuit drawn as a 2-port network with output across C12.

Using RLC Building Blocks

All of the building blocks are formed by selecting appropriate values with the rfckt.shuntrlc
object shown in Figure 4 or the rfckt.seriesrlc object shown in Figure 5. The 2-port building
blocks are then connected using rfckt.cascade object as shown in Figure 6 or rfckt.series
object as shown in Figure 7.

6-40
 MOS Interconnect and Crosstalk Using RFCKT Objects

Figure 4: The 2-port network created using the rfckt.shuntrlc object.

Figure 5: The 2-port network created using the rfckt.seriesrlc object.

6-41
6 RF Toolbox Examples

Figure 6: Connect 2-port networks with the rfckt.cascade object.

Figure 7: Connect 2-port networks with the rfckt.series object.

Shared Pieces of 2-Port Networks

The following MATLAB code constructs the portion of the network shared between the two variants.

6-42
 MOS Interconnect and Crosstalk Using RFCKT Objects

R1 = rfckt.seriesrlc('R',10);
C1 = rfckt.shuntrlc('C',0.114e-12);
R9 = rfckt.shuntrlc('R',48);
C9 = rfckt.shuntrlc('C',0.007e-12);
R10 = rfckt.shuntrlc('R',24);
C10 = rfckt.shuntrlc('C',0.2e-12);
R10C10 = rfckt.series('Ckts',{R10,C10});
C9R10C10 = rfckt.cascade('Ckts',{C9,R10C10});
R9C9R10C10 = rfckt.series('Ckts',{R9,C9R10C10});
R2 = rfckt.seriesrlc('R',72);
C2 = rfckt.shuntrlc('C',1.238e-12);
R3 = rfckt.seriesrlc('R',34);
C3 = rfckt.shuntrlc('C',0.021e-12);
R4 = rfckt.seriesrlc('R',96);
C4 = rfckt.shuntrlc('C',0.028e-12);
R5 = rfckt.seriesrlc('R',72);
C5 = rfckt.shuntrlc('C',0.007e-12);
R6 = rfckt.seriesrlc('R',10);
C6 = rfckt.shuntrlc('C',1.048e-12);
R7 = rfckt.seriesrlc('R',120);
R8 = rfckt.shuntrlc('R',24);
C8 = rfckt.shuntrlc('C',0.2e-12);
R8C8 = rfckt.series('Ckts',{R8,C8});
sharedckt = rfckt.cascade('Ckts', ...
{R1,C1,R9C9R10C10,R2,C2,R3,C3,R4,C4,R5,C5,R6,C6,R7,R8C8});

% Additional shared building blocks used in both 2-port networks.

C7 = rfckt.shuntrlc('C',0.47e-12);
R11C12 = rfckt.shuntrlc('R',1000,'C',1e-12);

Construct Each 2-Port Network

Figure 2 shows that constructing a 2-port network with an output port across C7 requires creating
C11 using rfckt.shuntrlc object, then combining C11 with R11 and C12 using rfckt.series
object, and finally combining C11R11C12 with the rest of the network and C7 using rfckt.cascade
object.

Similarly, Figure 3 shows that constructing a 2-port network with an output port across C12 requires
creating another version of C11 (C11b) using rfckt.seriesrlc object and combining all the parts
together using rfckt.cascade object.

Construct shunt RLC circuit.

C11 = rfckt.shuntrlc('C',0.1e-12);
C11R11C12 = rfckt.series('Ckts',{C11,R11C12});
cktC7 = rfckt.cascade('Ckts',{sharedckt,C11R11C12,C7});

Construct series RLC circuit.

C11b = rfckt.seriesrlc('C',0.1e-12);
cktC12 = rfckt.cascade('Ckts',{sharedckt,C7,C11b,R11C12});

Simulation Setup

The input signal used by Pillage and Rohrer is a voltage ramp from 0 to 5 volts with a rise time of one
nanosecond and a duration of ten nanoseconds. The following MATLAB code models this signal with
1000 timepoints with a sampleTime of 0.01 nanoseconds.

6-43
6 RF Toolbox Examples

The following MATLAB code also uses the logspace function to generate a vector of 101
logarithmically spaced analysis frequencies between 1 Hz and 100 GHz. Specifying a wide set of
frequency points improves simulation accuracy.

sampleTime = 1e-11;
t = (0:1000)'*sampleTime;
input = [(0:100)'*(5/100); (101:1000)'*0+5];
freq = logspace(0,11,101)';

Simulate Each 2-Port Network

To simulate each network:

1 The analyze function extracts S-parameters over the specified frequency range.
2 The s2tf function, with option = 2, computes the gain from the source voltage to the output
voltage. It allows arbitrary source and load impedances, in this case Zsource = 0 and Zload =
Inf. The resulting transfer functions tfC7 and tfC12 are frequency-dependent data vectors that
can be fit with rational-function approximation.
3 The rationalfit function generates high-accuracy rational-function approximations. The
resulting approximations match the networks to machine accuracy.
4 The timeresp function computes the analytic solution to the state-space equations defined by a
rational-function approximation. This methodology is fast enough to enable one to push a million
bits through a channel.

Simulate cktC7 circuit.

analyze(cktC7,freq);
sparamsC7 = cktC7.AnalyzedResult.S_Parameters;
tfC7 = s2tf(sparamsC7,50,0,Inf,2);
fitC7 = rationalfit(freq,tfC7);
outputC7 = timeresp(fitC7,input,sampleTime);

Simulate cktC12 circuit.

analyze(cktC12,freq);
sparamsC12 = cktC12.AnalyzedResult.S_Parameters;
tfC12 = s2tf(sparamsC12,50,0,Inf,2);
fitC12 = rationalfit(freq,tfC12);
outputC12 = timeresp(fitC12,input,sampleTime);

Plot Transient Responses

The outputs match Figures 23 and 24 of the Pillage and Rohrer paper. Plot the ramp response of low-
to mid-frequency MOS circuit interconnect with crosstalk.

figure
plot(t,input,t,outputC7,'LineWidth',2)
axis([0 2.5e-9 0 5.5]);
title('Ramp Response of Low- to Mid-Frequency MOS Circuit Interconnect with Crosstalk');
xlabel('Time (sec)');
ylabel('Voltage (volts)');
legend('Vinput','V(C7)','Location','NorthWest');

6-44
 MOS Interconnect and Crosstalk Using RFCKT Objects

Plot the crosstalk in low- to mid-frequency MOS circuit interconnect with ramp input.

figure
plot(t,input,t,outputC12,'LineWidth',2)
axis([0 5e-9 0 .5]);
title('Crosstalk in Low- to Mid-frequency MOS Circuit Interconnect with Ramp Input');
xlabel('Time (sec)');
ylabel('Voltage (volts)');
legend('Vinput','V(C12)','Location','NorthEast');

6-45
6 RF Toolbox Examples

Verify Rational Fit Outside Fit Range

Though not shown in this example, you can also use the freqresp function to check the behavior of
rationalfit function well outside the specified frequency range. The fit outside the specified range
can sometimes cause surprising behavior, especially if frequency data near 0 Hz (DC) was not
provided.

To perform this check for the rational-function approximation in this example, uncomment and run
the following MATLAB code.

% widerFreqs = logspace(0,12,1001);
% respC7 = freqresp(fitC7,widerFreqs);
% figure
% loglog(freqs,abs(tfC7),'+',widerFreqs,abs(respC7))
% respC12 = freqresp(fitC12,widerFreqs);
% figure
% loglog(freqs,abs(tfC12),'+',widerFreqs,abs(respC12))

See Also

More About
• “MOS Interconnect and Crosstalk” on page 6-27

6-46
 Model and Analyze High-Speed Backplanes

Model and Analyze High-Speed Backplanes

This five-part tutorial shows you how to model and analyze high-speed backplanes using
sparameters and rationalfit functions. High-speed backplanes, as shown in the diagram, are
vital in enabling data transfer between multiple boards connected to a larger system. In this example,
you model a high-speed backplane using S-parameters and fit the S-parameter data for analysis.

Rational Fitting

RF Toolbox™ uses the rationalfit function to fit data defined in the frequency domain with an
equivalent Laplace transfer function. Using rational function fitting you can create simple models for
the required accuracy, model order reduction, zero phase on extrapolation to DC, and causal
modeling system.

This type of modeling is useful to signal integrity engineers, whose goal is to reliably connect high-
speed semiconductor devices with multi-Gbps serial data streams across backplanes and printed
circuit boards.

Compared to traditional techniques such as linear interpolation, rational function fitting provides
more insight into the physical characteristics of a high-speed backplane. It enables you to trade off
between complexity and accuracy using model order reduction. For a given accuracy, rational
functions are less complex than other types of models such as FIR filters generated by IFFT
techniques. In addition, rational function models inherently constrain the phase to be zero on
extrapolation to DC. Methods that do not provide as much insight into physical characteristics
require elaborate constraint algorithms to force the extrapolated phase to zero at DC.

6-47
6 RF Toolbox Examples

Design Workflow

To model, analyze, and export high-speed backplanes, follow these five steps.

1 Import S-parameters and port reduction: Use RF Toolbox to import N-port S-parameters
representing high-speed backplane channels to model the channels and the crosstalk between
channels. This example imports 16-port S-parameters that represent a 16-port high-speed
backplane and reduces them to 4-port S-parameters for analysis. For more information, see “Use
S-parameters with Port Reduction” on page 6-49.
2 Fit S-parameters: Fit 4-port S-parameters from the step 1 using the rataionalfit function to
create a simpler high-speed backplane model for accuracy, model order reduction, and zero
phase on extrapolation to DC. For more information, see “Fit S-Parameters with a Rational
Function” on page 6-53.
3 Analyze your high-speed backplane: Calculate the time-domain reflectometry and time-domain
transmission of a differential high-speed backplane channel for further analysis. For more
information, see “Compute Differential TDR and TDT” on page 6-60.
4 Build your model in Simulink: Build your model in Simulink® to simulate your high-speed
backplane and study the impact of the differential backplane on the random input signal. For
more information, see “Build Simulink Model from Rational Function” on page 6-63.
5 Export your model to Verilog-A: Export your high-speed backplane model in S-parameters to a
Verilog-A module that models the high-level behavior of a high-speed backplane. For more
information, see “Export Verilog-A module from Rational Function” on page 6-67.

See Also
rationalfit | rational | rfmodel.rational

6-48
 Use S-parameters with Port Reduction

Use S-parameters with Port Reduction

This example shows how to use RF Toolbox™ to import N-port S-parameters representing high-speed
backplane channels, and converts 16-port S-parameters to 4-port S-parameters to model the channels
and the crosstalk between the channels.

Figure 1: 16-Port differential backplane

Read Single-Ended 16-Port S-Parameters

Read a Touchstone® data file into an sparameters object. The data in this file are the 50-ohm S-
parameters of a 16-port differential backplane designed for a 2-Gbps high-speed signal, shown in
Figure 1, measured at 1496 frequencies ranging from 50 MHz to 15 GHz.
filename = 'default.s16p';
backplane = sparameters(filename)

backplane =
sparameters with properties:

Impedance: 50
NumPorts: 16
Parameters: [16x16x1496 double]
Frequencies: [1496x1 double]

freq = backplane.Frequencies;

Convert 16-Port S-Parameters to 4-Port S-Parameters to Model Differential Channel

Use the snp2smp function to convert 16-port S-parameters to 4-port S-parameters that represent the
first differential channel. The port index of this differential channel, N2M, specifies how the ports of
the 16-port S-parameters map to the ports of the 4-port S-parameters, is [1 16 2 15]. (The port
indices of the second, third and fourth channels are [3 14 4 13], [5 12 6 11] and [7 10 8 9],
respectively). The other 12 ports, [3 4 5 6 7 8 9 10 11 12 13 14], are terminated with the
characteristic Impedance specified by the sparameters object. Then, create an sparameters
object with 4-port S-parameters for the first differential channel.
(Port 1) (Port 16)
Port 1 > ----->| |<----- < Port 2
| DUT |

6-49
6 RF Toolbox Examples

Port 3 > ----->| |<----- < Port 4

(Port 2) (Port 15)

n2m = [1 16 2 15];
z0 = backplane.Impedance;
first4portdata = snp2smp(backplane.Parameters,z0,n2m,z0);
first4portsparams = sparameters(first4portdata,freq,z0)

first4portsparams =
sparameters with properties:

Impedance: 50
NumPorts: 4
Parameters: [4x4x1496 double]
Frequencies: [1496x1 double]

Plot S21 and S43 of the first differential channel.

figure
rfplot(first4portsparams,2,1)
hold on
rfplot(first4portsparams,4,3,'-r')

% % If you want to write the 4-port S-parameters of the differential

% % channel into a |.s4p| file, then uncomment the line below.
%
% rfwrite(first4portsparams,'firstchannel.s4p')

6-50
 Use S-parameters with Port Reduction

Convert 16-Port S-Parameters to 4-Port S-Parameters to Model Crosstalk Between Two

Differential Channels

Use the snp2smp function to convert 16-port S-parameters to 4-port S-parameters that represent the
crosstalk between port [3 4] and port [16 15]. As shown in Figure 1, these ports are on different
channels. The other 12 ports, [1 2 5 6 7 8 9 10 11 12 13 14], are terminated with the
characteristic Impedance specified by the sparameters object. Then, create an sparameters
object with 4-port S-parameters for the crosstalk.

(Port 3) (Port 16)

Port 1 > ----->| |<----- < Port 2
| DUT |
Port 3 > ----->| |<----- < Port 4
(Port 4) (Port 15)

n2m = [3 16 4 15];
crosstalk4portdata = snp2smp(backplane.Parameters,z0,n2m,z0);
crosstalk4portsparams = sparameters(crosstalk4portdata,freq,z0)

crosstalk4portsparams =
sparameters with properties:

Impedance: 50
NumPorts: 4
Parameters: [4x4x1496 double]
Frequencies: [1496x1 double]

Plot S21, S43, S12 and S34 to show the crosstalk between these two channels.

figure
rfplot(crosstalk4portsparams,2,1)
hold on
rfplot(crosstalk4portsparams,4,3,'-r')
rfplot(crosstalk4portsparams,1,2,'-k')
rfplot(crosstalk4portsparams,3,4,'-g')

6-51
6 RF Toolbox Examples

% % If you want to write the 4-port S-parameters of the crosstalk into an

% % .s4p file, then uncomment the line below.
%
% rfwrite(crosstalk4portsparams,'crosstalk.s4p')

See Also

More About
• “Fit S-Parameters with a Rational Function” on page 6-53
• “Compute Differential TDR and TDT” on page 6-60
• “Export Verilog-A module from Rational Function” on page 6-67

6-52
 Fit S-Parameters with a Rational Function

Fit S-Parameters with a Rational Function

This example shows how to use RF Toolbox™ to model a differential high-speed backplane channel
using rational functions. This type of model is useful to signal integrity engineers, whose goal is to
reliably connect high-speed semiconductor devices with, multi-Gbps serial data streams across
backplanes and printed circuit boards.

Compared to traditional techniques such as linear interpolation, rational function fitting provides
more insight into the physical characteristics of a high-speed backplane. It provides a means, called
model order reduction, of making a trade-off between complexity and accuracy. For a given accuracy,
rational functions are less complex than other types of models such as FIR filters generated by IFFT
techniques. In addition, rational function models inherently constrain the phase to be zero on
extrapolation to DC. Less physically-based methods require elaborate constraint algorithms in order
to force the extrapolated phase to zero at DC.

Figure 1: A differential high-speed backplane channel

Read Single-Ended 4-Port S-Parameters and Convert Them to Differential 2-Port S-

Parameters

Read a Touchstone® data file, default.s4p, into an sparameters object. The parameters in this
data file are the 50-ohm S-parameters of the single-ended 4-port passive circuit shown in Figure 1, at
1496 frequencies ranging from 50 MHz to 15 GHz. Then, get the single-ended 4-port S-parameters
and use the matrix conversion function s2sdd to convert them to differential 2-port S-parameters.
Finally, plot the differential S11 parameter on a Smith chart.

6-53
6 RF Toolbox Examples

filename = 'default.s4p';
backplane = sparameters(filename);
data = backplane.Parameters;
freq = backplane.Frequencies;
z0 = backplane.Impedance;

Convert to 2-port differential S-parameters.

diffdata = s2sdd(data);
diffz0 = 2*z0;

By default, s2sdd expects ports 1 and 3 to be input and ports 2 and 4 to be output. However if your
data has ports 1 and 2 as input and ports 3 and 4 as output, then use 2 as the second input argument
to s2sdd function to specify this alternate port arrangement. For example, diffdata =
s2sdd(data,2);

diffsparams = sparameters(diffdata,freq,diffz0)

diffsparams =
sparameters with properties:

Impedance: 100
NumPorts: 2
Parameters: [2x2x1496 double]
Frequencies: [1496x1 double]

figure
smithplot(diffsparams,1,1)

6-54
 Fit S-Parameters with a Rational Function

Compute Transfer Function and Its Rational Function Object Representation

First, use the s2tf function to compute the differential transfer function. Then, use the rational
function to compute the analytical form of the transfer function and store it in an object. The
rational function fits a rational function object to the specified data over the specified frequencies.
The run time depends on the computer, the fitting tolerance, the number of data points, among other
factors.

difftransfunc = s2tf(diffdata,diffz0,diffz0,diffz0);
rationalfunc = rational(freq,difftransfunc,Tolerance=-60)

rationalfunc =
rational with properties:

NumPorts: 1
NumPoles: 386
Poles: [386x1 double]
Residues: [1x1x386 double]
DirectTerm: 0
ErrDB: -60.9431

npoles = rationalfunc.NumPoles;
fprintf('The derived rational function contains %d poles.\n',npoles);

The derived rational function contains 386 poles.

6-55
6 RF Toolbox Examples

Validate Differential-Mode Frequency Response

Use the freqresp method of the rational object to get the frequency response of the rational
function object. Then, create a plot to compare the frequency response of the rational function object
and that of the original data.

freqsforresp = linspace(0,20e9,2000)';
resp = freqresp(rationalfunc,freqsforresp);

figure
subplot(2,1,1)
plot(freq*1.e-9,20*log10(abs(difftransfunc)),'r',freqsforresp*1.e-9, ...
20*log10(abs(resp)),'b--','LineWidth',2)
title(sprintf('Rational Fitting with %d poles',npoles),'FontSize',12)
ylabel('Magnitude (decibels)')
xlabel('Frequency (GHz)')
legend('Original data','Fitting result')

subplot(2,1,2)
origangle = unwrap(angle(difftransfunc))*180/pi;
plotangle = unwrap(angle(resp))*180/pi;
plot(freq*1.e-9,origangle,'r',freqsforresp*1.e-9,plotangle,'b--', ...
'LineWidth',2)
ylabel('Phase (deg.)')
xlabel('Frequency (GHz)')
legend('Original data','Fitting result')

6-56
 Fit S-Parameters with a Rational Function

Calculate and Plot Differential Input and Output Signals of High-Speed Backplane

Generate a random 2 Gbps pulse signal. Then, use the timeresp method of the rational object to
compute the response of the rational function object to the random pulse. Finally, plot the input and
output signals of the rational function model that represents the differential circuit.

datarate = 2*1e9; % Data rate: 2 Gbps

samplespersymb = 100;
pulsewidth = 1/datarate;
ts = pulsewidth/samplespersymb;
numsamples = 2^17;
numplotpoints = 10000;

t_in = double((1:numsamples)')*ts;
input = sign(randn(1,ceil(numsamples/samplespersymb)));
input = repmat(input,[samplespersymb, 1]);
input = input(:);
[output,t_out] = timeresp(rationalfunc,input,ts);

figure
subplot(2,1,1)
plot(t_in(1:numplotpoints)*1e9,input(1:numplotpoints),'LineWidth',2)
title([num2str(datarate*1e-9),' Gbps signal'],'FontSize',12)
ylabel('Input signal')
xlabel('Time (ns)')
axis([-inf,inf,-1.5,1.5])

subplot(2,1,2)
plot(t_out(1:numplotpoints)*1e9,output(1:numplotpoints),'LineWidth',2)
ylabel('Output signal')
xlabel('Time (ns)')
axis([-inf,inf,-1.5,1.5])

6-57
6 RF Toolbox Examples

Plot Eye Diagram of 2 Gbps Output Signal

Estimate and remove the delay from the output signal and create an eye diagram by using
Communications Toolbox™ functions.

if ~isempty(which('eyediagram'))
ignoreBits = 1500;
eyediagram(output(ignoreBits:end),2*samplespersymb,2/datarate)
end

6-58
 Fit S-Parameters with a Rational Function

See Also

More About
• “Use S-parameters with Port Reduction” on page 6-49
• “Compute Differential TDR and TDT” on page 6-60
• “Build Simulink Model from Rational Function” on page 6-63

6-59
6 RF Toolbox Examples

Compute Differential TDR and TDT

This example shows how to use RF Toolbox™ functions to calculate the time-domain reflectometry
(TDR) and time-domain transmission (TDT) of a differential high-speed backplane channel.

You can use TDR and TDT waveforms to characterize high-speed channels, transmission lines, and
other electrical systems.

Read Single-Ended 4-Port S-Parameters and Convert Them to Differential 2-Port S-

Parameters

Use the sparameters object to extract S-parameters from the default.s4p Touchstone file. The
parameters in this data file are the 50-ohm S-parameters of a single-ended 4-port passive circuit,
measured at 1496 frequencies ranging from 50 MHz to 15 GHz. Next, extract the single-ended four-
port S-parameters from the data object.

filename = 'default.s4p';
backplane = sparameters(filename);
data = backplane.Parameters;
freq = backplane.Frequencies;
z0 = backplane.Impedance;

Use the matrix conversion function s2sdd to convert the 4-port S-parameters to 2-port differential S-
parameters.

diffdata = s2sdd(data);
diffsparams = sparameters(diffdata,freq,2*z0);
z0differential = diffsparams.Impedance;

Calculate and Plot the Differential TDR

TDR is the time-domain reflected signal from any impedance discontinuities within a circuit. First,
extract the differential S11 (reflected signal in the frequency domain) data using the rfparam
function. Convert the S11 data to a time-domain signal by creating a rational function of the data
using the rational function. Compute the TDR Signal using the stepresp function of the
rfmodel.rational object.

s11 = rfparam(diffsparams,1,1);
s11fit = rational(freq,s11);
Ts = 5e-12;
N = 5000;
Trise = 5e-11;
[tdr,tdrT] = stepresp(s11fit,Ts,N,Trise);

Calculate the impedance over time using the gamma2z function and plot the calculated TDR
impedance versus time.

zLt = gamma2z(tdr, z0differential);

figure
plot(tdrT*1e9,zLt,'r','LineWidth',2)
ylabel('Differential TDR (Ω)')
xlabel('Time (ns)')
legend('Calculated TDR')

6-60
 Compute Differential TDR and TDT

Calculate and Plot Differential TDT

TDT is the time-domain transmitted voltage signal from an input in an electrical circuit. Use the
rational function perform rational fitting on the TDT frequency data, then use the stepresp
function to compute the TDT signal. Calculate the final TDT voltage signal by using the input voltage
(S-parameters are normalized to 1 V by the equipment). Plot the calculated TDT.

s21 = rfparam(diffsparams,2,1);
s21fit= rational(freq,s21);
Ts = 5e-12;
N = 5000;
Trise = 5e-11;
[tdt,tdtT] = stepresp(s21fit, Ts, N, Trise);
Vin = 1;
tdtV = Vin*tdt/2;
figure
plot(tdtT*1e9,tdtV,'r','LineWidth',2)
ylabel('Differential TDT (V)')
xlabel('Time (ns)')
legend('Calculated TDT')

6-61
6 RF Toolbox Examples

References

[1] Zeng. R, J. Sinsky. "Modified Rational Function Modeling Technique for High Speed Circuits."
MTT-S Int. Microwave Symp. Dig,, San Fancisco, CA, June 11-16, 2006.

[2] A. S. Ali, R. Mittra. "Time-Domain Reflectometry using Scattering Parameters and a De-
Embedding Application" Technical Report, Electromagnetic Communication Laboratory Report No.
86-4, May 1986.

See Also

More About
• “Use S-parameters with Port Reduction” on page 6-49
• “Fit S-Parameters with a Rational Function” on page 6-53
• “Build Simulink Model from Rational Function” on page 6-63

6-62
 Build Simulink Model from Rational Function

Build Simulink Model from Rational Function

This example shows how to use Simulink® to simulate a differential high-speed backplane channel.
The example first reads a Touchstone® data file that contains single-ended 4-port S-parameters for a
differential high-speed backplane and converts them to 2-port differential S-parameters. It computes
the transfer function of the differential circuit and uses the rationalfit function to fit a closed-
form rational function to the circuit's transfer function. Then, the example converts the poles and
residues of the rational function object into the numerators and denominators of the Laplace
Transform S-Domain transfer functions that it uses to build the Simulink model of the rational
function object.

To run this example, you must have Simulink installed.

Read Single-Ended 4-Port S-Parameters and Convert Them to Differential 2-Port S-

Parameters

Read a Touchstone data file, default.s4p, into an sparameters object. The parameters in this data
file are the 50-ohm S-parameters of a single-ended 4-port passive circuit, measured at 1496
frequencies ranging from 50 MHz to 15 GHz. Then, get the single-ended 4-port S-parameters from
the data object, and use the matrix conversion function s2sdd to convert them to differential 2-port
S-parameters.
filename = 'default.s4p';
backplane = sparameters(filename);
data = backplane.Parameters;
freq = backplane.Frequencies;
z0 = backplane.Impedance;

Convert to 2-port differential S-parameters. This operation pairs together odd-numbered ports first,
followed by the even-numbered ports. If a different configuration has been used to measure the single
ended S-parameters, you can specify a different second argument in the s2sdd command. For
example, option "2" will allow you to pair the input and output ports in ascending order. Alternatively,
you can use the command snp2smp to change the port order.
diffdata = s2sdd(data,1);
diffz0 = 2*z0;

Compute Transfer Function and Its Rational Function Representation

First, use the s2tf function to compute the differential transfer function. Then, use the
rationalfit function to compute the closed form of the transfer function and store it in an
rfmodel.rational object. The rationalfit function fits a rational function object to the specified
data over the specified frequencies.
difftf = s2tf(diffdata,diffz0,diffz0,diffz0);
fittol = -30; % Rational fitting tolerance in dB
delayfactor = 0.9; % Delay factor
rationalfunc = rationalfit(freq,difftf,fittol,'DelayFactor', delayfactor)
npoles = length(rationalfunc.A);
fprintf('The derived rational function contains %d poles.\n', npoles);

rationalfunc =

rfmodel.rational with properties:

6-63
6 RF Toolbox Examples

A: [20x1 double]
C: [20x1 double]
D: 0
Delay: 6.0172e-09
Name: 'Rational Function'

The derived rational function contains 20 poles.

Get Numerator and Denominator of Laplace Transform S-Domain Transfer Functions

This example uses Laplace Transform S-Domain transfer functions to represent the backplane in the
Simulink model. Convert the poles and corresponding residues of the rational function object into
numerator and denominator form for use in the Laplace Transform transfer function blocks. Each
transfer function block represents either one real pole and the corresponding real residue, or a pair
of complex conjugate poles and residues, so the transfer function block always has real coefficients.
For this example, the rational function object contains 2 real poles/residues and 6 pairs of complex
poles/residues, so the Simulink model contains 8 transfer function blocks.

A = rationalfunc.A;
C = rationalfunc.C;
den = cell(size(A));
num = cell(size(A));
k = 1; % Index of poles and residues
n = 0; % Index of numerators and denominators
while k <= npoles
if isreal(A(k)) % Real poles
n = n + 1;
num{n} = C(k);
den{n} = [1, -A(k)];
k = k + 1;
else % Complex poles
n = n + 1;
real_a = real(A(k));
imag_a = imag(A(k));
real_c = real(C(k));
imag_c = imag(C(k));
num{n} = [2*real_c, -2*(real_a*real_c+imag_a*imag_c)];
den{n} = [1, -2*real_a, real_a^2+imag_a^2];
k = k + 2;
end
end
den = den(1:n);
num = num(1:n);

Build Simulink Model of Backplane

Build a Simulink model of the backplane using the Laplace Transform transfer functions. Then,
connect a random source to the input of the backplane and a scope to its input and output.

modelname = fliplr(strtok(fliplr(tempname), filesep));

simulink_rfmodel_build_rational_system_helper(modelname , numel(num))
simulink_rfmodel_add_source_sink_helper(modelname)

6-64
 Build Simulink Model from Rational Function

Figure 1. Simulink model for a rational function

Simulate Simulink Model of Rational Function

When you simulate the model, the Scope shows the impact of the differential backplane on the
random input signal.

set_param([modelname,'/Rational Model Output'], 'Open', 'on')

sc = get_param([modelname,'/Rational Model Output'],'ScopeConfiguration');
sc.Position = [200, 216, 901, 442];
sim(modelname);

6-65
6 RF Toolbox Examples

Close Model

close_system(modelname, 0)

See Also

More About
• “Export Verilog-A module from Rational Function” on page 6-67
• “Use S-parameters with Port Reduction” on page 6-49

6-66
 Export Verilog-A module from Rational Function

Export Verilog-A module from Rational Function

This example shows how to use RF Toolbox™ functions to generate a Verilog-A module that models
the high-level behavior of a high-speed backplane. First, it reads the single-ended 4-port S-
parameters for a differential high-speed backplane and converts them to 2-port differential S-
parameters. Then, it computes the transfer function of the differential circuit and fits a rational
function to the transfer function. Next, the example exports a Verilog-A module that describes the
model. Finally, it plots the unit step response of the generated Verilog-A module in a third-party
circuit simulation tool.

Use Rational Function Object to Describe High-Level Behavior of High-Speed Backplane

Read a Touchstone® data file, default.s4p, into an sparameters object. The parameters in this
data file are the 50-ohm S-parameters of a single-ended 4-port passive circuit, measured at 1496
frequencies ranging from 50 MHz to 15 GHz. Then, extract the single-ended 4-port S-parameters
from the data stored in the Parameters property of the sparameters object, use the s2sdd
function to convert them to differential 2-port S-parameters, and use the s2tf function to compute
the transfer function of the differential circuit. Then, use the rationalfit function to generate an
rfmodel.rational object that describes the high-level behavior of this high-speed backplane. The
rfmodel.rational object is a rational function object that expresses the circuit's transfer function
in closed form using poles, residues, and other parameters, as described in the rationalfit
reference page.

filename = 'default.s4p';
backplane = sparameters(filename);
data = backplane.Parameters;
freq = backplane.Frequencies;
z0 = backplane.Impedance;

Convert to 2-port differential S-parameters.

diffdata = s2sdd(data);
diffz0 = 2*z0;
difftf = s2tf(diffdata,diffz0,diffz0,diffz0);

Fit the differential transfer function into a rational function.

fittol = -30; % Rational fitting tolerance in dB

delayfactor = 0.9; % Delay factor
rationalfunc = rationalfit(freq,difftf,fittol,'DelayFactor',delayfactor)

rationalfunc =
rfmodel.rational with properties:

A: [20x1 double]
C: [20x1 double]
D: 0
Delay: 6.0172e-09
Name: 'Rational Function'

Export Rational Function Object as Verilog-A Module

Use the writeva method of the rfmodel.rational object to export the rational function object as
a Verilog-A module, called samplepassive1, that describes the rational model. The input and output

6-67
6 RF Toolbox Examples

nets of samplepassive1 are called line_in and line_out. The predefined Verilog-A discipline,
electrical, describes the attributes of these nets. The format of numeric values, such as the
Laplace transform numerator and denominator coefficients, is %12.10e. The electrical discipline is
defined in the file disciplines.vams, which is included in the beginning of the
samplepassive1.va file.
workingdir = tempname;
mkdir(workingdir)
writeva(rationalfunc, fullfile(workingdir,'samplepassive1'), ...
'line_in', 'line_out', 'electrical', '%12.10e', 'disciplines.vams');

type(fullfile(workingdir,'samplepassive1.va'));

// Module: samplepassive1

// Generated by MATLAB(R) 24.2 and the RF Toolbox 24.2.

// Generated on: 06-Sep-2024 15:23:42

`include "disciplines.vams"

module samplepassive1(line_in, line_out);

electrical line_in, line_out;
electrical node1;

real nn1[0:1], nn2[0:1], nn3[0:1], nn4[0:1], nn5[0:1], nn6[0:1], nn7[0:1], nn8[0:1], nn9[0:1], n
real dd1[0:2], dd2[0:2], dd3[0:2], dd4[0:2], dd5[0:2], dd6[0:2], dd7[0:2], dd8[0:2], dd9[0:2], d

analog begin

@(initial_step) begin
nn1[0] = -3.8392614832e+18;
nn1[1] = 5.2046393014e+07;
dd1[0] = 2.8312609831e+21;
dd1[1] = 3.5124823781e+09;
dd1[2] = 1.0000000000e+00;
nn2[0] = -2.0838483814e+19;
nn2[1] = 5.3487174018e+08;
dd2[0] = 1.8020362314e+21;
dd2[1] = 7.8266367089e+09;
dd2[2] = 1.0000000000e+00;
nn3[0] = 1.7726270794e+19;
nn3[1] = 2.5185716022e+09;
dd3[0] = 1.2157471895e+21;
dd3[1] = 8.1132784895e+09;
dd3[2] = 1.0000000000e+00;
nn4[0] = 2.3112282793e+20;
nn4[1] = 9.2690544437e+08;
dd4[0] = 7.9582429152e+20;
dd4[1] = 1.1379108659e+10;
dd4[2] = 1.0000000000e+00;
nn5[0] = 8.9321469721e+19;
nn5[1] = -1.4945928109e+10;
dd5[0] = 4.1473706594e+20;
dd5[1] = 1.1346735824e+10;
dd5[2] = 1.0000000000e+00;
nn6[0] = -3.5180951909e+20;
nn6[1] = -1.9895507212e+10;

6-68
 Export Verilog-A module from Rational Function

dd6[0] = 1.9080843811e+20;
dd6[1] = 1.0434555792e+10;
dd6[2] = 1.0000000000e+00;
nn7[0] = -1.0593240107e+20;
nn7[1] = 1.9248932577e+10;
dd7[0] = 6.1152960549e+19;
dd7[1] = 1.0001203231e+10;
dd7[2] = 1.0000000000e+00;
nn8[0] = 5.4441539403e+16;
nn8[1] = -9.7818749687e+06;
dd8[0] = 4.3821946493e+19;
dd8[1] = 6.6700188623e+08;
dd8[2] = 1.0000000000e+00;
nn9[0] = 2.2556903052e+16;
nn9[1] = 7.9711163023e+06;
dd9[0] = 2.1228807651e+19;
dd9[1] = 4.9531801417e+08;
dd9[2] = 1.0000000000e+00;
nn10[0] = 1.1592988960e+10;
dd10[0] = 3.0829914556e+09;
dd10[1] = 1.0000000000e+00;
nn11[0] = 1.2852839051e+08;
dd11[0] = 5.9779845807e+08;
dd11[1] = 1.0000000000e+00;
end

V(node1) <+ laplace_nd(V(line_in), nn1, dd1);

V(node1) <+ laplace_nd(V(line_in), nn2, dd2);
V(node1) <+ laplace_nd(V(line_in), nn3, dd3);
V(node1) <+ laplace_nd(V(line_in), nn4, dd4);
V(node1) <+ laplace_nd(V(line_in), nn5, dd5);
V(node1) <+ laplace_nd(V(line_in), nn6, dd6);
V(node1) <+ laplace_nd(V(line_in), nn7, dd7);
V(node1) <+ laplace_nd(V(line_in), nn8, dd8);
V(node1) <+ laplace_nd(V(line_in), nn9, dd9);
V(node1) <+ laplace_nd(V(line_in), nn10, dd10);
V(node1) <+ laplace_nd(V(line_in), nn11, dd11);
V(line_out) <+ absdelay(V(node1), 6.0171901584e-09);
end
endmodule

Plot Unit Step Response of Generated Verilog-A Module

Many third-party circuit simulation tools support the Verilog-A standard. These tools simulate
standalone components defined by Verilog-A modules and circuits that contain these components. The
following figure shows the unit step response of the samplepassive1 module. The figure was
generated with a third-party circuit simulation tool.

6-69
6 RF Toolbox Examples

Figure 1: The unit step response.

delete(fullfile(workingdir,'samplepassive1.va'));
rmdir(workingdir)

See Also

More About
• “Build Simulink Model from Rational Function” on page 6-63
• “Use S-parameters with Port Reduction” on page 6-49

6-70
 Using 'NPoles' Parameter With rationalfit

Using 'NPoles' Parameter With rationalfit

This example shows how to use the 'NPoles' parameter to improve the quality of the output of
rationalfit. By default, the rationalfit function uses 48 or fewer poles to find the rational
function that best matches the data. If 48 poles is not enough, it may be advantageous to change the
range of the number of poles used by rationalfit.

First, read in the bandpass filter data contained in the file npoles_bandpass_example.s2p, and
plot the S21 data. Next, use the rationalfit function to fit a rational function to the S21 data, with
the 'NPoles' parameter set to its default value, and visually compare the results to the original data.
Lastly, use rationalfit again, this time specifying a larger number of poles, and see if the result
improves.

Read and Visualize Data

S = sparameters('npoles_bandpass_example.s2p');
figure
subplot(2,1,1)
rfplot(S,2,1,'db')
subplot(2,1,2)
rfplot(S,2,1,'angle')

Analyze Output of rationalfit When Using Default Value for 'NPoles'

Use the rfparam function to extract the S21 values, and then call rationalfit.

6-71
6 RF Toolbox Examples

s21 = rfparam(S,2,1);
datafreq = S.Frequencies;
defaultfit = rationalfit(datafreq,s21);

Warning: Achieved only -13.0 dB accuracy with 48 poles, not -40.0 dB. Consider specifying a larg

Use the freqresp function to calculate the response of the output of rationalfit.

respfreq = 2.25e9:2e5:2.75e9;
defaultresp = freqresp(defaultfit,respfreq);

Compare the original data against the frequency response of the default rational function calculated
by rationalfit.

subplot(2,1,1)
plot(datafreq,20*log10(abs(s21)),'.-')
hold on
plot(respfreq,20*log10(abs(defaultresp)))
hold off
xlabel('Frequency (Hz)')
ylabel('Magnitude (dB)')
defaultnpoles = numel(defaultfit.A);
defaultstr = ['Default NPoles (Uses ',num2str(defaultnpoles),' poles)'];
title(defaultstr)
legend('Original Data','Default rationalfit','Location','best')
subplot(2,1,2)
plot(datafreq,unwrap(angle(s21))*180/pi,'.-')
hold on
plot(respfreq,unwrap(angle(defaultresp))*180/pi)
hold off
xlabel('Frequency (Hz)')
ylabel('Angle (degrees)')
legend('Original Data','Default rationalfit','Location','best')

6-72
 Using 'NPoles' Parameter With rationalfit

Analyzing how well the output of rationalfit matches the original data, it appears that while the
default values of rationalfit do a reasonably good job in the center of the bandpass region, the fit
is poor on the edges of the bandpass region. It is possible that using a more complex rational function
will achieve a better fit.

Analyze Output of rationalfit When Using Custom Value for 'NPoles'

Fit the original S21 data, but this time, instruct rationalfit to use between 49 and 60 poles using
the 'NPoles' parameter.

customfit = rationalfit(datafreq,s21,'NPoles',[49 60]);

customresp = freqresp(customfit,respfreq);

Compare the original data against the frequency response of the custom rational function calculated
by rationalfit.

figure
subplot(2,1,1)
plot(datafreq,20*log10(abs(s21)),'.-')
hold on
plot(respfreq,20*log10(abs(customresp)))
hold off
xlabel('Frequency (Hz)')
ylabel('Magnitude (dB)')
customnpoles = numel(customfit.A);
customstr = ['NPoles = [49 60] (Uses ',num2str(customnpoles),' poles)'];
title(customstr)

6-73
6 RF Toolbox Examples

legend('Original Data','Custom rationalfit','Location','best')

subplot(2,1,2)
plot(datafreq,unwrap(angle(s21))*180/pi,'.-')
hold on
plot(respfreq,unwrap(angle(customresp))*180/pi)
hold off
xlabel('Frequency (Hz)')
ylabel('Angle (degrees)')
legend('Original Data','Custom rationalfit','Location','best')

The fit using a larger number of poles is clearly more precise.

See Also

More About
• “Using 'Weight' Parameter With rationalfit” on page 6-75
• “Improve Quality of Fit Using 'DelayFactor' Parameter” on page 6-81
• “Rational Fit S-Parameters” on page 6-214

6-74
 Using 'Weight' Parameter With rationalfit

Using 'Weight' Parameter With rationalfit

This example shows how to use the 'Weight' parameter to improve the quality of the output of
rationalfit.

By default, the rationalfit function minimizes the absolute error between the data and the
rational function, treating all data points equally. If you want to emphasize some of the data points
more than the others, use the 'Weight' parameter.

For example, If the magnitude of the input data has a large dynamic range, it is often useful to be
more concerned with the relative error at each data point, rather than the absolute error at each data
point, so that the data points with relatively smaller magnitudes are fit accurately. The common way
to do this is to set the 'Weight' parameter to 1./abs(data).

To put the example above in practice, follow the steps below.

• Read in the SAW filter data contained in the file sawfilter.s2p, and plot the S21 data.
• Use the rationalfit function to fit a rational function to the S21 data, with the 'Weight'
parameter set to its default value, and visually compare the results to the original data.
• Use rationalfit again, this time specifying the 'Weight' parameter to be 1./abs(S21), and
see if the result improves.

Read and Visualize Data

Read and visualize SAW filter S-parameters data.

S = sparameters('sawfilter.s2p');
figure
subplot(2,1,1)
rfplot(S,2,1,'db')
subplot(2,1,2)
rfplot(S,2,1,'angle')

6-75
6 RF Toolbox Examples

Analyze Output of rationalfit When Using Default Value for 'Weight'

Use the rfparam function to extract the S21 values, and then call rationalfit.
s21 = rfparam(S,2,1);
datafreq = S.Frequencies;
defaultfit = rationalfit(datafreq,s21);

Use the freqresp function to calculate the response of the output of rationalfit.
respfreq = 1e9:1.5e6:4e9;
defaultresp = freqresp(defaultfit,respfreq);

Compare the original data against the frequency response of the default rational function calculated
by rationalfit.
subplot(2,1,1)
plot(datafreq,20*log10(abs(s21)),'.-')
hold on
plot(respfreq,20*log10(abs(defaultresp)))
hold off
xlabel('Frequency (Hz)')
ylabel('Magnitude (dB)')
defaultnpoles = numel(defaultfit.A);
defaultstr = ['Default Weight (Uses ',num2str(defaultnpoles),' poles)'];
title(defaultstr)
legend('Original Data','Default rationalfit','Location','best')
subplot(2,1,2)

6-76
 Using 'Weight' Parameter With rationalfit

plot(datafreq,unwrap(angle(s21))*180/pi,'.-')
hold on
plot(respfreq,unwrap(angle(defaultresp))*180/pi)
hold off
xlabel('Frequency (Hz)')
ylabel('Angle (degrees)')
legend('Original Data','Default rationalfit','Location','best')

While the output of rationalfit is not awful, it does not match the regions in the data that are very
small in magnitude.

figure
plot(datafreq,20*log10(abs(s21)),'.-')
hold on
plot(respfreq,20*log10(abs(defaultresp)))
hold off
axis([2.25e9 2.65e9 -75 -30])
xlabel('Frequency (Hz)')
ylabel('Magnitude (dB)')
title('Accuracy at Small Magnitudes Using Default Weight')
legend('Original Data','Default rationalfit','Location','best')

6-77
6 RF Toolbox Examples

Using the 'Weight' parameter to make that data relatively more important can help the accuracy of
the fit.

Analyze Output of rationalfit When Using Custom Value for 'Weight'

By using a 'Weight' of 1./abs(s21), rationalfit minimizes the relative error of the system,
instead of the absolute error of the system.
customfit = rationalfit(datafreq,s21,'Weight',1./abs(s21));

Warning: Achieved only -39.7 dB accuracy with 48 poles, not -40.0 dB. Consider specifying a larg

customresp = freqresp(customfit,respfreq);

Compare the original data against the frequency response of the custom rational function calculated
by rationalfit.
figure
subplot(2,1,1)
plot(datafreq,20*log10(abs(s21)),'.-')
hold on
plot(respfreq,20*log10(abs(customresp)))
hold off
xlabel('Frequency (Hz)')
ylabel('Magnitude (dB)')
customnpoles = numel(customfit.A);
customstr = ['Weight = 1./abs(s21) (Uses ',num2str(customnpoles),' poles)'];
title(customstr)

6-78
 Using 'Weight' Parameter With rationalfit

legend('Original Data','Custom rationalfit','Location','best')

subplot(2,1,2)
plot(datafreq,unwrap(angle(s21))*180/pi,'.-')
hold on
plot(respfreq,unwrap(angle(customresp))*180/pi)
hold off
xlabel('Frequency (Hz)')
ylabel('Angle (degrees)')
legend('Original Data','Custom rationalfit','Location','best')

The plot shows that the custom 'Weight' parameter created a better fit for the data points with
smaller magnitudes.

figure
plot(datafreq,20*log10(abs(s21)),'.-')
hold on
plot(respfreq,20*log10(abs(customresp)))
hold off
axis([2.25e9 2.65e9 -75 -30])
xlabel('Frequency (Hz)')
ylabel('Magnitude (dB)')
title('Accuracy at Small Magnitudes Using Custom Weight')
legend('Original Data','Custom rationalfit','Location','best')

6-79
6 RF Toolbox Examples

See Also

More About
• “Improve Quality of Fit Using 'DelayFactor' Parameter” on page 6-81
• “Rational Fit S-Parameters” on page 6-214
• “Rational Fit S-Parameters” on page 6-214

6-80
 Improve Quality of Fit Using 'DelayFactor' Parameter

Improve Quality of Fit Using 'DelayFactor' Parameter

This example shows how to use the 'DelayFactor' parameter to improve the quality of the output of
rationalfit function.

The rationalfit function selects a rational function that matches frequency domain data. If that
data contains a significant "time delay", which would present itself as a phase shift in the frequency
domain, then it might be very difficult to fit using a reasonable number of poles.

In these cases, when the input data contains a large negative slope (i.e. data with a large enough time
delay), the rationalfit function first remove some of the delay from the data, and then find a
rational function that best fits the remaining "undelayed" data. The rationalfit function accounts
for the removed delay by storing it within the 'Delay' parameter of the output. By default,
rationalfit does not remove any delay from the data.

First, create differential transfer function data from 4-port backplane S-parameters. Next, attempt to
fit the data using the default settings of the rationalfit function. Lastly, use the 'DelayFactor'
parameter to improve the accuracy of the output of rationalfit.

Create Transfer Function

Read in the 4-port backplane S-parameter data from 'default.s4p'.

S = sparameters('default.s4p');
fourportdata = S.Parameters;
freq = S.Frequencies;
fourportZ0 = S.Impedance;

Convert 4-port single ended S-parameters into 2-port differential S-parameters

diffdata = s2sdd(fourportdata);
diffZ0 = 2*fourportZ0;

Create a transfer function from the differential 2-port data

tfdata = s2tf(diffdata,diffZ0,diffZ0,diffZ0);

Analyze Output of rationalfit When Using Default Value for 'DelayFactor'

Use the freqresp function to calculate the response of the output of rationalfit.
defaultfit = rationalfit(freq,tfdata)

Warning: Achieved only -10.2 dB accuracy with 48 poles, not -40.0 dB. Consider specifying a larg

defaultfit =
rfmodel.rational with properties:

A: [48x1 double]
C: [48x1 double]
D: 0
Delay: 0
Name: 'Rational Function'

respfreq = 0:4e6:20e9;
defaultresp = freqresp(defaultfit,respfreq);

6-81
6 RF Toolbox Examples

Note that the 'Delay' parameter is zero (no delay removed from the data).

Plot the original data vs. the default output of rationalfit.

figure
subplot(2,1,1)
tfdataDB = 20*log10(abs(tfdata));
plot(freq,tfdataDB,'.-')
hold on
plot(respfreq,20*log10(abs(defaultresp)))
hold off
xlabel('Frequency (Hz)')
ylabel('Magnitude (dB)')
defaultnpoles = numel(defaultfit.A);
defstr = ['Default DelayFactor (Uses ',num2str(defaultnpoles),' poles)'];
title(defstr)
legend('Original Data','Default rationalfit','Location','best')
subplot(2,1,2)
tfdataphase = 180*unwrap(angle(tfdata))/pi;
plot(freq,tfdataphase,'.-')
hold on
plot(respfreq,180*unwrap(angle(defaultresp))/pi)
hold off
xlabel('Frequency (Hz)')
ylabel('Angle (degrees)')
legend('Original Data','Default rationalfit','Location','best')

6-82
 Improve Quality of Fit Using 'DelayFactor' Parameter

Note that the results when using the default settings of rationalfit are poor. Because the phase of
the original data has a very large negative slope, it may be possible to improve the accuracy of the
rational function by using the 'DelayFactor' parameter.

Analyze Output of rationalfit When Using Custom Value for 'DelayFactor'

'DelayFactor' must be set to a value between 0 and 1. Choosing which value is an exercise in trial and
error. For some data sets (those whose phase has an overall upward slope), changing the value of
'DelayFactor' will have no effect on the outcome.

Holding all other possible parameters of rationalfit constant, 0.98 is found to create a good fit.

customfit = rationalfit(freq,tfdata,'DelayFactor',0.98)

customfit =
rfmodel.rational with properties:

A: [31x1 double]
C: [31x1 double]
D: 0
Delay: 6.5521e-09
Name: 'Rational Function'

customresp = freqresp(customfit,respfreq);

Note that the 'Delay' parameter is not zero (rationalfit removed some delay from the data).

Plot the original data vs. the custom output of rationalfit.

subplot(2,1,1)
plot(freq,tfdataDB,'.-')
hold on
plot(respfreq,20*log10(abs(customresp)))
hold off
xlabel('Frequency (Hz)')
ylabel('Magnitude (dB)')
customnpoles = numel(customfit.A);
customstr = ['DelayFactor = 0.98 (Uses ',num2str(customnpoles),' poles)'];
title(customstr)
legend('Original Data','Custom rationalfit','Location','best')
subplot(2,1,2)
plot(freq,tfdataphase,'.-')
hold on
plot(respfreq,180*unwrap(angle(customresp))/pi)
hold off
xlabel('Frequency (Hz)')
ylabel('Angle (degrees)')
legend('Original Data','Custom rationalfit','Location','best')

6-83
6 RF Toolbox Examples

The rational function created by using a custom value for 'DelayFactor' is much more accurate, and
uses fewer poles.

See Also

More About
• “Rational Fit S-Parameters” on page 6-214
• “Using 'NPoles' Parameter With rationalfit” on page 6-71
• “Using 'Weight' Parameter With rationalfit” on page 6-75

6-84
 Data Analysis on S-Parameters of RF Data Files

Data Analysis on S-Parameters of RF Data Files

This example shows how to perform statistical analysis on a set of S-parameter data files using
magnitude, mean, and standard deviation (STD).

First, read twelve S-parameter files, where these files represent the twelve similar RF filters into the
MATLAB® workspace and plot them. Next, plot and analyze the passband response of these filters to
ensure they meet statistical norms.

Read S-Parameters from Filter Data Files

Use built-in RF Toolbox™ functions for reading a set of S-Parameter data files. For each filter plot the
S21 dB values. The names of the files are AWS_Filter_1.s2p through AWS_Filter_12.s2p. These files
represent 12 passband filters with similar specifications.

numfiles = 12;
filename = "AWS_Filter_"+(1:numfiles)+".s2p"; % Construct filenames
S = sparameters(filename(1)); % Read file #1 for initial set-up
freq = S.Frequencies; % Frequency values are the same for all files
numfreq = numel(freq); % Number of frequency points
s21_data = zeros(numfreq,numfiles); % Preallocate for speed
s21_groupdelay = zeros(numfreq,numfiles); % Preallocate for speed

% Read Touchstone files

for n = 1:numfiles
S = sparameters(filename(n));
s21 = rfparam(S,2,1);
s21_data(:,n) = s21;
s21_groupdelay(:,n) = groupdelay(S,freq,2,1);
end
s21_db = 20*log10(abs(s21_data));

figure
plot(freq/1e9,s21_db)
xlabel('Frequency (GHz)')
ylabel('Filter Response (dB)')
title('Transmission performance of 12 filters')
axis on
grid on

6-85
6 RF Toolbox Examples

Filter Passband Visualization

In this section, find, store, and plot the S21 data from the AWS downlink band (2.11 - 2.17 GHz).

idx = (freq >= 2.11e9) & (freq <= 2.17e9);

s21_pass_data = s21_data(idx,:);
s21_pass_db = s21_db(idx,:);
freq_pass_ghz = freq(idx)/1e9; % Normalize to GHz

plot(freq_pass_ghz,s21_pass_db)
xlabel('Frequency (GHz)')
ylabel('Filter Response (dB)')
title('Passband variation of 12 filters')
axis([min(freq_pass_ghz) max(freq_pass_ghz) -1 0])
grid on

6-86
 Data Analysis on S-Parameters of RF Data Files

Basic Statistical Analysis of S21 Data

To determine whether the data follows a normal distribution and if there is an outlier, perform
statistical analysis on the magnitude and group delay of all passband S21 data sets.

abs_S21_pass_freq = abs(s21_pass_data);

Calculate the mean and the STD of the magnitude of the entire passband S21 data set.

mean_abs_S21 = mean(abs_S21_pass_freq,'all')

mean_abs_S21 =
0.9289

std_abs_S21 = std(abs_S21_pass_freq(:))

std_abs_S21 =
0.0104

Calculate the mean and STD of the passband magnitude response at each frequency point. This
determines if the data follows a normal distribution.

mean_abs_S21_freq = mean(abs_S21_pass_freq,2);
std_abs_S21_freq = std(abs_S21_pass_freq,0,2);

Plot all the raw passband magnitude data as a function of frequency, as well as the upper and lower
limits defined by the basic statistical analysis.

6-87
6 RF Toolbox Examples

plot(freq_pass_ghz,mean_abs_S21_freq,'m')
hold on
plot(freq_pass_ghz,mean_abs_S21_freq + 2*std_abs_S21_freq,'r')
plot(freq_pass_ghz,mean_abs_S21_freq - 2*std_abs_S21_freq,'k')
legend('Mean','Mean + 2*STD','Mean - 2*STD')
plot(freq_pass_ghz,abs_S21_pass_freq,'c','HandleVisibility','off')
grid on
axis([min(freq_pass_ghz) max(freq_pass_ghz) 0.9 1])
ylabel('Magnitude S21')
xlabel('Frequency (GHz)')
title('S21 (Magnitude) - Statistical Analysis')
hold off

Plot a histogram for the passband magnitude data. This determines if the upper and lower limits of
the data follow a normal distribution.

histfit(abs_S21_pass_freq(:))
grid on
axis([0.8 1 0 100])
xlabel('Magnitude S21')
ylabel('Distribution')
title('Compare filter passband response vs. a normal distribution')

6-88
 Data Analysis on S-Parameters of RF Data Files

Get the groupdelay of the passband S21 data. Use inner 60% of the bandwith for statistical analysis
of the groupdelay and normalize it to 10 ns.

idx_gpd = (freq >= 2.13e9) & (freq <= 2.15e9);

freq_pass_ghz_gpd = freq(idx_gpd)/1e9; % Normalize to GHz
s21_groupdelay_pass_data = s21_groupdelay(idx_gpd,:)/10e-9; % Normalize to 10 ns

Calculate the per-frequency mean and standard deviation of the normalized group delay response. All
the data is collected into a single vector for alter analysis.

mean_grpdelay_S21 = mean(s21_groupdelay_pass_data,2);
std_grpdelay_S21 = std(s21_groupdelay_pass_data,0,2);
all_grpdelay_data = reshape(s21_groupdelay_pass_data.',numel(s21_groupdelay_pass_data),1);

Plot all the normalized passband groupdelay data as a function of frequency, including the upper and
lower limits defined by the basic statistical analysis.

plot(freq_pass_ghz_gpd,mean_grpdelay_S21,'m')
hold on
plot(freq_pass_ghz_gpd,mean_grpdelay_S21 + 2*std_grpdelay_S21,'r')
plot(freq_pass_ghz_gpd,mean_grpdelay_S21 - 2*std_grpdelay_S21,'k')
legend('Mean','Mean + 2*STD','Mean - 2*STD')
plot(freq_pass_ghz_gpd,s21_groupdelay_pass_data,'c','HandleVisibility','off')
grid on
xlim([min(freq_pass_ghz_gpd) max(freq_pass_ghz_gpd)])
ylabel('Normalized group delay S21')
xlabel('Frequency (GHz)')

6-89
6 RF Toolbox Examples

title('S21 (Normalized group delay) - Statistical Analysis')

hold off

Plot a histogram for the normalized passband group delay data. This determines if the upper and
lower limits of the data follow a uniform distribution.

histogram(all_grpdelay_data,35)
grid on
xlabel('Group delay S21 (seconds)')
ylabel('Distribution')
title('Histogram of the normalized group delay')

6-90
 Data Analysis on S-Parameters of RF Data Files

See Also

More About
• “Bandpass Filter Response Using RFCKT Objects” on page 6-33
• “Bandpass Filter Response” on page 6-21

6-91
6 RF Toolbox Examples

Write S2P Touchstone Files

This example shows how to write the data in a circuit object created in the MATLAB® workspace
into an industry-standard data file, Touchstone®. You can use these files in third-party tools.

To write a touchstone file, in this example an RLGC transmission line object is created and analyzed
in the frequency domain. This analyzed results are written into a Touchstone file and the data is
compared with the original result.

Create RF Circuit Object to Represent an RLCG Transmission Line

Create a txlineRLCGLine object to represent a RLCG transmission line. This example uses Name-
Value pairs to implement the parameters in the RLCG transmission line shown in figure 1 [1].

Figure 1: RLCG transmission line.

ckt1 = txlineRLCGLine('R',100,'L',80e-9,'C',200e-12,'G',1.6);

Clone Circuit Object

Use the clone function to make a copy of the transmission line object.

ckt2 = clone(ckt1)

ckt2 =
txlineRLCGLine: RLCGLine element

Name: 'RLCGLine'
Frequency: 1.0000e+09
R: 100
L: 8.0000e-08
C: 2.0000e-10
G: 1.6000
IntpType: 'Linear'
LineLength: 0.0100

6-92
 Write S2P Touchstone Files

Termination: 'NotApplicable'
StubMode: 'NotAStub'

Cascade Two Circuit Objects

Use the circuit object to cascade the two transmission lines.

ckt = circuit([ckt1,ckt2]);

Analyze and Plot S-Parameter Data

Use the sparameters object to analyze the cascadeed transmission line in the frequency domain.

freq = linspace(0,10e9);
ckt_sparameters = sparameters(ckt,freq);

Use the smithplot method to plot the object's S11 on a Smith chart®.

figure
smithplot(ckt_sparameters,[1,1],'LegendLabels','S11 Original')

Write Data to S2P File

Use the rfwrite function to write the data to a file.

workingdir = tempname;
mkdir(workingdir);

6-93
6 RF Toolbox Examples

filename = fullfile(workingdir,'myrlcg.s2p');
if exist(filename,'file')
delete(filename)
end
rfwrite(ckt_sparameters,filename);

Compare Data

Read the data from the file myrlcg.s2p into a new sparameters object and plot input reflection
coefficient, S11 on a Smith chart. Visually compare the 'S11 original' and 'S11 from S2P' to confirm
that the data matches.

compare_ckt = sparameters(filename);
figure
smithplot(compare_ckt,[1,1],'LegendLabels','S11 from S2P')

[1] M. Steer, "Transmission Lines," in Microwave and RF Design: Transmission Lines. vol. 2, 3rd ed.
Raleigh, North Carolina, US: North Carolina State University, 2019, ch. 2, sec. 2, pp.58.

See Also

More About
• “Visualize Mixer Spurs” on page 6-95
• “Finding Free IF Bandwidths” on page 6-101

6-94
 Visualize Mixer Spurs

Visualize Mixer Spurs

This example shows how to create an rfckt.mixer object and plot the mixer spurs of that object.

Mixers are non-linear devices used in RF systems. They are typically used to convert signals from one
frequency to another. In addition to the desired output frequency, mixers also produce
intermodulation products (also called mixer spurs), which are unwanted side effects of their
nonlinearity. The output of the mixer occurs at the frequencies:

Fout(N, M) = NFin + MFLO

where:

• Fin is the input frequency.

• FLO is the local oscillator (LO) frequency.
• N is a nonnegative integer.
• M is an integer.

Only one of these output frequencies is the desired tone. For example, in a downconversion mixer (i.e.
Fin = FRF) with a low-side LO (i.e. FRF > FLO), the case N = 1, M = − 1 represents the desired output
tone. That is:

Fout(1, − 1) = FIF = NFin + MFLO = FRF − FLO

All other combinations of N and M represent the spurious intermodulation products.

Intermodulation tables (IMTs) are often used in system-level modeling of mixers. This example first
examines the IMT of a mixer. Then the example reads an .s2d format file containing an IMT, and
plots the output power at each output frequency, including the desired signal and the unwanted
spurs. The example also creates a cascaded circuit which contains a mixer with IMT followed by a
filter, whose purpose is to mitigate the spurs, and plots the output power before and after mitigation.

For more information on IMTs, see the OpenIF example “Finding Free IF Bandwidths” on page 6-101.

Create Mixer Object from Data File

Create an rfckt.mixer object to represent the downconverting mixer that is specified in the file,
samplespur1.s2d. The mixer is characterized by S-parameters, spot noise and IMT. These data are
stored in the NetworkData, NoiseData and MixerSpurData properties of the rfckt object,
respectively.
Mixer = rfckt.mixer('FLO', 1.7e9); % Flo = 1.7GHz
read(Mixer,'samplespur1.s2d');
disp(Mixer)

rfckt.mixer with properties:

MixerSpurData: [1x1 rfdata.mixerspur]

MixerType: 'Downconverter'
FLO: 1.7000e+09
FreqOffset: []
PhaseNoiseLevel: []
NoiseData: [1x1 rfdata.noise]

6-95
6 RF Toolbox Examples

NonlinearData: Inf
IntpType: 'Linear'
NetworkData: [1x1 rfdata.network]
nPort: 2
AnalyzedResult: [1x1 rfdata.data]
Name: 'Mixer'

IMT = Mixer.MixerSpurData.data

IMT = 16×16

99 26 35 39 50 41 53 49 51 42 62 51 60 47 77 50
24 0 35 13 40 24 45 28 49 33 53 42 60 47 63 99
73 73 74 70 71 64 69 64 69 62 74 62 72 60 99 99
67 64 69 50 77 47 74 44 74 47 75 44 70 99 99 99
86 90 86 88 88 85 86 85 90 85 85 85 99 99 99 99
90 80 90 71 90 68 90 65 88 65 85 99 99 99 99 99
90 90 90 90 90 90 90 90 90 90 99 99 99 99 99 99
90 90 90 90 90 87 90 90 90 99 99 99 99 99 99 99
99 95 99 95 99 95 99 95 99 99 99 99 99 99 99 99
90 95 90 90 90 99 90 99 99 99 99 99 99 99 99 99
⋮

Plot Mixer Output Signal and Spurs

Use the plot method of the rfckt object to plot the power of the desired output signal and the
spurs. The second input argument must be the string 'MIXERSPUR'. The third input argument must
be the index of the circuit for which to plot output power data. The rfckt.mixer object only
contains one circuit (the mixer), so index 0 corresponds to the mixer input and index 1 corresponds to
the mixer output.

CktIndex = 1; % Plot the output only

Pin = -10; % Input power is -10dBm
Fin = 2.1e9; % Input frequency is 2.1GHz
figure
plot(Mixer,'MIXERSPUR',CktIndex,Pin,Fin);

6-96
 Visualize Mixer Spurs

Use Data Cursor

Run the cursor over the plot to get the frequency and power level of each signal and spur.

6-97
6 RF Toolbox Examples

Create Cascaded Circuit

Create an amplifier object for LNA, mixer, and LC Bandpass Tee objects. Then build the cascade
shown in the following figure:

Figure 1: Cascaded Circuit

FirstCkt = rfckt.amplifier('NetworkData', ...

rfdata.network('Type','S','Freq',2.1e9,'Data',[0,0;10,0]), ...
'NoiseData',0,'NonlinearData',Inf); % 20dB LNA
SecondCkt = copy(Mixer); % Mixer with IMT table
ThirdCkt = rfckt.lcbandpasstee('L',[97.21 3.66 97.21]*1.0e-9, ...
'C',[1.63 43.25 1.63]*1.0e-12); % LC Bandpass filter
CascadedCkt = rfckt.cascade('Ckts',{FirstCkt,SecondCkt,ThirdCkt});

6-98
 Visualize Mixer Spurs

Plot Output Signal and Spurs of LC filter in Cascade

Use the plot method of the rfckt object to plot the power of the desired output signal and the
spurs. The third input argument is 3, which directs the toolbox to plot the power at the output of the
third component of the cascade (the LC filter).

CktIndex = 3; % Plot the output signal and spurs of the LC filter,

% which is the 3rd circuit in the cascade
Pin = -30; % Input power is -30dBm
Fin = 2.1e9; % Input frequency is 2.1GHz
plot(CascadedCkt,'MIXERSPUR',CktIndex,Pin,Fin)

Plot Cascade Signal and Spurs in 3D

Use the plot method of the rfckt object with a third input argument of 'all' to plot the input
power and the output power after each circuit component in the cascade. Circuit index 0 corresponds
to the input of the cascade. Circuit index 1 corresponds to the output of the LNA. Circuit index 2
corresponds to the output of the mixer, which was shown in the previous plot. Circuit index 3
corresponds to the output of the LC Bandpass Tee filter.

CktIndex = 'all'; % Plot the input signal, the output signal, and the
% spurs of the three circuits in the cascade: FirstCkt,
% SecondCkt and ThirdCkt
Pin = -30; % Input power is -30dBm
Fin = 2.1e9; % Input frequency is 2.1GHz
plot(CascadedCkt,'MIXERSPUR',CktIndex,Pin,Fin)
view([68.5 26])

6-99
6 RF Toolbox Examples

See Also

More About
• “Perform Frequency Planning to Find Spur-Free IF Bandwidths” on page 6-288
• “Finding Free IF Bandwidths” on page 6-101
• “Write S2P Touchstone Files” on page 6-92

6-100
 Finding Free IF Bandwidths

Finding Free IF Bandwidths

This example shows how to select an Intermediate Frequency (IF) that is free from any
intermodulation distortion. First, you create an OpenIF object and specify whether you are designing
a transmitter or receiver. Second, you use the addMixer function to define the properties of each
mixer as well as the specific Radio Frequency (RF) it interacts with. Lastly, you view the results using
the functions report and show.

Background Knowledge (Mixer Spurs)

When converting from RF to IF (receiver) or from IF to RF (transmitter), a mixer is used.

Unfortunately, mixers are nonlinear and their outputs contain energy at unwanted frequencies (we
call these unwanted outputs "spurs"). The OpenIF tool helps you to select an IF which avoids having
these spurious mixer outputs interfere with the mixer output. The output of the mixer is
characterized by the following equation:

Fout(N, M) = NFin + MFLO

where:

• Fin is the input frequency.

• FLO is the local oscillator (LO) frequency.
• N is a nonnegative integer.
• M is an integer.

Only one of these output frequencies is the desired tone. For example, in a downconversion mixer (i.e.
Fin = FRF) with a low-side LO (i.e. FRF > FLO), the case N = 1, M = − 1 represents the desired output
tone. That is:

Fout(1, − 1) = FIF = NFin + MFLO = FRF − FLO

All other combinations of N and M represent the spurious intermodulation products. To characterize
these intermodulation products, an Intermodulation Table (IMT) is used.

Background Knowledge (Intermodulation Tables)

The IMT provides information on the amount of power generated at each intermodulation product
frequency. For accurate mixer spurs analysis results, the IMT should be built from simulated or
measured data at the desired input signal and local oscillator frequency and power conditions.
Extrapolation to other conditions will lead to inaccuracies.

Here is the IMT of a downconverting mixer with a low side LO, measured at Fin = FRF = 2 . 1 GHz,
Pin = PRF = − 10 dBm, FLO = 1 . 7 GHz, and PLO = 7 dBm.

! Element (N,M) gives power of |N*Fin+M*Flo| in dBc

! Top indices give M =
! Left-hand indices give N =
%0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
0% 99 26 35 39 50 41 53 49 51 42 62 51 60 47 77 50
1% 24 0 35 13 40 24 45 28 49 33 53 42 60 47 63
2% 73 73 74 70 71 64 69 64 69 62 74 62 72 60
3% 67 64 69 50 77 47 74 44 74 47 75 44 70

6-101
6 RF Toolbox Examples

4% 86 90 86 88 88 85 86 85 90 85 85 85
5% 90 80 90 71 90 68 90 65 88 65 85
6% 90 90 90 90 90 90 90 90 90 90
7% 90 90 90 90 90 87 90 90 90
8% 99 95 99 95 99 95 99 95
9% 90 95 90 90 90 99 90
10% 99 99 99 99 99 99
11% 90 99 90 95 90
12% 99 99 99 99
13% 90 99 90
14% 99 99
15% 99

Notice that it is a convention in industry-standard IMTs to assume symmetry, namely:

Pout(N, M) = Pout(N, − M)

and RF Toolbox™ software follows this convention.

If the measurement reveals that in fact the mixer is asymmetric, i.e.:

Pout(N, M) ≠ Pout(N, − M)

there is no way of accommodating this information in an industry-standard IMT. In this situation, the
most common convention is to build an approximate model by placing the value:

max Pout(N, M), Pout(N, − M)

at position N, M.

Thus industry-standard IMTs in general and RF Toolbox in particular will over-estimate the power of
one spur in each pair of asymmetric spurs.

In the IMT, a 0 always appears in the table at the position N = 1, M = 1, which represents both the
desired signal and its symmetric image pair. All other entries are specified in dBc below the power of
the mixer output at the desired frequency. (In the unlikely case of a spur being above the power of the
desired, it will appear as a negative number, the magnitude of which is the spur power in dBc above
the desired.)

For example, in the IMT above, at row N = 1, column M = 3, the IMT value is 13. RF Toolbox will
place a pair of symmetric IM products at:

Fout(1, 3) = Fin + 3FLO

Fout(1, − 3) = Fin − 3FLO

each with a power level of -13 dBc. The absolute power of a spur in dBm is calculated by subtracting
the IMT dBc value from the output power (also in dBm) of the desired tone.

By convention, the special value of 99 means the tone at that index is negligible.

For more information on intermodulation tables, see [1] on page 6-109.

Design Requirements

Find a spur-free IF for a receiver. The receiver must be able to downconvert from three separate RF
bands to the same (shared) IF. To find an IF center frequency that is spur-free for all three RF bands,

6-102
 Finding Free IF Bandwidths

your requirements must specify the RF Center Frequency, the RF Bandwidth, and the IF Bandwidth
that goes with that particular RF:

% RF band 1
RFCF1 = 2400e6; % 2.4 GHz
RFBW1 = 200e6; % 200 MHz
IFBW1 = 20e6; % 20 MHz

% RF band 2
RFCF2 = 3700e6; % 3.7 GHz
RFBW2 = 250e6; % 250 MHz
IFBW2 = 20e6; % 20 MHz

% RF band 3
RFCF3 = 5400e6; % 5.4 GHz
RFBW3 = 250e6; % 250 MHz
IFBW3 = 50e6; % 50 MHz

Next we must have an IMT measured for each RF band. Assume you have tested and measured the
mixers you plan to use with the following results:

IMT1 = [99 0 21 17 26;

11 0 29 29 63;
60 48 70 86 41;
90 89 74 68 87;
99 99 95 99 99];

IMT2 = [99 1 9 12 15;

20 0 26 31 48;
55 70 51 70 53;
85 90 60 70 94;
96 95 94 93 92];

IMT3 = [99 2 11 15 16;

27 0 16 41 55;
25 61 66 65 47;
92 83 66 77 88;
97 94 91 92 99];

Find Spur-Free frequencies using the OpenIF object

Create the object using the OpenIF function. Specify you are designing a receiver by setting the
'IFLocation' property to 'MixerOutput'.

h = OpenIF('IFLocation', 'MixerOutput');

Use the addMixer method to input the information for each RF band. Here low-side injection is
assumed for each mixer, but high-side injection could be tried later.

addMixer(h,IMT1, RFCF1, RFBW1, 'low', IFBW1);

addMixer(h,IMT2, RFCF2, RFBW2, 'low', IFBW2);
addMixer(h,IMT3, RFCF3, RFBW3, 'low', IFBW3);

View the results textually using the report method.

report(h);

6-103
6 RF Toolbox Examples

Intermediate Frequency (IF) Planner

IF Location: MixerOutput

-- MIXER 1 --
RF Center Frequency: 2.4 GHz
RF Bandwidth: 200 MHz
IF Bandwidth: 20 MHz
MixerType: low
Intermodulation Table: 99 0 21 17 26
11 0 29 29 63
60 48 70 86 41
90 89 74 68 87
99 99 95 99 99

-- MIXER 2 --
RF Center Frequency: 3.7 GHz
RF Bandwidth: 250 MHz
IF Bandwidth: 20 MHz
MixerType: low
Intermodulation Table: 99 1 9 12 15
20 0 26 31 48
55 70 51 70 53
85 90 60 70 94
96 95 94 93 92

-- MIXER 3 --
RF Center Frequency: 5.4 GHz
RF Bandwidth: 250 MHz
IF Bandwidth: 50 MHz
MixerType: low
Intermodulation Table: 99 2 11 15 16
27 0 16 41 55
25 61 66 65 47
92 83 66 77 88
97 94 91 92 99

Spur-Free Zones:
2.00 - 2.50 MHz
2.50 - 3.33 MHz
3.33 - 6.25 MHz
6.25 - 8.33 MHz
8.33 - 458.00 MHz
502.00 - 572.50 MHz
627.50 - 713.00 MHz
836.67 - 891.25 MHz
1.00 - 1.05 GHz
1.11 - 1.15 GHz
1.28 - 1.31 GHz
1.67 - 1.72 GHz
2.00 - 2.10 GHz
2.55 - 2.62 GHz
4.17 - 4.21 GHz
7.74 - 7.90 GHz
8.30 - 8.34 GHz
10.20 - 10.53 GHz
11.49 - 11.49 GHz
15.31 - 15.80 GHz
16.60 - 17.87 GHz

6-104
 Finding Free IF Bandwidths

19.35 - 21.08 GHz

22.96 - 26.35 GHz

View the results graphically using the show method.

figure;
show(h);

Interpreting the Graphical Results

The figure created by the show method displays all relevant spurious frequency ranges as colored
horizontal rectangles. If there any spur-free zones (there may not be) it will be displayed as vertical
green rectangle.

In this example, as we can see in the figure, there are no spur-free zones. The legend in the upper
right-hand corner tells us which color each Mixer is associated with. If we wish more detailed
information about a spurious region, we can click on one of the rectangles:

6-105
6 RF Toolbox Examples

If we wish to find a spur-free zone, we will have to adjust some of the parameters of the setup.

Adjusting a Mixer Property to find Spur-Free Zones

In the current setup, there are no spur-free zones available. We will need to adjust some of the setup
parameters in order to find a spur-free zone. The values laid out in the design requirements (RF
Bandwidth, RF Center Frequency, and IF Bandwidth) cannot be changed. However, some parameters
(such as altering low- or high-side injection) are design decisions. We can see if changing the first
mixer to high-side injection will open up a spur-free zone:

h.Mixers(1).MixingType = 'high';
figure;
show(h);

6-106
 Finding Free IF Bandwidths

Adjusting the SpurFloor to find Spur-Free Zones

If we wish to use low-side injection in all of the mixers, we must find acceptable spur-free zones by
adjusting other parameters. Here we reset the OpenIF object to all low-side injection, and re-plot the
results:

h.Mixers(1).MixingType = 'low';
figure;
show(h);

6-107
6 RF Toolbox Examples

We notice there is a section around 500 MHz where there is an opening all the way down to roughly
-85 dBc. We can find that zone by adjusting the SpurFloor property:

h.SpurFloor = 85;
show(h);

6-108
 Finding Free IF Bandwidths

References

[1] Daniel Faria, Lawrence Dunleavy, and Terje Svensen. "The Use of Intermodulation Tables for
Mixer Simulations," Microwave Journal, Vol. 45, No. 4, December 2002, p. 60.

See Also

More About
• “Visualize Mixer Spurs” on page 6-95
• “Write S2P Touchstone Files” on page 6-92

6-109
6 RF Toolbox Examples

De-Embedding S-Parameters

This example shows you how to extract the S-parameters of a Device Under Test (DUT). First, read a
Touchstone® file into a sparameters object, second, calculate the S-parameters for the left and
right pads, third, de-embed the S-parameters using the deembedsparams function and finally display
the results.

This example uses the S-parameter data in the file samplebjt2.s2p that was collected from a
bipolar transistor in a fixture with a bond wire (series inductance of 1 nH) connected to a bond pad
(shunt capacitance of 100 fF) on the input, and a bond pad (shunt capacitance of 100 fF) connected to
a bond wire (series inductance of 1 nH) on the output, see Figure 1.

Figure 1: Device under test (DUT) and the test fixture.

This example will also show how to remove the effects of the fixture in order to extract the S-
parameters of the DUT.

Read Measured S-Parameters

Create a sparameters object for the measured S-parameters, by reading the Touchstone® data file,
samplebjt2.s2p.

S_measuredBJT = sparameters('samplebjt2.s2p');
freq = S_measuredBJT.Frequencies;

Calculate S-Parameters for Left Pad

Create a two port circuit object representing the left pad, containing a series inductor and a
shunt capacitor. Then calculate the S-parameters using the frequencies from samplebjt2.s2p.

leftpad = circuit('left');
add(leftpad,[1 2],inductor(1e-9));
add(leftpad,[2 3],capacitor(100e-15));
setports(leftpad,[1 3],[2 3]);
S_leftpad = sparameters(leftpad,freq);

Calculate S-Parameters for Right Pad

Create a two port circuit object representing the right pad, containing a series inductor and
shunt capacitor. Then, calculate the S-parameters using the frequencies from samplebjt2.s2p.

rightpad = circuit('right');
add(rightpad,[1 3],capacitor(100e-15));
add(rightpad,[1 2],inductor(1e-9));
setports(rightpad,[1 3],[2 3]);
S_rightpad = sparameters(rightpad,freq);

6-110
 De-Embedding S-Parameters

De-Embed S-Parameters

De-embed the S-parameters of the DUT from the measured S-parameters by removing the effects of
input and output pads (deembedsparams).
S_DUT = deembedsparams(S_measuredBJT,S_leftpad,S_rightpad);

Plot Measured and De-Embedded S11 Parameters on Z Smith® Chart

Use the smithplot function to plot the measured and de-embedded S11 parameters.
figure
hs = smithplot(S_measuredBJT,1,1);
hold on;
smithplot(S_DUT,1,1)
hs.ColorOrder = [1 0 0; 0 0 1];
hs.LegendLabels = {'Measured S11','De-Embedded S11'};

Plot Measured and De-Embedded S22 Parameters on Z Smith Chart

Use the smithplot function to plot the measured and de-embedded S22 parameters.
figure
hold off;
smithplot(S_measuredBJT,2,2)
hold on;
smithplot(S_DUT,2,2)
hs = smithplot('gco');

6-111
6 RF Toolbox Examples

hs.ColorOrder = [1 0 0; 0 0 1];
hs.LegendLabels = {'Measured S22','De-Embedded S22'};

Plot Measured and De-Embedded S21 Parameters in Decibels

Use the rfplot function to plot the measured and de-embedded S21 parameters.

figure
hold off;
h1 = rfplot(S_measuredBJT,2,1);
hold on;
h2 = rfplot(S_DUT,2,1);
legend([h1,h2],{'Measured S_{21}','De-Embedded S_{21}'});

6-112
 De-Embedding S-Parameters

See Also

More About
• “Extract S-Parameters from Circuit” on page 6-236
• “Extract S-Parameters from Mutual Inductor” on page 6-240
• “Bisect S-Parameters of Cascaded Probes” on page 6-114

6-113
6 RF Toolbox Examples

Bisect S-Parameters of Cascaded Probes

This example shows how to separate the S-parameters of two identical, passive, symmetric probes
connected in a cascade.

Introduction

Consider a DUT (device under test) connected to two probes. In order to de-embed the S-parameters
of DUT, you need to know the S-parameters of each individual probe. For accurate S-parameters of
the two probes, the calibration is done in the lab using SOLT (short, open, load, and thru) or TRL
(thru, reflect, line) measurements. However, if you assume the probes are identical and symmetric,
then you can approximate S-parameters quickly using the procedure sketched here.

The file connectedprobes.s2p contains the S-parameter data when the probes are connected
directly to each other.

ABCD-parameters

This example uses ABCD-parameters to bisect measured S-parameter data into the data for each
individual probe.

When you cascade two networks, you can calculate the ABCD-parameters of the combined network
by matrix multiplying the ABCD-parameters of the two individual networks.

Figure 1: Main network and network with two symmetric probes connected in a cascade

A B A1 B1 A2 B2
=
C D C1 D1 C2 D2

A1 B1 A2 B2 A1 B1 2
A B
If, = , then, =
C1 D1 C2 D2 C D C1 D1

From the above equation, you can find the ABCD-parameters of the two individual probes by taking
the matrix square root of the ABCD-parameters of main network.

Since both probes are identical, you can calculate the S-parameters of either one of the probes.

Extract Required S-Parameter Data from Given Touchstone file

Create an sparameters object from the Touchstone® data file connectedprobes.s2p.

filename = 'connectedprobes.s2p';
S = sparameters(filename);

6-114
 Bisect S-Parameters of Cascaded Probes

numports = S.NumPorts;
freq = S.Frequencies;
numfreq = numel(freq);
z0 = S.Impedance;

Calculate S-Parameter Data of Individual Probe

Create a zero matrix to store the ABCD-parameter data of the probe.

abcd_probe_data = zeros(numports,numports,numfreq);

To calculate S-Parameters of the probe, you need to know the S-parameters at every frequency it
operates. Convert the S-parameters extracted from connectedprobes.s2p to ABCD-parameters.
Then calculate the matrix square root of ABCD-parameters using sqrtm function to get the ABCD-
parameters of the probe. Convert these ABCD-parameters of the probe to S-parameters.

ABCD = abcdparameters(S);
for n = 1:numfreq
abcd_meas = ABCD.Parameters(:,:,n);
abcd_probe_data(:,:,n) = sqrtm(abcd_meas);
end
ABCD_probe = abcdparameters(abcd_probe_data,freq);

Create an S-parameter object from the calculated S-parameter data of the probe.

S_probe = sparameters(ABCD_probe,z0);

Compare Calculated S-Parameters with Expected S-Parameters

For this example, connectedprobes.s2p gives the S-Parameter data of this network.

Figure 2: Network derived from connectedprobes.s2p

Split the above network into two identical networks, probe1 and probe2. The S-parameters of these
probes represent the expected result.

6-115
6 RF Toolbox Examples

Figure 3: Identical Networks

Create probe1 using circuit, resistor, and capacitor objects from the RF Toolbox.

R1 = 1;
C1 = 1;
R2 = 1;
ckt = circuit('probe1');
add(ckt,[1 2],resistor(R1))
add(ckt,[2 4],capacitor(C1))
add(ckt,[2 3],resistor(R2))

Calculate the expected S-parameters of probe 1.

setports(ckt,[1 4],[3 4])

S_exp = sparameters(ckt,freq,z0);

Plot and compare the expected S-parameters from probe1 and those calculated using ABCD-
parameters and compare.

rfplot(S_exp,'+')
hold on
rfplot(S_probe,'--')
hold off
text(40,-30,{'+: Expected','--: Computed'})

6-116
 Bisect S-Parameters of Cascaded Probes

Compare Cascaded S-Parameters of Probe1 with S-Parameters of Combined Network

Cascade S-parameters of probe1 with itself using cascadesparams function and create an S-
parameter object with cascaded S-parameters.

S_combined = cascadesparams(S_probe,S_probe);

Plot and compare S-parameters from connectedprobes.s2p and those calculated from combined
probe1.

figure
rfplot(S,'+')
hold on
rfplot(S_combined,'--')
hold off
text(40,-27,{'+: Expected','--: Computed'})

Limitations

The procedure shown here cannot replace traditional calibration. We include it as an example of
using RF Toolbox™ and MATLAB™ to manipulate network parameters mathematically.

There are some limitations to using this procedure.

• There is no guaranteed solution. Some matrices do not have a square root.

6-117
6 RF Toolbox Examples

• The solution may not be unique. Often, there are two or more viable matrix square roots.

See Also

More About
• “De-Embedding S-Parameters” on page 6-110
• “Extract S-Parameters from Circuit” on page 6-236
• “Extract S-Parameters from Mutual Inductor” on page 6-240

6-118
 Designing Matching Networks for Low Noise Amplifiers

Designing Matching Networks for Low Noise Amplifiers

This example shows how to verify the design of input and output matching networks for a Low Noise
Amplifier (LNA) using gain and noise figure plot.

In wireless communications, receivers need to be able to detect and amplify incoming low-power
signals without adding much noise. Therefore, an LNA is often used as the first stage of these
receivers. To design an LNA, this example uses the available gain design technique, which involves
selecting an appropriate matching network that provides a suitable compromise between gain and
noise.

In this example, to design matching networks for an LNA, the rfckt.amplifier object and the
analyze method are used to examine the transducer power gains, the available power gain, and the
maximum available power gain. The method circle is used to determine optimal source reflection
coiefficent, GammaS and the function fzero is used in amplifier stabilization.

LNA Design Specifications

The LNA design specifications are as follows:

• Frequency range: 5.10 - 5.30 GHz

• Noise Figure <= 2.2 dB
• Transducer Gain > 11 dB
• Operating between 50-ohm terminations

Create rfckt.amplifier Object and Examine Amplifier Power Gains and Noise Figure

Create an rfckt.amplifier object to represent the amplifier that is specified in the file,
'samplelna1.s2p'. Analyze the amplifier using the analyze function the amplifier in the frequency
range from 2 - 10 GHz.

unmatched_amp = read(rfckt.amplifier, 'samplelna1.s2p');

analyze(unmatched_amp, 2e9:50e6:10e9);

Plot the transducer power gain (Gt), the available power gain (Ga) and the maximum available power
gain (Gmag).

figure
plot(unmatched_amp,'Gmag','Ga','Gt','dB')

6-119
6 RF Toolbox Examples

Examine the power gains at 5.2 GHz in order to design the input and output matching networks 5.2
GHz. Without the input and output matching networks, the transducer power gain at 5.2 GHz is about
7.2 dB. This is below the gain requirement of 11 dB in the design specifications and less than the
available power gain. This amplifier is also potentially unstable at 5.2 GHz, since the maximum
available gain does not exist at 5.2 GHz.

Plot the measured minimum noise figure (Fmin) and the noise figure (NF) calculated when there is no
input matching network. Specify an x-axis range of 4.9 GHz to 6 GHz, where the minimum noise
figure is measured.

plot(unmatched_amp,'Fmin','NF','dB')
axis([4.9 6 1.5 4])
legend('Location','NorthWest')

6-120
 Designing Matching Networks for Low Noise Amplifiers

In the absence of an input matching network, the noise figure is between 5.10 - 5.30 GHz which is
above the noise figure requirement of 2.2 dB in the specification.

Plot Gain, Noise Figure, and Stability Circles

Both the available gain and the noise figure are functions of the source reflection coefficient,
GammaS. To select an appropriate GammaS that provides a suitable compromise between gain and
noise, use the circle method of the rfckt.amplifier object to place the constant available gain
and the constant noise figure circles on the Smith chart. As mentioned earlier, the amplifier is
potentially unstable at 5.2 GHz. Therefore, the following circle command also places the input and
output stability circles on the Smith chart.

fc = 5.2e9;
hsm = smithplot;
circle(unmatched_amp,fc,'Stab','In','Stab','Out','Ga',10:2:20, ...
'NF',1.8:0.2:3,hsm);
legend('Location','SouthEast')

6-121
6 RF Toolbox Examples

Enable the data cursor and click on the constant available gain circle. The data tip displays the
following data:

• Available power gain (Ga)

• Noise figure (NF)
• Source reflection coefficient (GammaS)
• Output reflection coefficient (GammaOut)
• Normalized source impedance (ZS)

Ga, NF, GammaOut and ZS are all functions of the source reflection coefficient, GammaS. GammaS is the
complex number that corresponds to the location of the data cursor. A star ('*') and a circle-in-dashed-
line will also appear on the Smith chart. The star represents the matching load reflection coefficient
(GammaL) that is the complex conjugate of GammaOut. The gain is maximized when GammaL is the
complex conjugate of GammaOut. The circle-in-dashed-line represents the trajectory of the matching
GammaL when the data cursor moves on a constant available gain or noise figure circle.

6-122
 Designing Matching Networks for Low Noise Amplifiers

Because both the S11 and S22 parameters of the amplifier are less than unity in magnitude, both the
input and output stable region contain the center of the Smith chart. In order to make the amplifier
stable, GammaS must be in the input stable region and the matching GammaL must be in the output
stable region. The output stable region is shaded in the above figure. However, when a GammaS that
gives a suitable compromise between gain and noise is found, the matching GammaL always falls
outside the output stable region. This makes amplifier stabilization necessary.

Amplifier Stabilization

One way to stabilize an amplifier is to cascade a shunt resistor at the output of the amplifier.
However, this approach will also reduce gain and add noise. At the end of the example, you will notice
that the overall gain and noise still met the requirement.

To find the maximum shunt resistor value that makes the amplifier unconditionally stable, use the
fzero function to find the resistor value that makes stability MU equal to 1. The fzero function
always tries to achieve a value of zero for the objective function, so the objective function should
return MU-1.
type('lna_match_stabilization_helper.m')

function mu_minus_1 = lna_match_stabilization_helper(propval, fc, ckt, element, propname)

%LNA_MATCH_STABILIZATION_HELPER Return Stability MU-1.
% MU_MINUS_1 = LNA_MATCH_STABILIZATION_HELPER(PROPVALUE, FC, CKT,
% ELEMENT, PROPNAME) returns stability parameter MU-1 of a circuit, CKT
% when the property called PROPNAME of an element, ELEMENT is set to
% PROPVAL.
%

6-123
6 RF Toolbox Examples

% LNA_MATCH_STABILIZATION_HELPER is a helper function of RF

% Toolbox demo: Designing Matching Networks (Part 1: Networks with an LNA
% and Lumped Elements).

% Copyright 2007-2008 The MathWorks, Inc.

set(element, propname, propval)

analyze(ckt, fc);
mu_minus_1 = stabilitymu(ckt.AnalyzedResult.S_Parameters) - 1;

Compute the parameters for objective function and pass the objective function to fzero to get the
maximum shunt resistor value.

stab_amp = rfckt.cascade('ckts', {unmatched_amp, rfckt.shuntrlc});

R1 = fzero(@(R1) lna_match_stabilization_helper(R1,fc,stab_amp,stab_amp.Ckts{2},'R'),[1 1e5])

R1 =
118.6213

Find GammaS and GammaL

Cascade a 118-ohm resistor at the output of the amplifier and analyze the cascaded network. Place
the new constant available gain and the constant noise figure circles on the Smith chart.

shunt_r = rfckt.shuntrlc('R',118);
stab_amp = rfckt.cascade('ckts',{unmatched_amp,shunt_r});
analyze(stab_amp,fc);
hsm = smithplot;
circle(stab_amp,fc,'Ga',10:17,'NF',1.80:0.2:3,hsm)
legend('Location','SouthEast')

Use the data cursor to locate a GammaS. You can find that there is a suitable compromise between
gain and noise.

6-124
 Designing Matching Networks for Low Noise Amplifiers

The example is designined to select a GammaS that gives a gain of 14 dB and noise figure of 1.84 dB.
Compute the matching GammaL, which is the complex conjugate of GammaOut on the data tip.

GammaS = 0.67*exp(1j*153.6*pi/180)

GammaS =
-0.6001 + 0.2979i

Compute the normalized source impedance.

Zs = gamma2z(GammaS,1)

Zs =
0.2080 + 0.2249i

Compute the matching GammaL that is equal to the complex conjugate of GammaOut.

GammaL = 0.7363*exp(1j*120.1*pi/180)

GammaL =
-0.3693 + 0.6370i

Compute the normalized load impedance.

Zl = gamma2z(GammaL,1)

Zl =
0.2008 + 0.5586i

6-125
6 RF Toolbox Examples

Design Input Matching Network Using GammaS

In this example, the lumped LC elements are used to build the input and output matching networks as
follows:

The input matching network consists of one shunt capacitor, Cin, and one series inductor, Lin. Use
the Smith chart and the data cursor to find component values. To do this, start by plotting the
constant conductance circle that crosses the center of the Smith chart and the constant resistance
circle that crosses GammaS.

hsm = smithplot;
circle(stab_amp,fc,'G',1,'R',real(Zs),hsm);
hsm.GridType = 'YZ';
hold on
plot(GammaS,'k.','MarkerSize',16)
text(real(GammaS)+0.05,imag(GammaS)-0.05,'\Gamma_{S}','FontSize', 12, ...
'FontUnits','normalized')
plot(0,0,'k.','MarkerSize',16)
hold off

Then, find the intersection points of the constant conductance and the constant resistance circle.
Based on the circuit diagram above, the intersection point in the lower half of the Smith chart should
be used. Mark it as point A.

6-126
 Designing Matching Networks for Low Noise Amplifiers

GammaA = 0.6983*exp(1j*(-134.3)*pi/180);
Za = gamma2z(GammaA,1);
Ya = 1/Za;

Determine the value of Cin from the difference in susceptance from the center of the Smith chart to
point A. Namely,

Ya
2πf cCin = Im
50

where 50 is the reference impedance.

Cin = imag(Ya)/50/2/pi/fc

Cin =
1.1945e-12

Determine the value of Lin from the difference in reactance from point A to GammaS. Namely,

2πf cLin = 50 Im(Zs) − Im(Za)

Lin = (imag(Zs) - imag(Za))*50/2/pi/fc

Lin =
9.6522e-10

6-127
6 RF Toolbox Examples

Design Output Matching Network Using GammaL

Use the approach described in the previous section on designing the input matching network to
design the output matching network and get the values of Cout and Lout.

GammaB = 0.7055*exp(1j*(-134.9)*pi/180);
Zb = gamma2z(GammaB, 1);
Yb = 1/Zb;
Cout = imag(Yb)/50/2/pi/fc

Cout =
1.2194e-12

Lout = (imag(Zl) - imag(Zb))*50/2/pi/fc

Lout =
1.4682e-09

Verify Design

Create the input and output matching networks. Cascade the input matching network, the amplifier,
the shunt resistor and the output matching network to build the LNA.

input_match = rfckt.cascade('Ckts', ...

{rfckt.shuntrlc('C',Cin),rfckt.seriesrlc('L',Lin)});
output_match = rfckt.cascade('Ckts', ...
{rfckt.seriesrlc('L',Lout),rfckt.shuntrlc('C',Cout)});
LNA = rfckt.cascade('ckts', ...
{input_match,unmatched_amp,shunt_r,output_match});

Analyze the LNA around the design frequency range and plot the available and transducer power
gain. The available and transducer power gain at 5.2 GHz are both 14 dB as the design intended. The
transducer power gain is above 11 dB in the design frequency range, which meets the requirement in
the specification.

analyze(LNA,5.05e9:10e6:5.35e9);
plot(LNA,'Ga','Gt','dB');

Plot the noise figure around the design frequency range.

plot(LNA,'NF','dB')

The noise figure is below 2.2 dB in the design frequency range, which also meets the requirement in
the specification. The noise figure of the LNA at 5.2 GHz is about 0.1 dB above that of the amplifier
(1.84 dB), which demonstrates added noise by the shunt resistor.

The available gain design method is often used in LNA matching. In the second part of the example --
“Designing Matching Networks Using Single Stub Transmission Lines” on page 6-129, a simultaneous
conjugate matching example is presented.

See Also

More About
• “Designing Matching Networks Using Single Stub Transmission Lines” on page 6-129
• “Design Broadband Matching Networks for Antennas” on page 6-139

6-128
 Designing Matching Networks Using Single Stub Transmission Lines

Designing Matching Networks Using Single Stub Transmission

Lines

This example shows how to design an input and output matching networks for an amplifier. The input
and output matching networks for this amplifier element is designed using singe stub transmission
lines and this matching networks aims to maximize power delivered to a 50-ohm load and system. To
do so, this example:

• Calculates the reflection factors for simultaneous conjugate match.

• Determines the placement of a shunt stub in each matching network at a specified frequency.
• Cascades the matching networks with the amplifier and plots the results.

Create Amplifier Element

Create an amplifier object to represent the amplifier described by the measured frequency-
dependent S-parameter data in the file samplebjt2.s2p. Then, extract the frequency-dependent
S-parameter data from the rfckt.amplifier object.

amp = amplifier(UseNetworkData=true, FileName='samplebjt2.s2p');

S = amp.NetworkData;
Freq = S.Frequencies;

Check for Amplifier Stability

Before proceeding with the design, determine the measured frequencies at which the amplifier is
unconditionally stable. Use the stabilitymu function to calculate mu and muprime at each
frequency. Then, check that the returned values for mu are greater than one. This criteria is a
necessary and sufficient condition for unconditional stability. If the amplifier is not unconditionally
stable, print out the corresponding frequency value.

[mu,muprime] = stabilitymu(S);
figure
plot(Freq/1e9,mu,'--',Freq/1e9,muprime,'r')
legend('MU',"MU'",'Location','Best')
title("Stability Parameters MU and MU'")
xlabel('Frequency [GHz]')

6-129
6 RF Toolbox Examples

disp('Measured Frequencies where the amplifier is not unconditionally stable:')

Measured Frequencies where the amplifier is not unconditionally stable:

fprintf('\tFrequency = %.1e\n',Freq(mu<=1))

Frequency = 1.0e+09
Frequency = 1.1e+09

For this example, the amplifier is unconditionally stable at all measured frequencies except 1.0 GHz
and 1.1 GHz.

Determine Source and Load Matching Networks for Simultaneous Conjugate Match

Begin designing the input and output matching networks by transforming the reflection coefficients
for simultaneous conjugate match at the amplifier interfaces into the appropriate source and load
admittance. This example uses the following lossless transmission line matching scheme:

6-130
 Designing Matching Networks Using Single Stub Transmission Lines

The design parameters for this single stub matching scheme are the location of the stubs with
reference to the amplifier interfaces and the stub lengths. The procedure uses the following design
principles:

• The center of the Smith chart represents a normalized source or load immittance.
• Movement along a transmission line is equivalent to traversing a circle centered at the origin of
the Smith chart with radius equal to a reflection coefficient magnitude.
• A single transmission line stub can be inserted at the point on a transmission line when its
admittance (transmission line) intersects the unity conductance circle. At this location, the stub
will negate the transmission line susceptance, resulting in a conductance that equals the load or
source terminations.

This example uses the YZ Smith chart because it's easier to add a stub in parallel with a transmission
line using this type of Smith chart.

Calculate and Plot the Complex Load and Source Reflection Coefficients

Calculate and plot all complex load and source reflection coefficients for simultaneous conjugate
match at all measured frequency data points that are unconditionally stable. These reflection
coefficients are measured at the amplifier interfaces.

AllGammaL = gammaml(S);
AllGammaS = gammams(S);
hsm = smithplot(figure, [AllGammaL AllGammaS]);
hsm.LegendLabels = {'#Gamma ML','#Gamma MS'};

6-131
6 RF Toolbox Examples

Determine Load Reflection Coefficient at Single Frequency

Find the load reflection coefficient, GammaL, for the output matching network at the design frequency
1.9 GHz.

freq = Freq(Freq == 1.9e9);

GammaL = AllGammaL(Freq == 1.9e9)

GammaL =
-0.0421 + 0.2931i

Draw Constant Magnitude Circle for Load Reflection Coefficient GammaL

Draw a circle that is centered at the normalized admittance Smith chart origin and whose radius
equals the magnitude of GammaL. A point on this circle represents the reflection coefficient at a
particular location on the transmission line. The reflection coefficient for the transmission line at the
amplifier interface is GammaL, while the center of the chart represents the normalized load
admittance, y_L.

hsm = smithplot(figure);
hold on
d = (1:360);
x = abs(GammaL)*cosd(d);
y = abs(GammaL)*sind(d);
plot(x,y)
hsm.GridType = 'yz';
plot(0,0,'k.','MarkerSize',16)

6-132
 Designing Matching Networks Using Single Stub Transmission Lines

plot(GammaL,'k.','MarkerSize',16)
txtstr = sprintf('\\Gamma_{L}\\fontsize{8}\\bf=\\mid%s\\mid%s^\\circ', ...
num2str(abs(GammaL),4),num2str((angle(GammaL)*180/pi),4));
text(real(GammaL),imag(GammaL)+.1,txtstr,'FontSize',10, ...
'FontUnits','normalized');
plot(0,0,'r',0,0,'k.','LineWidth',2,'MarkerSize',16);
text(0.05,0,'y_L','FontSize',12,'FontUnits','normalized')

Draw Unity Constant Conductance Circle and Find Intersection Points

To determine the stub wavelength (susceptance) and its location with respect to the amplifier load
matching interface, plot the normalized unity conductance circle and the constant magnitude circle
and figure out where the two circles intersect. Find the points of intersection interactively using the
data cursor or analytically using the helper function,
imped_match_find_circle_intersections_helper. This example uses the helper function. The
circles intersect at two points. The example uses the third-quadrant point, which is labeled "A". The
unity conductance circle is centered at (-.5,0) with radius .5. The constant magnitude circle is
centered at (0,0) with radius equal to the magnitude of GammaL.

d = (1:360);
x = 0.5*cosd(d)-0.5;
y = 0.5*sind(d);
plot(x,y,'r');
% hsm.ColorOrder(2,:) = [1 0 0];
% figure
[~,pt2] = imped_match_find_circle_intersections_helper([0 0], ...
abs(GammaL),[-.5 0],.5);
GammaMagA = sqrt(pt2(1)^2 + pt2(2)^2);
GammaAngA = atan2(pt2(2),pt2(1));
% ax = hsm.Parent.CurrentAxes;
% hold (ax,"on");
plot(pt2(1),pt2(2),'k.','MarkerSize',16);
txtstr = sprintf('A=\\mid%s\\mid%s^\\circ',num2str(GammaMagA,4), ...
num2str(GammaAngA*180/pi,4));
text(pt2(1),pt2(2)-.07,txtstr,'FontSize',8,'FontUnits','normalized', ...
'FontWeight','Bold')
container = hsm.Parent;
annotation('textbox','VerticalAlignment','middle',...
'String',{'Unity','Conductance','Circle'},...
'HorizontalAlignment','center','FontSize',8,...
'EdgeColor',[0.04314 0.5176 0.7804],...
'BackgroundColor',[1 1 1],'Position',[0.1403 0.1608 0.1472 0.1396])
annotation('arrow',[0.2786 0.3286],[0.2778 0.3310])
annotation('textbox','VerticalAlignment','middle',...
'String',{'Constant','Magnitude','Circle'},...
'HorizontalAlignment','center','FontSize',8,...
'EdgeColor',[0.04314 0.5176 0.7804],...
'BackgroundColor',[1 1 1],'Position',[0.8107 0.3355 0.1286 0.1454])
annotation('arrow',[0.8179 0.5761],[0.4301 0.4887]);

6-133
6 RF Toolbox Examples

Calculate Stub Location and Stub Length for Output Matching Network

The open-circuit stub location in wavelengths from the amplifier load interface is a function of the
clockwise angular difference between point "A" and GammaL. When point "A" appears in the third
quadrant and GammaL falls in the second quadrant, the stub position in wavelengths is calculated as
follows:

StubPositionOut = ((2*pi + GammaAngA) - angle(GammaL))/(4*pi)

StubPositionOut =
0.2147

The stub value is the amount of susceptance that is required to move the normalized load admittance
(the center of the Smith chart) to point "A" on the constant magnitude circle. An open stub
transmission line can be used to supply this value of susceptance. Its wavelength is defined by the
amount of angular rotation from the open-circuit admittance point on the Smith chart (point "M" on
the following figure) to the required susceptance point "N" on the outer edge of the chart. Point "N" is
where a constant susceptance circle with a value equal to the susceptance of point "A" intersects the
unit circle. In addition, the StubLengthOut formula used below requires "N" to fall in the third or
fourth quadrant.

6-134
 Designing Matching Networks Using Single Stub Transmission Lines

Calculate the stub length of the output matching metwork.

GammaA = GammaMagA*exp(1j*GammaAngA);
bA = imag((1 - GammaA)/(1 + GammaA));
StubLengthOut = -atan2(-2*bA/(1 + bA^2),(1 - bA^2)/(1 + bA^2))/(4*pi)

StubLengthOut =
0.0883

Calculate Stub Location and Stub Length for Input Matching Network

In the previous sections, the example calculated the required lengths and placements, in
wavelengths, for the output matching transmission network. Following the same approach, the line
lengths for the input matching network are calculated:
GammaS = AllGammaS(Freq == 1.9e9)

GammaS =
-0.0099 + 0.2501i

Calculate the stub location of the input matching network.

[pt1,pt2] = imped_match_find_circle_intersections_helper([0 0], ...
abs(GammaS),[-.5 0],.5);
GammaMagA = sqrt(pt2(1)^2 + pt2(2)^2);
GammaAngA = atan2(pt2(2),pt2(1));
GammaA = GammaMagA*exp(1j*GammaAngA);
bA = imag((1 - GammaA)/(1 + GammaA));
StubPositionIn = ((2*pi + GammaAngA) - angle(GammaS))/(4*pi)

6-135
6 RF Toolbox Examples

StubPositionIn =
0.2267

Calculate the stub length of the input matching metwork.

StubLengthIn = -atan2(-2*bA/(1 + bA^2),(1 - bA^2)/(1 + bA^2))/(4*pi)

StubLengthIn =
0.0759

Verify Design

To verify the design, assemble a circuit using 50-Ohm microstrip transmission lines for the matching
networks. First, determine if the microstrip line is a suitable choice by analyzing the default
microstrip transmission line at a design frequency of 1.9 GHz.

stubTL4 = txlineMicrostrip;
[Z0, Eeff_f] = getZ0(stubTL4);

This characteristic impedance is close to the desired 50-Ohm impedance, so the example can proceed
with the design using these microstrip lines.

To calculate the required transmission line lengths in meters for the placement of the stubs, analyze
the microstrip to obtain a phase velocity value.

phase_vel = rf.physconst("LightSpeed")./sqrt(Eeff_f);

Use the phase velocity value, which determines the transmission line wavelength and the stub
location to set the appropriate transmission line lengths for the two microstrip transmission lines,
TL2 and TL3.

TL2 = txlineMicrostrip('LineLength',phase_vel/freq*StubPositionIn);
TL3 = txlineMicrostrip('LineLength',phase_vel/freq*StubPositionOut);

Use the phase velocity again to specify stub length and stub mode for each stub.

stubTL1 = txlineMicrostrip('LineLength',phase_vel/freq*StubLengthIn, ...

'StubMode','shunt','Termination','open');
set(stubTL4,'LineLength',phase_vel/freq*StubLengthOut, ...
'StubMode','shunt','Termination','open')

Now cascade the circuit elements and analyze the amplifier with and without the matching networks
over the frequency range of 1.5 to 2.3 GHz.

matched_amp = circuit([stubTL1 TL2 amp TL3 stubTL4]);

Samp = sparameters(matched_amp, (1.5e9:1e7:2.3e9));

To verify the simultaneous conjugate match at the input of the amplifier, plot the S11 parameters in
dB for both the matched and unmatched circuits.

clf
rfplot(S,1,1,'dB')
hold on
rfplot(Samp,1,1,'dB')
legend('S_{11} - Original Amplifier', 'S_{11} - Matched Amplifier')
legend('Location','SouthEast')
xlim([1.5 2.3])
hold off

6-136
 Designing Matching Networks Using Single Stub Transmission Lines

To verify the simultaneous conjugate match at the output of the amplifier, plot the S22 parameters in
dB for both the matched and unmatched circuits.

rfplot(S,2,2,'dB')
hold on
rfplot(Samp,2,2,'dB')
legend('S_{22} - Original Amplifier', 'S_{22} - Matched Amplifier')
legend('Location','SouthEast')
xlim([1.5 2.3])
hold off

Finally, plot the transducer gain (Gt) and the maximum available gain (Gmag) in dB for the matched
circuit.

Gt = 10*log10(powergain(Samp,'Gt'));
Gmag = 10*log10(powergain(Samp,'Gmag'));
figure;
plot(Samp.Frequencies/1e9, [Gt Gmag])
legend({'Gt','Gmag'})
xlabel('Frequency (GHz)')
ylabel('Gain (dB)')

6-137
6 RF Toolbox Examples

You can see that the transducer gain and the maximum available gain are very close to each other at
1.9 GHz.

See Also

More About
• “Designing Matching Networks for Low Noise Amplifiers” on page 6-119
• “Design Broadband Matching Networks for Antennas” on page 6-139
• “Design Broadband Matching Networks for Amplifier” on page 6-148

6-138
 Design Broadband Matching Networks for Antennas

Design Broadband Matching Networks for Antennas

This example shows how to design a broadband matching network between a resistive source and
inductive load using optimization with direct search methods.

In an RF system, a matching network circuit plays a vital role in transferring maximum power
between source and the load of the system. In most RF systems, such as wireless devices, a design
parameter called operation bandwidth is specified. By taking the operation bandwidth into
consideration, the purpose of the matching network is further extended to provide maximum power
transfer over a range of frequencies. Alternatively, you can use the L - section matching (conjugate
match) approach, guarantees maximum power transfer it does so only at a single frequency.

Figure 1: Impedance matching of an antenna to a source

To design a broadband matching network, first set the design parameters such as center frequency,
bandwidth, and impedances of source, load and reference. Then calculate the load reflection
coefficient and power gain to determine the frequency at which the matching network of the antenna
must operate and once the design is complete, optimize the derived network.

Specify Frequency and Impedance

Specify the center frequency, 350 MHz, and bandwidth, 110 MHz, of match to build a matching
network with a bandpass response.

fc = 350e6;
BW = 110e6;

Specify the source impedance, the reference impedance and the load resistance. In this example the
load Zl is modeled as a series R-L circuit. Instead of calculating the load impedance, you could
measure the impedance of the load.

Zs = 50; % Source impedance (ohm)

Z0 = 50; % Reference impedance (ohm)
Rl = 40; % Load resistance (ohm)
L = 12e-8; % Load inductance (Henry)

Define the number of frequency points to use for analysis and set up the frequency vector.

6-139
6 RF Toolbox Examples

nfreq = 256; % Number of frequency points

fLower = fc - (BW/2); % Lower band edge
fUpper = fc + (BW/2); % Upper band edge
freq = linspace(fLower,fUpper,nfreq); % Frequency array for analysis
w = 2*pi*freq; % Frequency (radians/sec)

Understand Load Behavior using Reflection Coefficient and Power Gain

Use two simple expressions for calculating the load reflection coefficient and the power gain. This
corresponds to directly connecting the source to the input terminals of an antenna i.e. in Figure 1
there is no matching network.
Xl = w*L; % Reactance (ohm)
Zl = Rl + 1i*Xl; % Load impedance (ohm)
GammaL = (Zl - Z0)./(Zl + Z0); % Load reflection coefficient
unmatchedGt = 10*log10(1 - abs(GammaL).^2); % Power delivered to load

Use the smithplot function to plot the variation in the load reflection coefficient with frequency. An
input reflection coefficient closer to center of the Smith chart denotes a better matching
performance.
figure
smithplot(freq,GammaL,'LegendLabels','#Gamma load','LineWidth',2,...
'View','top-right');

This plot shows that the load reflection coefficient is far away from this point. Therefore, there is an
impedance mismatch. You can confirm this mismatch by plotting the transducer gain as a function of
frequency.

6-140
 Design Broadband Matching Networks for Antennas

figure
plot(freq.*1e-6,unmatchedGt,'r')
grid on;
title('Power delivered to load - No matching network');
xlabel('Frequency (MHz)');
ylabel('Magnitude (decibels)');
legend('G_t','Location','Best');

As the plot shows, there is approximately 10 dB power loss around the desired region of operation
(295 - 405 MHz). As a result, the antenna needs a matching network that operates over a 110 MHz
bandwidth centered at 350 MHz.

Design Matching Network

The matching network must operate between 295 MHz and 405 MHz, therefore you choose a
bandpass topology for the matching network shown below.

Type - I: Series LC first element followed by shunt LC

6-141
6 RF Toolbox Examples

Figure 2: Matching network topology

The approach is to design an odd order 0.5 dB Chebyshev bandpass to obtain the initial design for the
matching network shown in figure 2. This is a single match problem [1] on page 6-147, i.e. the source
is purely resistive while load is a combination of R and L, solution is you can begin by choosing a five-
element prototype network.

N = 5; % Order of matching network

filter = rffilter('FilterType',"Chebyshev","FilterOrder",N, ...
"Implementation","LC Tee","ResponseType","Bandpass",...
"PassbandFrequency",[fLower fUpper],"PassbandAttenuation",0.5);
Lvals = filter.DesignData.Inductors;

Use the lcladder object to build the bandpass tee matching network. Note that the topology
demands a bandpass tee prototype that begins with a series inductor. If the topology chosen is an LC
bandpass pi then you would begin with shunt C for the lowpass prototype.

% Create the matching network

matchingNW = lcladder(filter);

% Copy initial values for comparison

L_initial = Lvals;

Optimize Designed Matching Network

There are several factors to consider prior to the optimization.

• Objective function - The objective function can be built in different ways depending on the
problem at hand. For this example, the objective function is shown in the file below.
• Choice of cost function - The cost function is the function we would like to minimize (maximize) to
achieve near optimal performance. There could be several ways to choose the cost function. One
obvious choice is the input reflection coefficient, gammaIn. In this example we have chosen to
minimize the average reflection coefficient in the passband.
• Optimization variables - In this case it is a vector of values, for the specific elements to optimize in
the matching network.
• Optimization method - A direct search based technique, the MATLAB® function fminsearch, is
used in this example to perform the optimization.

6-142
 Design Broadband Matching Networks for Antennas

• Number of iterations/function evaluations - Set the maximum number of iterations and function
evaluations to perform, so as to tradeoff between speed and quality of match.

The objective function used during the optimization process by fminsearch is shown here.

type('antennaMatchObjectiveFun.m')

function output = antennaMatchObjectiveFun(matchingNW,Lvalues,freq,ZL,Z0)

%ANTENNAMATCHOBJECTIVEFUN is the objective function used by the example
% Designing Broadband Matching Networks for Antennas.
%
% OUTPUT = ANTENNAMATCHOBJECTIVEFUN(MATCHINGNW,LVALUES,FREQ,Z0)
% returns the current value of the objective function stored in OUTPUT
% evaluated after updating the inductor values in the object, MATCHINGNW.
% The inductor values are stored in the variable LVALUES.
%
% ANTENNAMATCHOBJECTIVEFUN is an objective function of RF Toolbox demo:
% Designing Broadband Matching Networks (Part I: Antenna)

% Copyright 2008-2020 The MathWorks, Inc.

% Ensure positive element values

if any(Lvalues <= 0)
output = Inf;
return
end

% Update the element values in the matching network

matchingNW.Inductances(1) = Lvalues(1);
matchingNW.Inductances(end) = Lvalues(end);

% Perform analysis on tuned matching network

S = sparameters(matchingNW,freq,Z0);

% Calculate input reflection coefficient 'gammaIn'

gIn = gammain(S,ZL);

% Cost function
output = mean(abs(gIn));

% Other possible choices for objective function could be : -

% output = max(abs(gIn));
% output = -1*mean(Gt_pass);

% Animate
smithplot(freq,gIn);
drawnow

There are several ways to choose the cost function and some options are shown within the objective
function above (in comments). The optimization variables are the first and last inductors, L1 and L5
respectively. The element values are stored in the variable L_Optimized.

niter = 125;
options = optimset('Display','iter','MaxIter',niter); % Set options structure
L_Optimized = [Lvals(1) Lvals(end)];
L_Optimized = ...
fminsearch(@(L_Optimized)antennaMatchObjectiveFun(matchingNW, ...
L_Optimized,freq,Zl,Z0),L_Optimized,options);

6-143
6 RF Toolbox Examples

Iteration Func-count f(x) Procedure

0 1 0.933982
1 3 0.933982 initial simplex
2 5 0.920323 expand
3 7 0.911353 expand
4 9 0.853255 expand
5 11 0.730444 expand
6 13 0.526448 reflect
7 15 0.526448 contract inside
8 17 0.421103 reflect
9 19 0.421103 contract inside
10 20 0.421103 reflect
11 22 0.421103 contract inside
12 24 0.421103 contract inside
13 26 0.339935 expand
14 27 0.339935 reflect
15 29 0.28528 reflect
16 31 0.28528 contract inside
17 32 0.28528 reflect
18 34 0.283527 reflect
19 36 0.283527 contract inside
20 38 0.278939 contract inside
21 40 0.278123 reflect
22 41 0.278123 reflect
23 43 0.27636 contract inside
24 45 0.275782 contract inside
25 47 0.275637 contract inside
26 49 0.275498 reflect
27 51 0.275282 contract inside
28 52 0.275282 reflect
29 54 0.275282 contract inside
30 56 0.275282 contract inside

Optimization terminated:
the current x satisfies the termination criteria using OPTIONS.TolX of 1.000000e-04
and F(X) satisfies the convergence criteria using OPTIONS.TolFun of 1.000000e-04

6-144
 Design Broadband Matching Networks for Antennas

Update Matching Network Elements with Optimal Values

When the optimization routine stops, the optimized element values are stored in L_Optimized. The
following code updates the input and output matching network with these values.
matchingNW.Inductances(1) = L_Optimized(1); % Update the matching network inductor L1
matchingNW.Inductances(end) = L_Optimized(end); % Update the matching network inductor L5

Analyze and Display Optimization Results

Compare and plot the input reflection coefficient of the matched and unmatched results.
S = sparameters(matchingNW,freq,Z0);
gIn = gammain(S,Zl);
smithplot(freq,[gIn transpose(GammaL)],'LegendLabels',...
{'#Gamma in (Matched)','#Gamma in (Unmatched)'})

The optimized matching network improves the performance of the circuit. In the passband (295 - 405
MHz), the input reflection coefficient is closer to the center of the Smith chart. Plot the power
delivered to load for both the matched and unmatched system.
matchedGt = powergain(S,Zs,Zl,'Gt');
figure;
plot(freq*1e-6,matchedGt)
hold on;
plot(freq*1e-6,unmatchedGt,'r')
grid on;
hold off;

6-145
6 RF Toolbox Examples

title('Power delivered to load');

legend('Optimized network','No matching network','Location','Best');

The power delivered to the load is approximately 1 dB down for the optimized matching network.

Display Optimized Element Values

The following code shows the initial and optimized values for inductors L1 and L5.

L1_Initial = L_initial(1)

L1_Initial =
1.2340e-07

L1_Optimized = L_Optimized(1)

L1_Optimized =
1.2111e-07

L5_Initial = L_initial(end)

L5_Initial =
1.2340e-07

L5_Optimized = L_Optimized(end)

L5_Optimized =
1.7557e-09

6-146
 Design Broadband Matching Networks for Antennas

There are few things to consider when setting up an optimization:

• Choosing a different objective function would change the result.

• You can use advanced direct search optimization functions such as patternsearch and
simulannealband in your optimization, but you must have the Global Optimization Toolbox™
installed to access them.

References

1 Cuthbert, Thomas R. Broadband Direct-Coupled and Matching RF Networks. TRCPEP, 1999.

2 Ludwig, Reinhold, and Pavel Bretchko. RF Circuit Design: Theory and Applications. Prentice-Hall,
2000.
3 Pozar, David. Microwave Engineering. 2nd ed., John Wiley and Sons, 1999.

See Also

“Design Broadband Matching Networks for Amplifier” on page 6-148.

See Also

More About
• “Design Broadband Matching Networks for Amplifier” on page 6-148
• “Impedance Matching of Small Monopole Antenna” on page 6-166

6-147
6 RF Toolbox Examples

Design Broadband Matching Networks for Amplifier

This example shows how to design broadband matching networks for a low noise amplifier (LNA) with
ideal and real-world lumped LC elements. The real-world lumped LC elements are obtained from the
Modelithics SELECT+ Library™. The LNA is designed to the target gain and noise figure
specifications over a specified bandwidth. A direct-search based approach is used to arrive at the
optimum element values in the input and output matching networks.

In an RF receiver front end, the LNA is commonly found immediately after the antenna or after the
first bandpass filter that follows the antenna. Its position in the receiver chain ensures that it deals
with weak signals that have significant noise content. As a result the LNA has to not only provide
amplification to such signals but also minimize its own noise footprint on the amplified signal. This is
depicted in figure 1.

Figure 1: Impedance matching of an amplifier

Set Design Parameters

The design specifications are as follows.

• Amplifier is an LNA amplifier

• Center Frequency = 250 MHz
• Bandwidth = 100 MHz
• Transducer Gain greater than or equal to 10 dB
• Noise Figure less than or equal to 2.0 dB
• Operating between 50-Ohm terminations

Specify Design Parameters

You are building the matching network for an LNA with a bandpass response, so specify the
bandwidth of match, center frequency, gain, and noise figure targets.
BW = 100e6; % Bandwidth of matching network (Hz)
fc = 250e6; % Center frequency (Hz)
Gt_target = 10; % Transducer gain target (dB)
NFtarget = 2; % Max noise figure target (dB)

Specify the source impedance, reference impedance, and the load impedance.
Zs = 50; % Source impedance (Ohm)
Z0 = 50; % Reference impedance (Ohm)
Zl = 50; % Load impedance (Ohm)

6-148
 Design Broadband Matching Networks for Amplifier

Create Amplifier Object and Perform Analysis

Create an amplifier object from lnadata.s2p.

Mismatched_Amp = amplifier(FileName='lnadata.s2p');

Define the number of frequency points to use for analysis and set up the frequency vector.

Npts = 32; % No. of analysis frequency points

fLower = fc - (BW/2); % Lower band edge
fUpper = fc + (BW/2); % Upper band edge
freq = linspace(fLower,fUpper,Npts); % Frequency array for analysis

Perform frequency domain analysis on the mismatched amplifier.

sparam = sparameters(Mismatched_Amp,freq);
bMismatched = rfbudget(Mismatched_Amp,freq',-30,BW);
[K,~,~,Delta] = stabilityk(sparam);
Ga = powergain(sparam,'Ga');

Examine Stability, Power Gain, and Noise Figure

Plot the LNA stability parameters and examine stability, power gain and noise figure.

figure
plot(freq*1e-6,abs(Delta))
hold on
plot(freq*1e-6,K)
legend('Delta','K','Location','best')
title('Device stability parameters')
xlabel('Frequency [MHz]')
ylabel('Magnitude')
grid on
hold off

6-149
6 RF Toolbox Examples

As the plot shows, K > 1 and Δ < 1 for all frequencies in the bandwidth of interest. This means that
the device is unconditionally stable. It is also important to view the power gain and noise figure
behavior across the same bandwidth. Together with the stability information this data allows you to
determine if the gain and noise figure targets can be met.

figure
plot(freq*1e-6,10*log10(Ga))
hold on
plot(freq*1e-6,bMismatched.TransducerGain)
legend('Ga','Gt','Location','east')
xlabel('Frequency [MHz]')
ylabel('Magnitude [dB]')
hold off
grid on

6-150
 Design Broadband Matching Networks for Amplifier

This plot, shows the power gain across the 100-MHz bandwidth. It indicates that the transducer gain
varies linearly between 5.5 dB to about 3.1 dB and achieves only 4.3 dB at band center. It also
suggests there is sufficient headroom between the transducer gain Gt and the available gain Ga to
achieve our target Gt of 10 dB.

figure
plot(Mismatched_Amp.NoiseData.Frequencies*1e-6,Mismatched_Amp.NoiseData.Fmin)
hold on
plot(freq*1e-6,bMismatched.NF)
hold off
axis([200 300 0 2])
legend('Fmin','NF','Location','NorthEast')
xlabel('Frequency [MHz]')
ylabel('Magnitude [dB]')
grid on

6-151
6 RF Toolbox Examples

This plot shows the variation of the noise figure with frequency. The unmatched amplifier clearly
meets the target noise figure requirement. However this would change once the input and output
matching networks are included. Most likely, the noise figure of the LNA would exceed the
requirement.

Design Input and Output Matching Networks

The region of operation is between 200 — 300 MHz. Therefore, choose a bandpass topology for the
matching networks which is shown here.

6-152
 Design Broadband Matching Networks for Amplifier

Figure 2: Matching network topology

The topology chosen, as seen in Figure 2, is a direct-coupled prototype bandpass network of parallel
resonator type with top coupling [2], that is initially tuned to the geometric mean frequency with
respect to the bandwidth of operation.
N = 3; % Order of input/output matching network
wU = 2*pi*fUpper; % Upper band edge
wL = 2*pi*fLower; % Lower band edge
w0 = sqrt(wL*wU); % Geometric mean

For the initial design all the inductors are assigned the same value on the basis of the first series
inductor. As mentioned in [3], choose the prototype value to be unity and use standard impedance
and frequency transformations to obtain denormalized values [1]. The value for the capacitor in the
parallel trap is set using this inductor value to make it resonate at the geometric mean frequency.
Note that there are many ways of designing the initial matching network. This example shows one
possible approach.
LvaluesIn = (Zs/(wU-wL))*ones(N,1); % Series and shunt L's [H]
CvaluesIn = 1 / ( (w0^2)*LvaluesIn(2)); % Shunt C [F]
LvaluesOut = LvaluesIn;
CvaluesOut = CvaluesIn;

Build Matching Networks

Build each branch of the matching network and then define the circuit as a two-port network. This
example uses identical input and output matching networks.
InputMatchingNW = circuit;
add(InputMatchingNW,[1 2],inductor(LvaluesIn(1)));
add(InputMatchingNW,[2 0],inductor(LvaluesIn(2)));
add(InputMatchingNW,[2 0],capacitor(CvaluesIn));
add(InputMatchingNW,[2 3],inductor(LvaluesIn(3)));
setports(InputMatchingNW,[1 0],[3 0]);

OutputMatchingNW = circuit;
add(OutputMatchingNW,[1 2],inductor(LvaluesOut(1)));
add(OutputMatchingNW,[2 0],inductor(LvaluesOut(2)));
add(OutputMatchingNW,[2 0],capacitor(CvaluesOut));
add(OutputMatchingNW,[2 3],inductor(LvaluesOut(3)));
setports(OutputMatchingNW,[1 0],[3 0]);

Optimize Input & Output Matching Network

There are several points to consider prior to the optimization.

• Objective function: The objective function can be built in different ways depending on the problem
at hand. For this example, the objective function is shown in the file below.
• Choice of cost function: The cost function is the function you would like to minimize (maximize) to
achieve near optimal performance. There could be several ways to choose the cost function. For
this example you have two requirements to satisfy simultaneously, i.e. gain and noise figure. To
create the cost function you first, find the difference, between the most current optimized network
and the target value for each requirement at each frequency. The cost function is the L2-norm of
the vector of gain and noise figure error values.
• Optimization variables: In this case it is a vector of values, for the specific elements to optimize in
the matching network.

6-153
6 RF Toolbox Examples

• Optimization method: A direct search based technique, the MATLAB® function fminsearch, is
used in this example to perform the optimization.
• Number of iterations/function evaluations: Set the maximum no. of iterations and function
evaluations to perform, so as to tradeoff between speed and quality of match.
• Tolerance value: Specify the variation in objective function value at which the optimization
process should terminate.

The objective function used during the optimization process by fminsearch is shown here.
type('broadband_match_amplifier_objective_function.m')

function output = broadband_match_amplifier_objective_function(AMP,inMNW,outMNW,LC_Optim,freq,Gt_

%BROADBAND_MATCH_AMPLIFIER_OBJECTIVE_FUNCTION is the objective function.
% OUTPUT = BROADBAND_MATCH_AMPLIFIER_OBJECTIVE_FUNCTION(AMP,INMNW,OUTMNW,LC_OPTIM,FREQ,GT_TARGET,
% returns the current value of the objective function stored in OUTPUT
% evaluated after updating the element values in the object, inMNW and
% outMNW. The inductor and capacitor values are stored in the variable
% LC_OPTIM.
%
% BROADBAND_MATCH_AMPLIFIER_OBJECTIVE_FUNCTION is an objective function of RF Toolbox demo:
% Designing Broadband Matching Networks (Part II: Amplifier)

% Copyright 2023 The MathWorks, Inc.

% Ensure positive element values

if any(LC_Optim<=0)
output = inf;
return;
end
% Update matching network elements - The object inMNW and outMNW have
% several properties among which the cell array 'Elements' consists of all
% LC elements in the circuit.
inMNW.Elements(1).Inductance = LC_Optim(1);
inMNW.Elements(2).Inductance = LC_Optim(2);
inMNW.Elements(4).Inductance = LC_Optim(3);
inMNW.Elements(3).Capacitance = LC_Optim(4);
outMNW.Elements(1).Inductance = LC_Optim(5);
outMNW.Elements(2).Inductance = LC_Optim(6);
outMNW.Elements(4).Inductance = LC_Optim(7);
outMNW.Elements(3).Capacitance = LC_Optim(8);

% Perform analysis on tuned matching network

Npts = length(freq);
BW = max(freq) - min(freq);
np1 = nport(sparameters(inMNW,freq));
np2 = nport(sparameters(outMNW,freq));
b = rfbudget([np1 clone(AMP) np2],freq',-30,BW);

% Calculate transducer power gains and noise figures of the amplifier

Gt = b.TransducerGain(:,end);
NF_amp = b.NF(:,end);

% Calculate target gain and noise figure error

errGt = (Gt - Gt_target);
errNF = (NF_amp - NF);

% Check to see if gain and noise figure target are achieved by specifying

6-154
 Design Broadband Matching Networks for Amplifier

% bounds for variation.

deltaG = 0.40;
deltaNF = -0.05;
errGt(abs(errGt)<=deltaG) = 0;
errNF(errNF<deltaNF) = 0;

% Cost function
err_vec = [errGt;errNF];
output = norm((err_vec),2);

% Animate
Gmax = (Gt_target + deltaG).*ones(1,Npts);
Gmin = (Gt_target - deltaG).*ones(1,Npts);
plot(freq*1e-6,Gt)
hold on
plot(freq*1e-6,NF_amp)
plot(freq.*1e-6,Gmax,'r-*')
plot(freq.*1e-6,Gmin,'r-*')
legend('G_t','NF','Gain bounds','Location','East');
xlabel('Freq [MHz]')
ylabel('Magnitude [dB]')
axis([freq(1)*1e-6 freq(end)*1e-6 0 Gt_target+2]);
hold off
drawnow;

The optimization variables are all the elements (inductors and capacitors) of the input and output
matching networks.

nIter = 125; % Max No of Iterations

options = optimset('Display','iter','TolFun',1e-2,'MaxIter',nIter); % Set options structure
LC_Optimized = [LvaluesIn;CvaluesIn;LvaluesOut;CvaluesOut];
LC_Optimized = fminsearch(@(LC_Optimized) broadband_match_amplifier_objective_function(Mismatched
OutputMatchingNW,LC_Optimized,freq,Gt_target,NFtarget),LC_Optimized,

Iteration Func-count f(x) Procedure

0 1 30.4869
1 9 28.3549 initial simplex
2 11 25.5302 expand
3 12 25.5302 reflect
4 13 25.5302 reflect
5 14 25.5302 reflect
6 16 22.8228 expand
7 17 22.8228 reflect
8 19 19.0289 expand
9 20 19.0289 reflect
10 21 19.0289 reflect
11 22 19.0289 reflect
12 24 14.8785 expand
13 25 14.8785 reflect
14 27 10.721 expand
15 28 10.721 reflect
16 29 10.721 reflect
17 31 9.84796 expand
18 32 9.84796 reflect
19 33 9.84796 reflect
20 34 9.84796 reflect
21 35 9.84796 reflect

6-155
6 RF Toolbox Examples

22 37 9.84796 contract outside

23 39 9.84796 contract outside
24 41 9.84796 contract inside
25 43 9.64666 reflect
26 45 9.64666 contract inside
27 46 9.64666 reflect
28 48 9.64666 contract inside
29 49 9.64666 reflect
30 51 9.64666 contract inside
31 53 7.9372 expand
32 55 7.9372 contract outside
33 56 7.9372 reflect
34 57 7.9372 reflect
35 58 7.9372 reflect
36 59 7.9372 reflect
37 60 7.9372 reflect
38 62 5.98211 expand
39 63 5.98211 reflect
40 64 5.98211 reflect
41 65 5.98211 reflect
42 66 5.98211 reflect
43 68 4.31973 expand
44 70 4.31973 contract inside
45 71 4.31973 reflect
46 72 4.31973 reflect
47 73 4.31973 reflect
48 74 4.31973 reflect
49 75 4.31973 reflect
50 77 2.83135 expand
51 79 1.17624 expand
52 80 1.17624 reflect
53 81 1.17624 reflect
54 82 1.17624 reflect
55 84 0.691645 reflect
56 85 0.691645 reflect
57 86 0.691645 reflect
58 88 0.691645 contract inside
59 90 0.691645 contract outside
60 91 0.691645 reflect
61 93 0.691645 contract inside
62 95 0.691645 contract inside
63 96 0.691645 reflect
64 97 0.691645 reflect
65 98 0.691645 reflect
66 100 0.691645 contract inside
67 102 0.691645 contract outside
68 103 0.691645 reflect
69 105 0.691645 contract inside
70 107 0.497434 reflect
71 109 0.497434 contract inside
72 111 0.497434 contract inside
73 112 0.497434 reflect
74 114 0.497434 contract inside
75 116 0.497434 contract inside
76 118 0.444957 reflect
77 120 0.402851 expand
78 122 0 reflect
79 123 0 reflect

6-156
 Design Broadband Matching Networks for Amplifier

80 125 0 contract inside

81 127 0 contract inside
82 128 0 reflect
83 129 0 reflect
84 130 0 reflect
85 131 0 reflect
86 132 0 reflect
87 133 0 reflect
88 134 0 reflect
89 135 0 reflect
90 137 0 contract inside
91 139 0 contract outside

Optimization terminated:
the current x satisfies the termination criteria using OPTIONS.TolX of 1.000000e-04
and F(X) satisfies the convergence criteria using OPTIONS.TolFun of 1.000000e-02

Update Matching Network and Analyze LNA

Update the input and output matching networks with optimized element values, LC_Optimized.

InputMatchingNW.Elements(1).Inductance = LC_Optimized(1);
InputMatchingNW.Elements(2).Inductance = LC_Optimized(2);
InputMatchingNW.Elements(4).Inductance = LC_Optimized(3);
InputMatchingNW.Elements(3).Capacitance = LC_Optimized(4);
OutputMatchingNW.Elements(1).Inductance = LC_Optimized(5);
OutputMatchingNW.Elements(2).Inductance = LC_Optimized(6);

6-157
6 RF Toolbox Examples

OutputMatchingNW.Elements(4).Inductance = LC_Optimized(7);
OutputMatchingNW.Elements(3).Capacitance = LC_Optimized(8);

Analyze the S-parameters of input and output matching networks and then convert the S-parameter
data into an nport object.

s1 = sparameters(InputMatchingNW,freq);
np1 = nport(s1);
s2 = sparameters(OutputMatchingNW,freq);
np2 = nport(s2);

Use the rfbudget object to calculate the transducer gain and the noise figure across the bandwidth.
Compute RF budget of the LNA network. The LNA network comprises of matching network and low
noise amplifier.

bMatched = rfbudget([np1 clone(Mismatched_Amp) np2],freq',-30,BW);

Verify Design

Plot the transducer gain and the noise figure and then compare the optimization results with the
mismatched network.

plot(freq*1e-6,bMatched.TransducerGain(:,end))
hold on
plot(freq*1e-6,bMismatched.TransducerGain(:,end))
plot(freq*1e-6,bMatched.NF(:,end))
plot(freq*1e-6,bMismatched.NF(:,end))
legend('G_t - Matched','G_t - Mismatched','NF - Matched',...
'NF - Mismatched','Location','East')
xlabel('Frequency [MHz]')
ylabel('Magnitude [dB]')
axis([freq(1)*1e-6 freq(end)*1e-6 0 12])
grid on
hold off

6-158
 Design Broadband Matching Networks for Amplifier

The plot shows that the target requirement for both gain and noise figure have been met. To
understand the effect of optimizing with respect to only the transducer gain, use the first choice for
the cost function. The cost function involves only the gain term within the objective function.

Display Optimized Element Values

Display the optimized inductor and capacitor values for the input matching network.
Lin_Optimized = LC_Optimized(1:3)

Lin_Optimized = 3×1
10-7 ×

0.5722
0.9272
0.3546

Cin_Optimized = LC_Optimized(4)

Cin_Optimized =
6.8526e-12

Display the optimized inductor and capacitor values for the output matching network
Lout_Optimized = LC_Optimized(5:7)

Lout_Optimized = 3×1
10-6 ×

6-159
6 RF Toolbox Examples

0.0517
0.1275
0.0581

Cout_Optimized = LC_Optimized(8)

Cout_Optimized =
5.4408e-12

Matching Network with Real-world Lumped Component Models

The real-world lumped components are selected from the Modelithics Select+ Library [5]. You must
have the Modelithics Select+ Library license to run the following codes. Note that the Modelithics®
LC elements are scaled with respect to the substrate characteristics. This example shows one
possible approach using a 59 mil FR4 substrate.

Create RF Component from Modelithics SELECT+ Library

Set up the Modelithics Select+ Library by specifying the full path to the library.

mdlxSetup('C:\mdlx_library\SELECT')

Create the Modelithics® library object.

mdlx = mdlxLibrary;

Search the library for inductors with the optimized value for the input matching network.

search(mdlx,'FR4Standard59mil',Type='Inductors',Value=Lin_Optimized(1))

6-160
 Design Broadband Matching Networks for Amplifier

Search the library for capacitors with the optimized value for the input matching network.

search(mdlx,'FR4Standard59mil',Type='Capacitors',Value=Cin_Optimized)

6-161
6 RF Toolbox Examples

Create an array of Modelithics® LC elements for the input and output matching networks.

Lin1_mdlx = search(mdlx,'FR4Standard59mil',Type='Inductors',Value=Lin_Optimized(1));
Lin2_mdlx = search(mdlx,'FR4Standard59mil',Type='Inductors',Value=Lin_Optimized(2));
Lin3_mdlx = search(mdlx,'FR4Standard59mil',Type='Inductors',Value=Lin_Optimized(3));
Cin_mdlx = search(mdlx,'FR4Standard59mil',Type='Capacitors',Value=Cin_Optimized);

Lout1_mdlx = search(mdlx,'FR4Standard59mil',Type='Inductors',Value=Lout_Optimized(1));
Lout2_mdlx = search(mdlx,'FR4Standard59mil',Type='Inductors',Value=Lout_Optimized(2));
Lout3_mdlx = search(mdlx,'FR4Standard59mil',Type='Inductors',Value=Lout_Optimized(3));
Cout_mdlx = search(mdlx,'FR4Standard59mil',Type='Capacitors',Value=Cout_Optimized);

Create and Anlyze Matching Networks

Create the input and output matching networks with Modelithics® capacitors and inductors. Note
that most of the Modelithics® lumped components have two ports. Use the setports function to
define the circuit as a 2-port network.

InputMatchingNWMdlx = circuit;
add(InputMatchingNWMdlx,[1 2 0 0],Lin1_mdlx(1));
add(InputMatchingNWMdlx,[2 0 0 0],Lin2_mdlx(1));
add(InputMatchingNWMdlx,[2 0 0 0],Cin_mdlx(1));
add(InputMatchingNWMdlx,[2 3 0 0],Lin3_mdlx(1));
setports(InputMatchingNWMdlx,[1 0],[3 0]);

6-162
 Design Broadband Matching Networks for Amplifier

OutputMatchingNWMdlx = circuit;
add(OutputMatchingNWMdlx,[1 2 0 0],Lout1_mdlx(1));
add(OutputMatchingNWMdlx,[2 0 0 0],Lout2_mdlx(1));
add(OutputMatchingNWMdlx,[2 0 0 0],Cout_mdlx(1));
add(OutputMatchingNWMdlx,[2 3 0 0],Lout3_mdlx(1));
setports(OutputMatchingNWMdlx,[1 0],[3 0]);

Analyze the S-parameters of input and output matching networks and then convert the S-parameter
data into an nport object.

sm1 = sparameters(InputMatchingNWMdlx,freq);
npm1 = nport(sm1);

sm2 = sparameters(OutputMatchingNWMdlx,freq);
npm2 = nport(sm2);

Use the rfbudget object to calculate the transducer gain and the noise figure across the bandwidth.
Compute RF budget of the LNA network. The LNA network comprises of matching network and low
noise amplifier.

bMatchedMdlx = rfbudget([npm1 clone(Mismatched_Amp) npm2],freq',-30,BW);

Verify Design

Plot the transducer gain and the noise figure of the real-world matching network and then compare
the optimization results with the mismatched network.

figure
plot(freq*1e-6,bMatchedMdlx.TransducerGain(:,end))
hold on
plot(freq*1e-6,bMatched.TransducerGain(:,end))
plot(freq*1e-6,bMismatched.TransducerGain(:,end))
plot(freq*1e-6,bMatchedMdlx.NF(:,end))
plot(freq*1e-6,bMatched.NF(:,end))
plot(freq*1e-6,bMismatched.NF(:,end))
axis([freq(1)*1e-6 freq(end)*1e-6 0 12])
legend('G_t - Matched w/ Modelithics®','G_t - Matched w/ ideal',...
'G_t - Mismatched','NF - Matched w/ Modelithics®','NF - Matched w/ ideal',...
'NF - Mismatched','Position', [0.22643,0.47762,0.49643,0.25119])
xlabel('Frequency [MHz]')
ylabel('Magnitude [dB]')
grid on
hold off

6-163
6 RF Toolbox Examples

The most noticeable aspect of the matching performance when using the Modelithics® library is the
large deviation in transducer gain especially beyond 275 MHz. The noise figure performance has also
worsened by approximately 0.25 dB. This can be explained by the fact that the inductor and capacitor
models now represent real-world parasitic effects such as those due to the presence of a substrate,
pads and orientation. This analysis highlights the importance of being able to predict such behaviors
prior to realization of the design in hardware.

References

[1] Ludwig, Reinhold, and Gene Bogdanov. RF Circuit Design: Theory and Applications. Upper Saddle
River, NJ: Prentice-Hall, 2009.

[2] Cuthbert, Thomas R. Broadband Direct-Coupled and Matching RF Networks. Greenwood, Ark.:
T.R. Cuthbert, 1999.

[3] Cuthbert, T.R. “A Real Frequency Technique Optimizing Broadband Equalizer Elements.” In 2000
IEEE International Symposium on Circuits and Systems. Emerging Technologies for the 21st Century.
Proceedings (IEEE Cat No.00CH36353), 5:401–4. Geneva, Switzerland: Presses Polytech. Univ.
Romandes, 2000. https://doi.org/10.1109/ISCAS.2000.857453.

[4] Pozar, David M. Microwave Engineering. 4th ed. Hoboken, NJ: Wiley, 2012.

6-164
 Design Broadband Matching Networks for Amplifier

[5] https://www.modelithics.com/model/freemodels

See Also

More About
• “Design Broadband Matching Networks for Antennas” on page 6-139
• “Impedance Matching of Small Monopole Antenna” on page 6-166

6-165
6 RF Toolbox Examples

Impedance Matching of Small Monopole Antenna

This example shows how to design a double tuning L-section matching network between a resistive
source and capacitive load in the form of a small monopole by using the matchingnetwork object.
The L-section consists of two inductors. The network achieves conjugate match and guarantees
maximum power transfer at a single frequency.

Create Monopole

Create a quarter-wavelength monopole antenna via the Antenna Toolbox with the resonant frequency
around 1 GHz. For the purpose of this example, we choose a square ground plane of side 0 . 75λ.

fres = 1e9;
speedOfLight = physconst('lightspeed');
lambda = speedOfLight/fres;
L = 0.25*lambda;
dp = monopole('Height',L,'Width',L/50,...
'GroundPlaneLength',0.75*lambda,...
'GroundPlaneWidth',0.75*lambda);

Calculate Monopole Impedance

Specify the source (generator) impedance, the reference (transmission line) impedance and the load
(antenna) impedance. In this example, the load Zl0 will be the non-resonant (small) monopole at the
frequency of 500 MHz, which is the half of the resonant frequency. The source has the equivalent
impedance of 50 ohms.

f0 = fres/2;
Zs = 50;
Z0 = 50;
Zl0 = impedance(dp,f0);
Rl0 = real(Zl0);
Xl0 = imag(Zl0);

Define the number of frequency points for the analysis and set up a frequency band about 500 MHz.

Npts = 30;
fspan = 0.1;
fmin = f0*(1 - (fspan/2));
fmax = f0*(1 + (fspan/2));
freq = unique([f0 linspace(fmin,fmax,Npts)]);

Understand Load Behavior using Reflection Coefficient and Power Gain

Calculate the load reflection coefficient and the power gain between the source and the antenna.

S = sparameters(dp, freq);
GammaL = rfparam(S, 1,1);
Gt = 10*log10(1 - abs(GammaL).^2);

Plotting the input reflection coefficient on a Smith chart shows the capacitive behavior of this
antenna around the operating frequency of 500 MHz. The center of the Smith chart represents the
matched condition to the reference impedance. The location of the reflection coefficient trace around
− j5 . 0 Ω confirms that there is a severe impedance mismatch.

6-166
 Impedance Matching of Small Monopole Antenna

fig1 = figure;
hsm = smithplot(fig1,freq,GammaL,'LineWidth',2.0,'Color','m',...
'View','bottom-right','LegendLabels',{'#Gamma L'});

Plot the power delivered to the load.

fig2 = figure;
plot(freq*1e-6,Gt,'m','LineWidth',2);
grid on
xlabel('Frequency [MHz]')
ylabel('Magnitude (dB)')
title('Power delivered to load')

6-167
6 RF Toolbox Examples

As the power gain plot shows, there is approximately 20 dB power loss around the operating
frequency (500 MHz).

Design Matching Network

The matching network must ensure maximum power transfer at 500 MHz. The L-section double
tuning network achieves this goal [1]. The network topology, shown in the figure consists of an
inductor in series with the antenna, that cancels the large capacitance at 500 MHz, and a shunt
inductor that further boosts the output resistance to match the source impedance of 50 Ω.

Use the matchingnetwork object to create various matching network circuits based on the source
impedance, load impedance, and center frequency.

6-168
 Impedance Matching of Small Monopole Antenna

matchnw = matchingnetwork('CenterFrequency',f0,'LoadImpedance',Zl0,'Bandwidth',50e6);
matchnw.clearEvaluationParameter(1); % clear default constraint

The values of each element for the circuits generated are shown below.

[circuit_list, performance] = circuitDescriptions(matchnw)

circuit_list=4×5 table
circuitName component1Type component1Value component2Type component2
___________ ______________ _______________ ______________ __________

Circuit 1 "auto_1" "Series L" 2.9835e-07 "Shunt L" 1.0551e-

Circuit 2 "auto_2" "Series C" 3.396e-13 "Shunt L" 6.1873e-
Circuit 3 "auto_3" "Shunt C" 2.3541e-11 "Series L" 8.1999e-
Circuit 4 "auto_4" "Shunt L" 4.304e-09 "Series L" 7.3978e-

performance=4×4 table
circuitName evaluationPassed testsFailed performanceScore
___________ ________________ ____________ ________________

Circuit 1 "auto_1" {["Yes"]} {0x0 double} {[0]}

Circuit 2 "auto_2" {["Yes"]} {0x0 double} {[0]}
Circuit 3 "auto_3" {["Yes"]} {0x0 double} {[0]}
Circuit 4 "auto_4" {["Yes"]} {0x0 double} {[0]}

Create Matching Network and Calculate S-parameters

The matching network circuit is created via the RF Toolbox™ and it consists of the two inductors
whose values have been calculated above. The S-parameters of this network are calculated over the
frequency band centered at the operating frequency.

Calculate the S-parameters based on desired matching network (this example uses circuit #2).

ckt_no = 4;
Smatchnw = sparameters(matchnw, freq, Z0, ckt_no);

The circuit element representation of the matching network is shown below.

disp(matchnw.Circuit(ckt_no))

circuit: Circuit element

ElementNames: {'L' 'L_1'}

Elements: [1x2 inductor]
Nodes: [1 2 3]
Name: 'auto_4'

Reflection Coefficient and Power Gain with Matching Network

Calculate the input reflection coefficient/power gain for the antenna load with the matching network.

Zl = impedance(dp,freq);
GammaIn = gammain(Smatchnw,Zl);
Gtmatch = powergain(Smatchnw,Zs,Zl,'Gt');
Gtmatch = 10*log10(Gtmatch);

6-169
6 RF Toolbox Examples

Compare Results

Plot the input reflection coefficient and power delivered to the antenna, with and without the
matching network. The Smith® chart plot shows the reflection coefficient trace going through its
center thus confirming the match. At the operation frequency of 500 MHz, the generator transfers
maximum power to the antenna. The match degrades on either side of the operating frequency.

add(hsm,freq,GammaIn);
hsm.LegendLabels(2) = {'#Gamma In'};
hsm.View = 'full';

Plot power delivered to the load.

figure(fig2)
hold on
plot(freq*1e-6,Gtmatch,'LineWidth',2);
axis([min(freq)*1e-6,max(freq)*1e-6,-25,0])
legend('No matching network','Double tuning','Location','Best');

6-170
 Impedance Matching of Small Monopole Antenna

Matching Network Designer

The Matching Network Designer app allows to design matching networks or view an existing
matchingnetwork object. Type this command at the command line to open the Matching Network
Designer app. Use the matchnw object and select auto_4 node to view the corresponding circuit.

matchingNetworkDesigner(matchnw)

6-171
6 RF Toolbox Examples

References

[1] M. M. Weiner, Monopole Antennas, Marcel Dekker, Inc.,CRC Press, Rev. Exp edition, New York,
pp.110-118, 2003.

See Also

More About
• “Design Broadband Matching Networks for Antennas” on page 6-139
• “Design Broadband Matching Networks for Amplifier” on page 6-148

6-172
 Operations with RF Circuit Objects

Operations with RF Circuit Objects

This example shows how to create and use RF Toolbox™ circuit objects. In this example, you create
three circuit (rfckt) objects: two transmission lines and an amplifier.

You visualize the amplifier data using RF Toolbox™ functions and retrieve the frequency data that
was read from a file into the amplifier rfckt object. Then you analyze the amplifier over a different
frequency range and visualize the results. Next, you cascade the three circuits, analyze the cascaded
network and visualize its S-parameters over the original frequency range of the amplifier. Finally, you
plot the S11, S22, and S21 parameters, and noise figure of the cascaded network.

Create rfckt Objects

Create three circuit objects: two transmission lines, and an amplifier using data from default.amp
data file.
FirstCkt = rfckt.txline;
SecondCkt = rfckt.amplifier('IntpType','cubic');
read(SecondCkt,'default.amp');
ThirdCkt = rfckt.txline('LineLength',0.025,'PV',2.0e8);

View Properties of rfckt Objects

You can use the get function to view an object's properties. For example,
PropertiesOfFirstCkt = get(FirstCkt)

PropertiesOfFirstCkt = struct with fields:

LineLength: 0.0100
StubMode: 'NotAStub'
Termination: 'NotApplicable'
Freq: 1.0000e+09
Z0: 50.0000 + 0.0000i
PV: 299792458
Loss: 0
IntpType: 'Linear'
nPort: 2
AnalyzedResult: []
Name: 'Transmission Line'

PropertiesOfSecondCkt = get(SecondCkt)

PropertiesOfSecondCkt = struct with fields:

NoiseData: [1x1 rfdata.noise]
NonlinearData: [1x1 rfdata.power]
IntpType: 'Cubic'
NetworkData: [1x1 rfdata.network]
nPort: 2
AnalyzedResult: [1x1 rfdata.data]
Name: 'Amplifier'

PropertiesOfThirdCkt = get(ThirdCkt)

PropertiesOfThirdCkt = struct with fields:

LineLength: 0.0250

6-173
6 RF Toolbox Examples

StubMode: 'NotAStub'
Termination: 'NotApplicable'
Freq: 1.0000e+09
Z0: 50.0000 + 0.0000i
PV: 200000000
Loss: 0
IntpType: 'Linear'
nPort: 2
AnalyzedResult: []
Name: 'Transmission Line'

List Methods of rfckt Objects

You can use the methods function to list an object's methods. For example,
MethodsOfThirdCkt = methods(ThirdCkt)

MethodsOfThirdCkt = 82x1 cell

{'addlistener' }
{'analyze' }
{'calcgroupdelay' }
{'calckl' }
{'calcpout' }
{'calculate' }
{'calczin' }
{'checkbool' }
{'checkchar' }
{'checkenum' }
{'checkenumexact' }
{'checkfrequency' }
{'checkproperty' }
{'checkproptype' }
{'checkreadonlyproperty'}
{'checkrealscalardouble'}
{'circle' }
{'convertfreq' }
{'copy' }
{'delete' }
{'destroy' }
{'disp' }
{'eq' }
{'extract' }
{'findimpedance' }
{'findobj' }
{'findprop' }
{'ge' }
{'get' }
{'getdata' }
⋮

Change Properties of rfckt Objects

Use the get function or Dot Notation to get the line length of the first transmission line.
DefaultLength = FirstCkt.LineLength;

Use the set function or Dot notation to change the line length of the first transmission line.

6-174
 Operations with RF Circuit Objects

FirstCkt.LineLength = .001;
NewLength = FirstCkt.LineLength;

Plot Amplifier S11 and S22 Parameters

Use the smithplot method of circuit object to plot the original S11 and S22 parameters of the
amplifier (SecondCkt) on a Z Smith chart. The original frequencies of the amplifier's S-parameters
range from 1.0 GHz to 2.9 GHz.

figure
smithplot(SecondCkt,[1 1;2 2]);

Plot Amplifier Pin-Pout Data

Use the plot method of circuit object to plot the amplifier (SecondCkt) Pin-Pout data, in dBm, at 2.1
GHz on an X-Y plane.

figure
plot(SecondCkt,'Pout','dBm')

6-175
6 RF Toolbox Examples

legend('show','Location','northwest');

Get Original Frequency Data and Result of Analyzing Amplifier over Original Frequencies

When the RF Toolbox reads data from default.amp into an amplifier object (SecondCkt), it also
analyzes the amplifier over the frequencies of network parameters in default.amp file and store the
result at the property AnalyzedResult. Here are the original amplifier frequency and analyzed
result over it.

f = SecondCkt.AnalyzedResult.Freq;
data = SecondCkt.AnalyzedResult

data =
rfdata.data with properties:

Freq: [191x1 double]

S_Parameters: [2x2x191 double]
GroupDelay: [191x1 double]
NF: [191x1 double]
OIP3: [191x1 double]
Z0: 50.0000 + 0.0000i
ZS: 50.0000 + 0.0000i
ZL: 50.0000 + 0.0000i
IntpType: 'Cubic'
Name: 'Data object'

6-176
 Operations with RF Circuit Objects

Analyze and Plot S11 and S22 of Amplifier Circuit with Different Frequencies

To visualize the S-parameters of a circuit over a different frequency range, you must first analyze it
over the specified frequency range.

analyze(SecondCkt,1.85e9:1e7:2.55e9);
smithplot(SecondCkt,[1 1;2 2],'GridType','ZY')

Create and Analyze Cascaded rfckt Object

Cascade three circuit objects to create a cascaded circuit object, and then analyze it at the original
amplifier frequencies which range from 1.0 GHz to 2.9 GHz.

CascadedCkt = rfckt.cascade('Ckts',{FirstCkt,SecondCkt,ThirdCkt});
analyze(CascadedCkt,f)

ans =
rfckt.cascade with properties:

Ckts: {[1x1 rfckt.txline] [1x1 rfckt.amplifier] [1x1 rfckt.txline]}

nPort: 2
AnalyzedResult: [1x1 rfdata.data]
Name: 'Cascaded Network'

6-177
6 RF Toolbox Examples

Figure 1: The cascaded circuit.

Plot S11 and S22 Parameters of Cascaded Circuit

Use the smithplot method of circuit object to plot S11 and S22 of the cascaded circuit
(CascadedCkt) on a Z Smith chart.

smithplot(CascadedCkt,[1 1;2 2],'GridType','Z')

Plot S21 Parameters of Cascaded Circuit

Use the plot method of circuit object to plot S21 of the cascaded circuit (CascadedCkt) on an X-Y
plane.

plot(CascadedCkt,'S21','dB')
legend show;

Plot Budget S21 Parameters and Noise Figure of Cascaded Circuit

Use the plot method of circuit object to plot the budget S21 parameters and noise figure of the
cascaded circuit (CascadedCkt) on an X-Y plane.

plot(CascadedCkt,'budget','S21','NF')
legend show;

See Also

More About
• “Bandpass Filter Response Using RFCKT Objects” on page 6-33

6-178
 Operations with RF Data Objects

Operations with RF Data Objects

This example shows you how to manipulate RF data directly using rfdata objects. First, you create
an rfdata.data object by reading in the S-parameters of a two-port passive network stored in the
Touchstone® format data file, passive.s2p. Next, you create a circuit object, rfckt.amplifier,
and you update the properties of this object using three data objects.

Read Touchstone® Data File

Use the read method of the rfdata.data object to read the Touchstone data file passive.s2p.
The parameters in this data file are the 50-Ohm S-parameters of a 2-port passive network at
frequencies ranging from 315 kHz to 6.0 GHz.

data = rfdata.data;
data = read(data,'passive.s2p')

data =
rfdata.data with properties:

Freq: [202x1 double]

S_Parameters: [2x2x202 double]
GroupDelay: [202x1 double]
NF: [202x1 double]
OIP3: [202x1 double]
Z0: 50.0000 + 0.0000i
ZS: 50.0000 + 0.0000i
ZL: 50.0000 + 0.0000i
IntpType: 'Linear'
Name: 'Data object'

Use the extract method of the rfdata.data object to get other network parameters. For example,
here are the frequencies, 75-Ohm S-parameters, and Y-parameters which are converted from the
original 50-Ohm S-parameters in passive.s2p data file.

[s_params,freq] = extract(data,'S_PARAMETERS',75);
y_params = extract(data,'Y_PARAMETERS');

Use the RF utility function, smithplot to plot the 75-Ohm S11 on a Smith chart.

s11 = s_params(1,1,:);
figure
smithplot(freq, s11(:))

6-179
6 RF Toolbox Examples

Here are the four 75-Ohm S-parameters and four Y-parameters at 6.0 GHz, the last frequency.
f = freq(end)

f =
6.0000e+09

s = s_params(:,:,end)

s = 2×2 complex

-0.0764 - 0.5401i 0.6087 - 0.3018i

0.6094 - 0.3020i -0.1211 - 0.5223i

y = y_params(:,:,end)

y = 2×2 complex

0.0210 + 0.0252i -0.0215 - 0.0184i

-0.0215 - 0.0185i 0.0224 + 0.0266i

Create RF Data Objects for Amplifier with Your Own Data

In this example, you create a circuit object, rfckt.amplifier. Then you create three data objects
and use them to update the properties of the circuit object.

The rfckt.amplifier object has properties for network parameters, noise data and nonlinear data:

6-180
 Operations with RF Data Objects

• NetworkData is an rfdata.network object for network parameters.

• NoiseData is for noise parameters which could be a scalar NF (dB), an rfdata.noise, or an
rfdata.nf object.
• NonlinearData is for nonlinear parameters which could be a scalar OIP3 (dBm), an
rfdata.power, or an rfdata.ip3 object.

By default, these properties of rfckt.amplifier contain data from the default.amp data file.
NetworkData is an rfdata.network object that contains 50-Ohm 2-port S-Parameters at 191
frequencies ranging from 1.0 GHz to 2.9 GHz. NoiseData is an rfdata.noise object that contains
spot noise data at 9 frequencies ranging from 1.9 GHz to 2.48 GHz. The NonlinearData parameter
is an rfdata.power object that contains Pin/Pout data at 2.1 GHz.
amp = rfckt.amplifier

amp =
rfckt.amplifier with properties:

NoiseData: [1x1 rfdata.noise]

NonlinearData: [1x1 rfdata.power]
IntpType: 'Linear'
NetworkData: [1x1 rfdata.network]
nPort: 2
AnalyzedResult: [1x1 rfdata.data]
Name: 'Amplifier'

Use the following code to create an rfdata.network object that contains 2-port Y-parameters of an
amplifier at 2.08 GHz, 2.10 GHz and 2.15 GHz. Later in this example, you use this data object to
update the NetworkData property of the amplifier object.
f = [2.08 2.10 2.15] * 1.0e9;
y(:,:,1) = [-.0090-.0104i, .0013+.0018i; -.2947+.2961i, .0252+.0075i];
y(:,:,2) = [-.0086-.0047i, .0014+.0019i; -.3047+.3083i, .0251+.0086i];
y(:,:,3) = [-.0051+.0130i, .0017+.0020i; -.3335+.3861i, .0282+.0110i];
netdata = rfdata.network('Type','Y_PARAMETERS','Freq',f,'Data',y)

netdata =
rfdata.network with properties:

Type: 'Y_PARAMETERS'
Freq: [3x1 double]
Data: [2x2x3 double]
Z0: 50.0000 + 0.0000i
Name: 'Network parameters'

Use the following code to create an rfdata.nf object that contains noise figures of the amplifier, in
dB, at seven frequencies ranging from 1.93 GHz to 2.40 GHz. Later in this example, you use this data
object to update the NoiseData property of the amplifier object.
f = [1.93 2.06 2.08 2.10 2.15 2.3 2.4] * 1.0e+009;
nf = [12.4521 13.2466 13.6853 14.0612 13.4111 12.9499 13.3244];
nfdata = rfdata.nf('Freq',f,'Data',nf)

nfdata =
rfdata.nf with properties:

6-181
6 RF Toolbox Examples

Freq: [7x1 double]

Data: [7x1 double]
Name: 'Noise figure'

Use the following code to create an rfdata.ip3 object that contains the output third-order intercept
points of the amplifier, which is 8.45 watts at 2.1 GHz. Later in this example, you use this data object
to update the NonlinearData property of the amplifier object.

ip3data = rfdata.ip3('Type','OIP3','Freq',2.1e9,'Data',8.45)

ip3data =
rfdata.ip3 with properties:

Type: 'OIP3'
Freq: 2.1000e+09
Data: 8.4500
Name: '3rd order intercept'

Use the following code to update the properties of the amplifier object with three data objects you
created in the previous steps. To get a good amplifier object, the data in these data objects must be
accurate. These data could be obtained from RF measurements, or circuit simulation using other
tools.

amp.NetworkData = netdata;
amp.NoiseData = nfdata;
amp.NonlinearData = ip3data

amp =
rfckt.amplifier with properties:

NoiseData: [1x1 rfdata.nf]

NonlinearData: [1x1 rfdata.ip3]
IntpType: 'Linear'
NetworkData: [1x1 rfdata.network]
nPort: 2
AnalyzedResult: [1x1 rfdata.data]
Name: 'Amplifier'

See Also

More About
• “Create RF Objects” on page 2-2

6-182
 Design IF Butterworth Bandpass Filter

Design IF Butterworth Bandpass Filter

This example shows how to design an Intermediate Frequency (IF) Butterworth bandpass filter with a
center frequency of 400 MHz, bandwidth of 5 MHz, and Insertion Loss (IL) of 1dB [1] on page 6-186.

Account for Mismatch/Insertion Loss (IL)

Practical circuits suffer a certain degree of mismatch. Mismatch happens when an unmatched circuit
is connected to an RF source leading to reflections that result in a loss of power delivered to the
circuit. You can use IL to define this mismatch. Calculate the load impedance mismatch to account for
the given IL. The IL and normalized load impedance (ZL) are related as follows [2] on page 6-186,[3]
on page 6-186:

IL (dB) = -10*log10(1-| γin |^2) = -10*log10(4*ZL/(1+ZL)^2)

The roots of the resulting polynomial return the value of normalized load impedance. The
unnormalized values are 132.986 Ohms and 18.799 Ohms. Choose the higher value for the filter
design to account for the IL.

syms ZL IL
eqn = -10*log10(4*ZL/(1+ZL)^2) - IL == 0;
[solx, ~, ~] = solve(eqn,ZL,'ReturnConditions', true);
IL_desired_dB = 1;
Zload = double(subs(solx,IL,IL_desired_dB))*50;

Load impedance:

ZL = Zload(2);

Design Filter

Use rffilter to design the filter for the desired specifications.

Fcenter = 400e6;
Bwpass = 5e6;
if_filter = rffilter('ResponseType','Bandpass',...
'FilterType','Butterworth','FilterOrder',4,...
'PassbandAttenuation',10*log10(2),...
'Implementation','Transfer function',...
'PassbandFrequency',[Fcenter-Bwpass/2 Fcenter+Bwpass/2],'Zout',ZL);

Plot S-parameters and Group Delay of Filter

Calculate S-parameters.

freq = linspace(370e6,410e6,2001);
Sf = sparameters(if_filter, freq);
figure;
line = rfplot(Sf);
lgd = legend;
lgd.Location = "best";
[~,freq_index] = min(abs(freq-Fcenter));
datatip(line(3),'DataIndex',freq_index);

6-183
6 RF Toolbox Examples

A datatip shows a 1dB IL at Fcenter = 400 MHz.

Calculate groupdelay:

gd = groupdelay(if_filter, freq);
figure;
plot(freq/1e6, gd);
xlabel('Frequency (MHz)');
ylabel('Group delay (s)');
grid on;

6-184
 Design IF Butterworth Bandpass Filter

Insert Filter into rfbudget Object

An rffilter object can be inserted directly into an rfbudget object to perform budget analysis.

rfb = rfbudget(if_filter,Fcenter,-30,Bwpass)

rfb =
rfbudget with properties:

Elements: [1x1 rffilter]

InputFrequency: 400 MHz
AvailableInputPower: -30 dBm
SignalBandwidth: 5 MHz
Solver: Friis
AutoUpdate: true

Analysis Results
OutputFrequency: 400 (MHz)
OutputPower: -30 (dBm)
TransducerGain: -0.0004646 (dB)
NF: 0 (dB)
IIP2: [] (dBm)
OIP2: [] (dBm)
IIP3: Inf (dBm)
OIP3: Inf (dBm)
SNR: 76.99 (dB)

6-185
6 RF Toolbox Examples

References

[1] Hongbao Zhou, Bin Luo. " Design and budget analysis of RF receiver of 5.8GHz ETC reader"
Published at Communication Technology (ICCT), 2010 12th IEEE International Conference, Nanjing,
China, November 2010.

[2] Electronic Filter Analysis and Synthesis, Michael G. Ellis, Sr., Artech House, Chapter 7.

[3] RF Circuit Design, R. Ludwig, G. Bogdanov, Pearson Education, Chapter 2.

See Also

“Superheterodyne Receiver Using RF Budget Analyzer App” on page 6-2

See Also

More About
• “Bandpass Filter Response” on page 6-21

6-186
 Test, Visualize, and Enforce Passivity of Rational Output

Test, Visualize, and Enforce Passivity of Rational Output

This example shows how to test, visualize, and enforce the passivity of output from the rational
function.

S-Parameter Data Passivity

Time-domain analysis and simulation depend critically on the ability to convert frequency-domain S-
parameter data into causal, stable, and passive time-domain representations. Because the rational
function guarantees that all poles are in the left half plane, rational output is both stable and
causal by construction. The problem is passivity.

N-port S-parameter data represents a frequency-dependent transfer function H(f). You can create an
S-parameters object in RF Toolbox™ by reading a Touchstone® file, such as passive.s2p, into the
sparameters function.

You can use the ispassive function to check the passivity of the S-parameter data, and the
passivity function to plot the 2-norm of the N-by-N matrices H(f) at each data frequency.

S = sparameters('passive.s2p');
ispassive(S)

ans = logical
1

passivity(S)

6-187
6 RF Toolbox Examples

Testing and Visualizing rational Output Passivity

The rational function converts N-port S-parameter data, S, into an object that represents a rational
fit to the data. Using the ispassive function on the output reports that even if input data S is
passive, the output fit is not passive. In other words, the norm H(f) is greater than 1 at some
frequency in the range [0,Inf].

The passivity function takes the output of the rational function as input and plots its passivity.
This is a plot of the upper bound of the norm(H(f)) on [0,Inf], also known as the H-infinity norm.

fit = rational(S);
ispassive(fit)

ans = logical
0

passivity(fit)

6-188
 Test, Visualize, and Enforce Passivity of Rational Output

The makepassive function takes as input a rational object, and produces a passive fit by using
convex optimization techniques to optimally match the data of the S-parameter input S while
satisfying passivity constraints. The residues and direct term of the output pfit are modified, but the
poles of the output pfit are identical to the poles of the input fit.

pfit = makepassive(fit,'Display','on');

Iter H-infinity norm Frequency Error (dB) Constraints

0 1+7.15e-03 516 MHz -40.4597
1 1+5.44e-04 319 MHz -41.7549 5
2 1+7.83e-05 383 MHz -41.7318 7
3 1+8.26e-06 360 MHz -41.7275 9
4 1-8.79e-07 359 MHz -41.7271 10

ispassive(pfit)

ans = logical
1

passivity(pfit)

6-189
6 RF Toolbox Examples

all(pfit.Poles == fit.Poles)

ans = logical
1

Generate Equivalent SPICE Circuit from Passive Fit

The generateSPICE function takes a passive fit and generates an equivalent circuit as a SPICE
subcircuit file. The input fit is an object as returned by rational with an S-parameters object as
input. The generated file is a SPICE model constructed solely of passive R, L, C elements and
controlled source elements E, F, G, and H.
generateSPICE(pfit,'mypassive.ckt')
type mypassive.ckt

* Equivalent circuit model for mypassive.ckt

.SUBCKT mypassive po1 po2
Vsp1 po1 p1 0
Vsr1 p1 pr1 0
Rp1 pr1 0 50
Ru1 u1 0 50
Fr1 u1 0 Vsr1 -1
Fu1 u1 0 Vsp1 -1
Ry1 y1 0 1
Gy1 p1 0 y1 0 -0.02
Vsp2 po2 p2 0
Vsr2 p2 pr2 0

6-190
 Test, Visualize, and Enforce Passivity of Rational Output

Rp2 pr2 0 50
Ru2 u2 0 50
Fr2 u2 0 Vsr2 -1
Fu2 u2 0 Vsp2 -1
Ry2 y2 0 1
Gy2 p2 0 y2 0 -0.02
Rx1 x1 0 1
Fxc1_2 x1 0 Vx2 1.73989860493562
Cx1 x1 xm1 4.40797245907321e-09
Vx1 xm1 0 0
Gx1_1 x1 0 u1 0 -0.0637703326841126
Rx2 x2 0 1
Fxc2_1 x2 0 Vx1 -1.1043097663947
Cx2 x2 xm2 4.40797245907321e-09
Vx2 xm2 0 0
Gx2_1 x2 0 u1 0 0.0704222011893046
Rx3 x3 0 1
Cx3 x3 0 2.46549209700263e-12
Gx3_1 x3 0 u1 0 -0.857223829748846
Rx4 x4 0 1
Cx4 x4 0 1.19630317726907e-11
Gx4_1 x4 0 u1 0 -2.76943895164145
Rx5 x5 0 1
Cx5 x5 0 1.85575919869105e-11
Gx5_1 x5 0 u1 0 -1.31126215630122
Rx6 x6 0 1
Cx6 x6 0 8.71015574884511e-11
Gx6_1 x6 0 u1 0 -0.691517548083117
Rx7 x7 0 1
Cx7 x7 0 5.76402178229167e-10
Gx7_1 x7 0 u1 0 -0.0721110184360492
Rx8 x8 0 1
Cx8 x8 0 1.32870599809626e-08
Gx8_1 x8 0 u1 0 -0.853432841202091
Rx9 x9 0 1
Fxc9_10 x9 0 Vx10 1.74905983790304
Cx9 x9 xm9 4.40797245907321e-09
Vx9 xm9 0 0
Gx9_2 x9 0 u2 0 -0.0646357714417864
Rx10 x10 0 1
Fxc10_9 x10 0 Vx9 -1.09852560806066
Cx10 x10 xm10 4.40797245907321e-09
Vx10 xm10 0 0
Gx10_2 x10 0 u2 0 0.0710040501255583
Rx11 x11 0 1
Cx11 x11 0 2.46549209700263e-12
Gx11_2 x11 0 u2 0 -0.903067046326257
Rx12 x12 0 1
Cx12 x12 0 1.19630317726907e-11
Gx12_2 x12 0 u2 0 -2.78921388996219
Rx13 x13 0 1
Cx13 x13 0 1.85575919869105e-11
Gx13_2 x13 0 u2 0 -1.32144704264833
Rx14 x14 0 1
Cx14 x14 0 8.71015574884511e-11
Gx14_2 x14 0 u2 0 -0.731804282602902
Rx15 x15 0 1
Cx15 x15 0 5.76402178229167e-10

6-191
6 RF Toolbox Examples

Gx15_2 x15 0 u2 0 -0.0762129482984789

Rx16 x16 0 1
Cx16 x16 0 1.32870599809626e-08
Gx16_2 x16 0 u2 0 -0.853727128838507
Gyc1_1 y1 0 x1 0 -1
Gyc1_2 y1 0 x2 0 -1
Gyc1_3 y1 0 x3 0 -1
Gyc1_4 y1 0 x4 0 0.133829740739509
Gyc1_5 y1 0 x5 0 -0.577026114870956
Gyc1_6 y1 0 x6 0 -1
Gyc1_7 y1 0 x7 0 1
Gyc1_8 y1 0 x8 0 0.998953434819797
Gyc1_9 y1 0 x9 0 0.981489576611876
Gyc1_10 y1 0 x10 0 0.985665919902024
Gyc1_11 y1 0 x11 0 0.820830479554872
Gyc1_12 y1 0 x12 0 -1
Gyc1_13 y1 0 x13 0 1
Gyc1_14 y1 0 x14 0 0.920286730209121
Gyc1_15 y1 0 x15 0 -0.916755172898503
Gyc1_16 y1 0 x16 0 -1
Gyd1_1 y1 0 u1 0 0.999378046395583
Gyd1_2 y1 0 u2 0 -0.0123723629900616
Gyc2_1 y2 0 x1 0 0.992951774589524
Gyc2_2 y2 0 x2 0 0.990662123377927
Gyc2_3 y2 0 x3 0 0.823600342834595
Gyc2_4 y2 0 x4 0 -1
Gyc2_5 y2 0 x5 0 1
Gyc2_6 y2 0 x6 0 0.974339502347319
Gyc2_7 y2 0 x7 0 -0.968877586361953
Gyc2_8 y2 0 x8 0 -1
Gyc2_9 y2 0 x9 0 -1
Gyc2_10 y2 0 x10 0 -1
Gyc2_11 y2 0 x11 0 -1
Gyc2_12 y2 0 x12 0 0.207630507657774
Gyc2_13 y2 0 x13 0 -0.66808999878785
Gyc2_14 y2 0 x14 0 -1
Gyc2_15 y2 0 x15 0 1
Gyc2_16 y2 0 x16 0 0.998717056545118
Gyd2_1 y2 0 u1 0 0.0128984144008306
Gyd2_2 y2 0 u2 0 0.999493034617814
.ENDS

See Also

More About
• “Using 'NPoles' Parameter With rationalfit” on page 6-71
• “Using 'Weight' Parameter With rationalfit” on page 6-75
• “Improve Quality of Fit Using 'DelayFactor' Parameter” on page 6-81

6-192
 Design, Visualize and Explore Inverse Chebyshev Filter - I

Design, Visualize and Explore Inverse Chebyshev Filter - I

This example shows how to determine the transfer function for a fifth-order inverse Chebyshev
lowpass filter with 1 dB passband attenuation, cutoff frequency of 1 rad/sec, and a minimum
attenuation of 50 dB in the stopband. Determine the amplitude response at 2 rad/sec [1].

The rffilter object is used to design a RF Filter. A filter requires a minimum set of parameters
for it to be completely defined. Refer to the table in the rffilter documentation page which reflects
this set of required parameters. Each set of parameters result in its corresponding syntax. Input
these parameters as name-value pairs to rffilter to design the specified filter. Note that the
parameters which are required but are not defined assume default values.

After initialization of an rffilter object, the property DesignData contains the complete solution
of the filter designed. It is a structure which contains fields such as the computed factorized
polynomials for the construction of the transfer function.

Design Chebyshev Type II filter

N = 5; % Filter order
Fp = 1/(2*pi); % Passband cutoff frequency
Ap = 1; % Passband attenuation
As = 50; % Stopband attenuation

Use rffilter object to create a desired filter. The only implementation type for Inverse Chebyshev
is 'Transfer function'.

r = rffilter('FilterType','InverseChebyshev','ResponseType','Lowpass', ...
'Implementation','Transfer function','FilterOrder',N, ...
'PassbandFrequency',Fp,'StopbandAttenuation',As, ...
'PassbandAttenuation',Ap);

Generate and Visualize Transfer Function Polynomial

Use tf function to generate transfer function polynomials.

[numerator, denominator] = tf(r);

format long g

Display Numerator21 polynomial coefficients.

disp('Numerator polynomial coefficients of Transfer function');

Numerator polynomial coefficients of Transfer function

disp(numerator{2,1});

0.0347736250821381 0 0.672768334081369

Display Denominator polynomial coefficients.

disp('Denominator polynomial coefficients of Transfer function');

Denominator polynomial coefficients of Transfer function

disp(denominator);

1 3.81150884154936 7.2631952221038 8.6134457

6-193
6 RF Toolbox Examples

Optionally, use Control System Toolbox™ to visualize all transfer functions.

G_s = tf(numerator,denominator)

G_s =

From input 1 to output...

s^5
1: --------------------------------------------------------
s^5 + 3.812 s^4 + 7.263 s^3 + 8.613 s^2 + 6.43 s + 2.603

0.03477 s^4 + 0.6728 s^2 + 2.603

2: --------------------------------------------------------
s^5 + 3.812 s^4 + 7.263 s^3 + 8.613 s^2 + 6.43 s + 2.603

From input 2 to output...

0.03477 s^4 + 0.6728 s^2 + 2.603
1: --------------------------------------------------------
s^5 + 3.812 s^4 + 7.263 s^3 + 8.613 s^2 + 6.43 s + 2.603

s^5
2: --------------------------------------------------------
s^5 + 3.812 s^4 + 7.263 s^3 + 8.613 s^2 + 6.43 s + 2.603

Continuous-time transfer function.

Visualize Amplitude Response of Filter

frequencies = linspace(0,1,1001);
Sparam = sparameters(r, frequencies);

Note: S-parameters computes the transfer function using quadratic (lowpass/highpass) or quartic
(bandpass/bandstop) factorized forms. These factors are used to construct the polynomials. The
polynomial form is numerically unstable for larger filter order so the preferred form is the factorized
quadratic/quartic forms. These factorized parts are present in r.DesignData. For example, the
numerator21 can be accessed using r.DesignData.Numerator21.

l = rfplot(Sparam,2,1);

6-194
 Design, Visualize and Explore Inverse Chebyshev Filter - I

Amplitude Response of Filter at Specified Frequency

freq = 2/(2*pi);
hold on;
setrfplot('noengunits',false);

Note: To use rfplot and plot on the same figure use setrfplot.

plot(freq*ones(1,101),linspace(-120,20,101));
setrfplot('engunits',false);
[~,freq_index]= min(abs(frequencies-freq));
datatip(l,'DataIndex',freq_index);

6-195
6 RF Toolbox Examples

Using the data tip, the magnitude at 2 rad/sec is found to be -36.59 dB.

Evaluate the exact value at 2 rad/sec.

S_freq = sparameters(r,freq);
As_freq = 20*log10(abs(rfparam(S_freq,2,1)));
sprintf('Amplitude response at 2 rad/sec is %d dB',As_freq)

ans =
'Amplitude response at 2 rad/sec is -3.668925e+01 dB'

Calculate Stopband Frequency at As

Fs = r.DesignData.Auxiliary.Wx*r.PassbandFrequency;
sprintf('Stopband frequency at -%d dB is: %d Hz',As, Fs)

ans =
'Stopband frequency at -50 dB is: 3.500241e-01 Hz'

6-196
 Design, Visualize and Explore Inverse Chebyshev Filter - I

References

[1] Ellis, Michael G. Electronic Filter Analysis and Synthesis. Boston: Artech House, 1994.

See Also

More About
• “Design, Visualize and Explore Inverse Chebyshev Filter - II” on page 6-198
• “Design IF Butterworth Bandpass Filter” on page 6-183

6-197
6 RF Toolbox Examples

Design, Visualize and Explore Inverse Chebyshev Filter - II

This example shows how to design a fourth-order inverse Chebyshev low-pass filter with stopband
frequency of 10000 rad/sec, and epsilon of 0.01 (please see the reference section) using rffilter.
This rffilter could be used in a circuit or in a rfbudget object.

The rffilter object is used to design a RF filter. A filter requires a minimum set for parameters to
completely define it.

The parameters to design an inverse Chebyshev filter can be one of the following:

• Filter order, Passband frequency, Passband and Stopband Attenuation

• Passband and Stopband frequencies, Passband and Stopband Attenuation
• Filter order, Stopband frequency, Stopband Attenuation

Design Filter

N = 4; % Filter order
Fs = 1000/(2*pi); % Stopband frequency
epsilon = 0.01;
Rs = 10*log10((1+epsilon^2)/epsilon^2); % Stopband attenuation

Use the first set of parameters to define the filter.

r = rffilter('FilterType','InverseChebyshev','ResponseType','Lowpass', ...
'Implementation','Transfer function','FilterOrder',N, ...
'PassbandFrequency',Fs,'PassbandAttenuation',Rs, ...
'StopbandAttenuation',Rs);

Note: Alternative, you can also use the third set of parameters to design the same filter:

r = rffilter('FilterType','InverseChebyshev','ResponseType','Lowpass', ...
'Implementation','Transfer function','FilterOrder',N, ...
'StopbandFrequency',Fs,'StopbandAttenuation',Rs);

The limitation of this parameter set is that it assumes the passband attenuation to be fixed at
10*log10(2) dB.

Visualize magnitude response, phase response, and phase delay of filter

frequencies = linspace(0,2*Fs,1001);
rfplot(r, frequencies);

6-198
Design, Visualize and Explore Inverse Chebyshev Filter - II

6-199
6 RF Toolbox Examples

6-200
 Design, Visualize and Explore Inverse Chebyshev Filter - II

Optionally, you can also use Signal Processing Toolbox to visualize the analog filter using:

freqs(numerator{2,1},denominator)

Find zeros, poles, and gain

[z,p,k] = zpk(r);

You can obtain zeros, poles, and gain of Transfer function (S21) by:

format long g
zeros_21 = z{2,1}

zeros_21 = 4×1 complex

0 + 1082.39220029239i
0 - 1082.39220029239i
0 + 2613.12592975275i
0 - 2613.12592975275i

poles_21 = p % Same denominator for S11, S12, S21 and S22

poles_21 = 4×1 complex

-171.158733950657 + 476.096694464131i

6-201
6 RF Toolbox Examples

-171.158733950657 - 476.096694464131i
-504.530434776367 + 240.786480832184i
-504.530434776367 - 240.786480832184i

k_21 = k{2,1}

k_21 =
0.00999950003749688

View transfer function in factorized form

View these factor forms directly from the filter r.

disp('Numerator of Transfer function as factors:');

Numerator of Transfer function as factors:

r.DesignData.Numerator21

ans = 2×3

1 0 1171572.87525381
0.00999950003749688 0 68280.8572899443

disp('Denominator of Transfer function as factors:');

Denominator of Transfer function as factors:

r.DesignData.Denominator

ans = 2×3

1 342.317467901314 255963.374687264
1 1009.06086955273 312529.088967178

Alternatively, use |zpk| from Control System Toolbox to view the transfer function in factorized form.

G_s = zpk(zeros_21,poles_21,k_21)

G_s =

0.0099995 (s^2 + 1.172e06) (s^2 + 6.828e06)

-------------------------------------------------
(s^2 + 1009s + 3.125e05) (s^2 + 342.3s + 2.56e05)

Continuous-time zero/pole/gain model.

6-202
 Design, Visualize and Explore Inverse Chebyshev Filter - II

References

[1] Paarmann, L. D. Design and Analysis of Analog Filters: A Signal Processing Perspective. SECS
617. Boston: Kluwer Academic Publishers, 2001.

See Also

More About
• “Design, Visualize and Explore Inverse Chebyshev Filter - I” on page 6-193
• “Design IF Butterworth Bandpass Filter” on page 6-183

6-203
6 RF Toolbox Examples

Design Matching Networks for Passive Multiport Network

This example shows how to design matching networks for 16-port passive network at 39 GHz for 5G
mmWave systems. Matching networks are designed independently for each port, and each generated
matching network is intended to function between two 1-port terminations.

Design Multiport Passive Network

Compute the S-Parameters of a patch antenna array designed at 39 GHz. Load the
sparams_patchArray.mat file. The s_params_circ_array function is obtained from the
supporting file designmultiport.mlx.

Fcenter = 39e9;
load('sparams_patchArray.mat')
Sparam_array = s_params_circ_array;
show(patchArray)
view([90 0])

Determine the index corresponding to the center frequency.

freq = Sparam_array.Frequencies;
fIndex = find(freq == Fcenter);

Create Matching Networks

Generate matching networks for each corresponding port independently, with a Loaded Q of 20 and
configure the topology to 'Pi'. This Q-factor is aligned with half power bandwidth of the patch antenna
array, which is approximately 2 GHz.

Define the number of ports in the network and specify the termination impedance.

6-204
 Design Matching Networks for Passive Multiport Network

numport = s_params_circ_array.NumPorts;
ZT = 50;
loadedQ = 20;
topology = 'Pi';
for i = 1 : numport
% reflection coefficient/Sii
gam_array = s_params_circ_array.Parameters(i,i,fIndex);
% Load impedance
Zout = gamma2z(gam_array);
% Matching networks generation
match_net(i) = matchingnetwork('SourceImpedance', ZT, ...
'LoadImpedance', Zout, 'CenterFrequency', Fcenter, ...
'LoadedQ', loadedQ, 'Components', topology);
end

The source is connected to the component located on left of the matching network circuit and the
load is connected to the component connected to the right of the matching network circuit. For the
matching networks generated, the source is terminated with ZT (50 Ohm) and the load impedance is
the impedance seen at the ith-port given by Zout.

View and Select Circuits

Select a topology from the sixteen matchingnetwork objects. To get an overview of the available
circuits, see circuitDescriptions function.

In this example, a Shunt C-Series L-Shunt C topology is used. If this topology is not available in your
network, use the best available matching network circuit.

selectedCircuits = repmat(circuit,1,numport);
cIndex = zeros(1,numport);

View the list of circuits generated.

for i = 1:numel(match_net)
c = circuitDescriptions(match_net(i));
% Perform a text search to choose the circuit with Shunt C-Series L-Shunt C topology
Index = strcmp(c.component1Type,"Shunt C") & ...
strcmp(c.component2Type,"Series L") & ...
strcmp(c.component3Type,"Shunt C");
if any(Index)
% ShuntC-SeriesL-ShuntC topology
cIndex(i) = find(Index, 1, 'first');
selectedCircuits(i) = match_net(i).Circuit(cIndex(i));
else
% Best available matchingnetwork
selectedCircuits(i) = match_net(i).Circuit(1);
end
selectedCircuits(i).Name = "N"+i;
end

To view the performance of a selected matching network circuit, use rfplot. For instance, to plot
the performance of the first matching network for the circuit with Shunt C-Series L-Shunt C topology
type this command.

rfplot(match_net(1),freq,cIndex(1));

6-205
6 RF Toolbox Examples

Add Matching Network Circuits to 16-Port Network

Create Circuit Object

Create a circuit object and an n-port object for the 16-port network.

ckt = circuit('patchArray');
array_net = nport(Sparam_array);

In this example, number of circuit nodes are shown as 17, as nodes 1 through 16 will be used for
adding the matching networks.

cktnodes = (1+numport):(numport+numport);

Add the n-port object to circuit object.

add(ckt, cktnodes, array_net);

View parent nodes of the 16-port network.

disp(array_net)

nport: N-port element

NetworkData: [1×1 sparameters]

Name: 'Sparams'
NumPorts: 16
Terminals: {1×32 cell}
ParentNodes: [17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 0 0 0 0 0 0 0 0 0 0 0 0 0 … ]
ParentPath: 'patchArray'

An illustration of the circuit object with 16-port n-port is provided.

6-206
 Design Matching Networks for Passive Multiport Network

Initialize the ports.

ports = cell(1,numport);

Add each matching network circuit to its corresponding port one at a time. Port numbers for
corresponding matching network circuit are also generated.

6-207
6 RF Toolbox Examples

for i=1:length(selectedCircuits)
add(ckt, [i, 0, i+numport, 0], selectedCircuits(i), ...
{'p1+', 'p1-', 'p2+', 'p2-'});
ports{i} = [i, 0];
end
% ports = arrayfun(@(x) [x 0],1:10,'UniformOutput',false);

Use the setports function to define the ports for each of the circuits.

setports(ckt,ports{:});

An illustration of the circuit object with n-port and matching network circuits are provided.

6-208
Design Matching Networks for Passive Multiport Network

6-209
6 RF Toolbox Examples

Generate and Plot S-Parameters

Generate and plot the S-Parameters of the passive 16-port matching network.

Sparam = sparameters(ckt, freq);

Plot Frequency Responses

Plot the frequency response of the 16-port network before matching.

figure; rfplot(s_params_circ_array); legend off

Plot the frequency response of the 16-port network after matching.

figure; rfplot(Sparam); legend off

6-210
 Design Matching Networks for Passive Multiport Network

See Also

More About
• “Design Two-Stage Low Noise Amplifier Using Microstrip Transmission Line Matching Network”
on page 6-218

6-211
6 RF Toolbox Examples

Frequency Sweep in RF Budget Analysis

This example shows how to sweep through frequency-dependent properties of the elements in an RF
Budget Analysis.

Use the nport and amplifier objects to specify the 2-port RF elements in the design and build an
RF budget element by cascading the elements together into an RF system with rfbudget.

Building Elements of RF Budget Cascade

Build and parameterize each of the 2-port RF elements.

f1 = nport('RFBudget_RF.s2p','RFBandpassFilter');
a1 = amplifier('Name','RFAmplifier', ...
'Gain',11.53, ...
'NF',1.53, ...
'OIP3',35);

Use rfbudget object to cascade the elements with input frequency 2.1 GHz, input power -30 dBm,
and input bandwidth 45 MHz. This example cascades a filter and an amplifier.

b = rfbudget('Elements',[f1 a1], ...

'InputFrequency',2.1e9, ...
'AvailableInputPower',-30, ...
'SignalBandwidth',45e6);

Read Frequency-Dependent Noise Figure

Read frequency-dependent Noise Figure (NF) values of the amplifier from the data-sheet. A similar
approach can be followed if the Output third-order intercept (OIP3) or Gain is frequency-dependent.

% Inputs from the data-sheet

freq_datasheet = [1.98;1.99;2.0;2.01;2.02;2.03;2.04;2.05;2.06;2.07;2.08;....
2.09;2.10].*1e9;

NF_datasheet = [1.0000;1.0442;1.0883;1.1325;1.1767;1.2208;1.2650;1.3092;...
1.3533;1.3975;1.4417;1.4858;1.5300];

% Interpolate the amplifier NF data based on existing filter frequencies

Freq = f1.NetworkData.Frequencies;
RFAmplifier_NF = interp1(freq_datasheet,NF_datasheet,Freq);

Plot RF Budget Results Versus Input Frequency

Loop over the desired frequencies, by setting NF of the RF Amplifier element in the rfbudget object.

TotalNF = zeros(size(Freq));
for i = 1:numel(Freq)
b.InputFrequency = Freq(i);

% Adjust frequency-dependent NF of the RF Amplifier

elems(2).NF = RFAmplifier_NF(i);

% Compute NF of the cascade

TotalNF(i) = b.NF(end);

6-212
 Frequency Sweep in RF Budget Analysis

end
plot(Freq/1e9,TotalNF)
grid on;
xlabel('Frequency (GHz)')
ylabel('NF (dB)')
title('Noise Figure vs. Input Frequency')

See Also

More About
• “RF Budget Harmonic Balance Analysis of Low-IF Receiver, IP2 and NF” on page 6-225
• “Visualizing RF Budget Analysis over Bandwidth” on page 6-14

6-213
6 RF Toolbox Examples

Rational Fit S-Parameters

This example shows how to use the rational object to create a rational fit to S-parameter data, and
the various properties and methods that are included in the rational object.

Create Rational Object

Read in the sparameters, and create the rational object from them. The rational function
automatically fits all entries of the S-parameter matrices.

S = sparameters('sawfilter.s2p')

S =
sparameters with properties:

Impedance: 50
NumPorts: 2
Parameters: [2x2x334 double]
Frequencies: [334x1 double]

r = rational(S)

r =
rational with properties:

NumPorts: 2
NumPoles: 24
Poles: [24x1 double]
Residues: [2x2x24 double]
DirectTerm: [2x2 double]
ErrDB: -40.9658

With the default settings on this example, the rational object achieves an accuracy of about -26 dB,
using 30 poles. By construction, the rational object is causal, with a non-zero direct term.

Compare Fit with Original Data

Generate the frequency response from the rational object, and compare one of the entries with the
original data.

resp = freqresp(r, S.Frequencies);

plot(S.Frequencies, real(rfparam(S, 1, 1)), ...
S.Frequencies, real(squeeze(resp(1,1,:))))

6-214
 Rational Fit S-Parameters

Limit Number of Poles

Redo the fit, limiting the number of poles to a maximum of 5. The rational object may use fewer
poles than specified. Notice that the quality of the fit is degraded as opposed to the original 30-pole
fit.

r5 = rational(S, 'MaxPoles', 5)

r5 =
rational with properties:

NumPorts: 2
NumPoles: 4
Poles: [4x1 double]
Residues: [2x2x4 double]
DirectTerm: [2x2 double]
ErrDB: -1.7376

resp5 = freqresp(r5, S.Frequencies);

plot(S.Frequencies, real(rfparam(S, 1, 1)), ...
S.Frequencies, real(squeeze(resp5(1,1,:))))

6-215
6 RF Toolbox Examples

Tighten Target Accuracy

Redo the fit, asking for a tighter tolerance (-60 dB), Notice that the fit is significantly improved,
particularly in the stopbands of the SAW filter.

rgood = rational(S, -60,'QLimit',inf,'TendsToZero',false)

rgood =
rational with properties:

NumPorts: 2
NumPoles: 85
Poles: [85x1 double]
Residues: [2x2x85 double]
DirectTerm: [2x2 double]
ErrDB: -56.2635

respgood = freqresp(rgood, S.Frequencies);

plot(S.Frequencies, real(rfparam(S, 1, 1)), ...
S.Frequencies, real(squeeze(respgood(1,1,:))))

6-216
 Rational Fit S-Parameters

See Also

More About
• “Test, Visualize, and Enforce Passivity of Rational Output” on page 6-187
• “Using 'NPoles' Parameter With rationalfit” on page 6-71
• “Using 'Weight' Parameter With rationalfit” on page 6-75

6-217
6 RF Toolbox Examples

Design Two-Stage Low Noise Amplifier Using Microstrip

Transmission Line Matching Network

This example shows how to use the RF Toolbox™ microstrip transmission line element to design two-
stage low noise amplifier (LNA) for wireless local area network (WLAN) with an input and the output
matching network (MNW) to maximize the power delivered through a 50-ohm load and the system.

Designing an input and output MNW is an important part of the amplifier design. The amplifier in this
example has high gain and low noise. To minimize parasitic effect, this example uses the microstrip
transmission line MNW with a single stub.

Define Microstrip Transmission Line Parameters

The microstrip transmission line parameters are chosen as follows.

• Physical Height of conductor or dielectric thickness — 1.524 mm

• Relative permittivity of dielectric — 3.48
• Loss angle tangent of dielectric — 0.0037
• Physical thickness of microstrip transmission line — 3.5 um

Design Input Matching Network Using Microstrip Transmission Line

The input matching network consists of one shunt stub and one series microstrip transmission line.

Create an input shunt stub microstrip transmission line with the physical length of 8.9 mm.

TL1 = txlineMicrostrip('Width',3.41730e-3,'Height',1.524e-3,'EpsilonR',3.48,'LossTangent',0.0037,
'LineLength',8.9e-3,'Thickness',0.0035e-3,'StubMode','Shunt','Termination','Open');

Create an input series microstrip transmission line with the physical length of 14.7 mm.

TL2 = txlineMicrostrip('Width',3.41730e-3,'Height',1.524e-3,'EpsilonR',3.48, 'LossTangent',0.0037

'LineLength',14.7e-3,'Thickness',0.0035e-3);

6-218
 Design Two-Stage Low Noise Amplifier Using Microstrip Transmission Line Matching Network

Create and Extract Amplifier Object

Create and extract an amplifier object from the frequency dependent S-parameter data available in
the specified file.

amp1 = nport('f551432p.s2p');

Define the frequency range.

freq = 2e9:10e6:3e9;

Create a two-stage amplifier and plot its S-parameter.

casamp = circuit([amp1,clone(amp1)],'amplifiers'); % amplifier ciruit without MNW.

Plot the S-Parameter over the frequency range from 2 – 3 GHz.

S2 = sparameters(casamp,freq);

Design Output Matching Network Using Microstrip Transmission Line

The output matching network consists of one shunt stub and one series microstrip transmission line.

Create an output series microstrip transmission line with the physical length of 22.47 mm.

TL3 = txlineMicrostrip('Width',3.41730e-3,'Height',1.524e-3,'EpsilonR',3.48, 'LossTangent',0.0037

'LineLength',22.47e-3,'Thickness',0.0035e-3);

Create an output shunt stub microstrip transmission line with the physical length of 5.66 mm.

TL4 = txlineMicrostrip('Width',3.41730e-3, 'Height',1.524e-3,'EpsilonR',3.48, 'LossTangent',0.003

'LineLength',5.66e-3,'Thickness',0.0035e-3,'StubMode','Shunt','Termination','Open');

Plot Input Reflection Coefficients of Two-Stage LNA

To verify the simultaneous conjugate match at the input of the amplifier, plot the input reflection
coefficients in dB for the amplifier circuit with and without a matching network.

Cascade the circuit elements by adding the input and the output MNW to the two-stage amplifier.

c = circuit([TL1, TL2,clone(amp1),clone(amp1),TL3, TL4]); % two-stage LNA with MNW

Plot the S-parameters and analyze the amplifier with and without the matching networks over the
frequency range of 2.4 – 2.5 GHz.

figure
S3 = sparameters(c,freq);
rfplot(S2,1,1)
hold on;
rfplot(S3,1,1)
legend('|S11| of Two-Stage LNA Without MNW','|S11| of Two-Stage LNA with MNW');
title('Input Reflection Coefficients of Two-Stage LNA');
grid on;

6-219
6 RF Toolbox Examples

The calculated input return loss for the two-stage LNA with the input MNW is around 13 dB.

Plot Output Reflection Coefficients of Two-Stage LNA

To verify the simultaneous conjugate match at the output of the amplifier, plot output reflection
coefficients in dB for both the two-stage LNA with and without a MNW.

figure
rfplot(S2,2,2)
hold on;
rfplot(S3,2,2)
legend('|S22| of Without MNW','|S22| of With MNW');
title('Output Reflection Coefficients of Two-Stage LNA');
grid on;

6-220
 Design Two-Stage Low Noise Amplifier Using Microstrip Transmission Line Matching Network

The calculated output return loss for the two-stage LNA with the output MNW is aroung 11 dB.

Plot Gain and Input Reflection Coefficients of Cascaded LNA

To verify the simultaneous conjugate match at the input and output of the amplifier, plot the input
reflection coefficient and the gain parameters in dB for the two-stage LNA with the MNW.

figure;
rfplot(S3,1,1)
hold on;
rfplot(S3,2,1)
title('Gain and Input Reflection Coefficients of Cascaded LNA');
grid on;

6-221
6 RF Toolbox Examples

The calculated amplifier gain, S21 is 34.5 dB, and the input reflection coefficient, S11 is around 13
dB.

Calculate and Plot Complex Load and Source Reflection Coefficients

Calculate and plot all the complex load and source reflection coefficients for simultaneous conjugate
match at all measured frequency data points that are unconditionally stable. These reflection
coefficients are measured at the amplifier interfaces.

figure
smithplot(S3,1,1,'LegendLabels','Measured S11')

6-222
 Design Two-Stage Low Noise Amplifier Using Microstrip Transmission Line Matching Network

Calculate Ampifier Noise Figure

Use an rfbudget object to calculate the amplifier noise figure.

b = rfbudget( ...
'Elements',[TL1 TL2 amp1 clone(amp1) TL3 TL4], ...
'InputFrequency',2.45e9, ...
'AvailableInputPower',0, ...
'SignalBandwidth',2e9, ...
'Solver','Friis', ...
'AutoUpdate',1);
rfplot(b,'NF')

6-223
6 RF Toolbox Examples

The amplifier noise figure is calculated as 0.7 dB.

Reference

[1] Maruddani, B, M Ma’sum, E Sandi, Y Taryana, T Daniati, and W Dara. “Design of Two Stage Low
Noise Amplifier at 2.4 - 2.5 GHz Frequency Using Microstrip Line Matching Network Method.”
Journal of Physics: Conference Series 1402 (December 2019): 044031.

See Also

More About
• “Analysis of Coplanar Waveguide Transmission Line in X-Band Application” on page 6-231
• “Design Matching Networks for Passive Multiport Network” on page 6-204
• “Design Broadband Matching Networks for Antennas” on page 6-139
• “Impedance Matching of Small Monopole Antenna” on page 6-166

6-224
 RF Budget Harmonic Balance Analysis of Low-IF Receiver, IP2 and NF

RF Budget Harmonic Balance Analysis of Low-IF Receiver, IP2

and NF

This example shows how to use the rfbudget object's harmonic balance solver to analyze a low-IF
(intermediate frequency) receiver RF budget for second-order intercept point (IP2), the second-order
intercept point and to compute a more accurate noise figure (NF) that correctly accounts for system
nonlinearity and noise-folding.

Use amplifier and modulator objects to construct the 2-port RF elements in a low-IF receiver
design, along with their output second-order intercept point (OIP2) specifications. You can turn off
the default ideal image reject and channel select filtering in the modulator with the ImageReject
and ChannelSelect logical name-value pairs.

Compute RF budget results by cascading the elements together into an RF system with rfbudget.
The rfbudget object enables design exploration and visualization at the MATLAB command-line. It
also enables automatic RF Blockset model and measurement testbench generation.
a1 = amplifier('Name','RFAmplifier', ...
'Gain',11.53, ...
'NF',1.53, ...
'OIP2',35);

d = modulator('Name','Demodulator', ...
'Gain',-6, ...
'NF',4, ...
'OIP2',50, ...
'LO',2.03e9, ...
'ConverterType','Down', ...
'ImageReject',false, ...
'ChannelSelect',false);

a2 = amplifier('Name','IFAmplifier', ...
'Gain',30, ...
'NF',8, ...
'OIP2',37);

b = rfbudget('Elements',[a1 d a2], ...

'InputFrequency',2.1e9, ...
'AvailableInputPower',-30, ...
'SignalBandwidth',45e6)

b =
rfbudget with properties:

Elements: [1x3 rf.internal.rfbudget.RFElement]

InputFrequency: 2.1 GHz
AvailableInputPower: -30 dBm
SignalBandwidth: 45 MHz
Solver: Friis
AutoUpdate: true

Analysis Results
OutputFrequency: (GHz) [ 2.1 0.07 0.07]
OutputPower: (dBm) [-18.47 -24.47 5.53]
TransducerGain: (dB) [ 11.53 5.53 35.53]

6-225
6 RF Toolbox Examples

NF: (dB) [ 1.53 1.843 4.793]

IIP2: (dBm) []
OIP2: (dBm) []
IIP3: (dBm) [ Inf Inf Inf]
OIP3: (dBm) [ Inf Inf Inf]
SNR: (dB) [ 65.91 65.6 62.65]

Why are OIP2 and IIP2 Empty in the Results?

The default Solver property of the rfbudget object is 'Friis', an equivalent baseband approximation
which is unable to compute IP2. To see the IP2 results, you can set the Solver property of the
budget object to 'HarmonicBalance'. This performs nonlinear circuit analysis to compute the steady-
state operating point, from which it is possible to compute IP2.

You can also select the 'HarmonicBalance' solver at rfbudget construction time by passing in a
Solver name-value pair after the other positional or name-value pair arguments, e.g.

b = rfbudget([a1 d a2],2.1e9,-30,45e6,'Solver','HarmonicBalance')

In general, the 'HarmonicBalance' solver is not as fast as the 'Friis' solver and does not compute
noise figure (NF) or signal-to-noise ratio (SNR).

b.Solver = 'HarmonicBalance'

b =
rfbudget with properties:

Elements: [1x3 rf.internal.rfbudget.RFElement]

InputFrequency: 2.1 GHz
AvailableInputPower: -30 dBm
SignalBandwidth: 45 MHz
Solver: HarmonicBalance
WaitBar: true
AutoUpdate: true

Analysis Results
OutputFrequency: (GHz) [ 2.1 0.07 0.07]
OutputPower: (dBm) [-18.47 -24.47 5.53]
TransducerGain: (dB) [ 11.53 5.53 35.53]
NF: (dB) [ 1.53 4.7 6.487]
IIP2: (dBm) [ 23.47 44.47 -4.581]
OIP2: (dBm) [ 35 50 30.95]
IIP3: (dBm) [ Inf Inf 19.45]
OIP3: (dBm) [ Inf Inf 54.98]
SNR: (dB) [ 65.91 62.74 60.96]

The rfbudget display above shows the results of the cascade computed by the 'HarmonicBalance'
solver. Comparing them to the 'Friis' results, the vector properties showing the OutputPower and
TransducerGain along the cascade match well.

As expected, the OIP2 and IIP2 properties have nonempty values. In addition, the output third-order
intercept point (OIP3) and input third-order intercept point (IIP3) properties have changed. The 'Friis'
solver is unable to capture the nonlinear bleeding through the IP2 properties of the cascade to affect
the third-order intercept point. Mathematically, this happens because cascading two second-order
polynomials results in a polynomial with a third-order term.

6-226
 RF Budget Harmonic Balance Analysis of Low-IF Receiver, IP2 and NF

Similarly, the NF results of Harmonic Balance are different (and more accurate) than the Friis results
because Harmonic Balance correctly captures the noise folding effects of nonlinearities.

Verifying HB Results Using RF Blockset Circuit Envelope Simulation

You can verify the harmonic balance NF, IP2 and IP3 results by exporting the budget to an RF
Blockset testbench model using the following command:

exportTestbench(b)

To verify NF, double-click on the RF Measurement Unit to open the mask, then select NF from the
Measured quantity pulldown. Then run the model. This verifies the Harmonic Balance NF calculation.

To verify IP2, double-click on the RF Measurement Unit to open its mask, then select IP2 from the
Measured quantity pulldown.

6-227
6 RF Toolbox Examples

Also uncheck the Simulate noise checkbox. Then run the model.

6-228
 RF Budget Harmonic Balance Analysis of Low-IF Receiver, IP2 and NF

To verify IP3, select IP3 from the Measured quantity pulldown and run the model again.

6-229
6 RF Toolbox Examples

Verifying HB results with RF Blockset Harmonic Balance

Rather than using the large machinery of circuit envelope and the RF Testbench, it is possible to
build a simpler model that computes the IP2 and IP3 using two tones and harmonic balance. Open the
model oipHB.slx found in the MATLAB/Examples folder. Simulate the model.

See Also

More About
• “Superheterodyne Receiver Using RF Budget Analyzer App” on page 6-2
• “Visualizing RF Budget Analysis over Bandwidth” on page 6-14
• “Frequency Sweep in RF Budget Analysis” on page 6-212

6-230
 Analysis of Coplanar Waveguide Transmission Line in X-Band Application

Analysis of Coplanar Waveguide Transmission Line in X-Band

Application

This example shows how to analyze a coplanar waveguide (cpw) transmission line for X-band
applications. CPW transmission line consists of a central metal strip separated by a narrow gap from
two ground planes on either side. The dimensions of the center strip, the gap, the thickness, and
permittivity of the dielectric substrate determine the characteristic impedance, group delay, and
noise. The gap in the cpw is usually very small and supports electric fields primarily concentrated in
the dielectric.

Define Parameters

The cpw transmission line has 200 mm slot width, 1600 mm conductor width, 635 mm height, 0.005
loss tangent, and 17 um of thickness. This example uses two different dielectric constants to simulate
the cpw transmission line. The dielectric constant values are 2.323 and 9.2.

cptxline1 = txlineCPW('EpsilonR',2.323,'SlotWidth',200e-6,'ConductorWidth',...
1600e-6,'Height',635e-6,'LossTangent',0.005,'Thickness',17e-6);
cptxline2 = txlineCPW('EpsilonR',9.2,'SlotWidth',200e-6,'ConductorWidth',...
1600e-6,'Height',635e-6,'LossTangent',0.005,'Thickness',17e-6);
% x band Frequency range 8 to 12GHz
freq = 5e9:10e6:14e9;

Plot Input Return Loss

The results for two different dielectric substrates indicates the impedance bandwidth increases with a
lower dielectric constant. The measurement results are for a frequency range of 5 GHz to 14GHz, and
magnitude of S11 < 10 dB.

figure;
sp1 = sparameters(cptxline1,freq);
sp2 = sparameters(cptxline2,freq);
rfplot(sp1,1,1);hold on;
rfplot(sp2,1,1);
title('Frequency Vs S-Parameters');
legend('EpsilonR 2.323','EpsilonR 9.2');
grid on;

6-231
6 RF Toolbox Examples

Group Delay

Group delay variations versus frequency is an essential factor when using phase modulation and high
data rates. This impairment causes distortion and degradation in wideband applications. In a cpw
transmission line the group delay increases with increase in the frequency for both dielectric
substrates.

gd1 = groupdelay(cptxline1,freq,'Impedance',50);
gd2 = groupdelay(cptxline2,freq,'Impedance',50);
figure;plot(freq,gd1);hold on;
plot(freq,gd2);
title('Frequency Vs Group delay');
legend('EpsilonR 2.323','EpsilonR 9.2');
xlabel('Frequency');
ylabel('Group delay');
grid on;

6-232
 Analysis of Coplanar Waveguide Transmission Line in X-Band Application

Noise Figure

The noise is generated primarily within the input stages of the receiver system itself. Cascaded stages
are not noisier than others. The noise generated at the input and amplified by the receiver's full gain
amplifier greatly exceeds the noise generated further along the receiver chain. In the results using
both lower and higher dielectric constant, noise figure increases with increasing frequency. The
variation is very less over the frequency range when using a lower dielectric constant.

nf1 = noisefigure(cptxline1,freq);
nf2 = noisefigure(cptxline2,freq);
figure;plot(freq,nf1);hold on;
plot(freq,nf2);
title('Frequency Vs Noise Figure');
legend('EpsilonR 2.323','EpsilonR 9.2');
xlabel('Frequency');
ylabel('Noise Figure');
grid on;

6-233
6 RF Toolbox Examples

Characteristic Impedance

Relative permittivity for a homogeneous dielectric affects the characteristic impedance of cpw
transmission line. You can compute this approximately by using the electrical model of the cpw to
clarify impedance behavior along the frequency band. Characteristic impedance determines the
amount of power transfer and attenuation effect along the cpw transmission line. The characteristic
impedance of a transmission line is usually written as Z0. In the simulation, the resulting
characteristic impedance decreases with increasing frequency in both dielectric constants. With
lower dielectric constant impedance value is below 50 ohms, with higher dielectric constant
impedance value is above 50 ohms.

ChImp1 = getZ0(cptxline1,freq);
ChImp2 = getZ0(cptxline2,freq);
figure; plot(freq,ChImp1);hold on;
plot(freq,ChImp2);
title('Frequency Vs Characteristics Impedance');
xlabel('Frequency');
ylabel('Characteristics Impedance');
legend('EpsilonR 2.323','EpsilonR 9.2');
grid on;

6-234
 Analysis of Coplanar Waveguide Transmission Line in X-Band Application

Conclusion

In RF and microwave circuit design the dielectric permittivity of the substrate plays an important role
and requires precise evaluation over a broad range of frequencies. With the above simulation you see
that, lower dielectric constant gives wider bandwidth, lower noise figure, and lower group delay.

Reference:

Sova, M., and I. Bogdan. "Coplanar Waveguide Resonator Design for Array Antenna Applications." In
6th International Conference on Telecommunications in Modern Satellite, Cable and Broadcasting
Service, 2003. TELSIKS 2003.,1:57 to 59. Serbia, Montenegro, Nis: IEEE, 2003.

See Also

More About
• “Design Two-Stage Low Noise Amplifier Using Microstrip Transmission Line Matching Network”
on page 6-218

6-235
6 RF Toolbox Examples

Extract S-Parameters from Circuit

This example uses the Symbolic Math Toolbox™ to explain how the RF Toolbox™ extracts two-port S-
parameters from an RF Toolbox circuit object.

Consider a two-port network as shown in the figure 1that you want to characterize with S-
parameters. S-parameters are defined as V – I × Z0 = S V + I × Z0 .

Figure 1: Two-Port Network

To extract the S-parameters from a circuit into an sparameters object, the RF Toolbox terminates
each port with the reference impedance Z0. Then, the RF Toolbox independently drives each port j,
1
with Z and solves for the port voltages V ij. Driving with current sources is the Norton equivalent of
0
driving with a 1 V source and a series resistance of Z0.

Measure the port voltage V ij at node i when node j is driven.

• If i ≠ j, the S-parameter entry Sij is simply twice the port voltage V ij, and this is given using the
equation Sij = 2 × V ij.
• The diagonal entries of S-parameters when i = j are given using the equation Sij = 2 × V ij − 1.

6-236
 Extract S-Parameters from Circuit

Figure 2: Circuit Driven at Port 1 with Current Source

Write Constitutive and Conservative Equations of Circuit

Circuits are represented in node-branch form in the RF Toolbox. There are four branches in the
circuit represented in figure 2, one for the input port, two for the two-port nport object, and one for
the output port. This means that the circuit has four branch current unknowns IS, I1, I2, and IL and
two node voltages V 11 and V 21. To represent the circuit described in figure 2 in node-branch form,
you need four constitutive equations to represent the branch currents and two conservative equations
to represent the node voltages.

syms F IS I1 I2 IL V1 V2 Z0
syms S11 S12 S21 S22

nI = 4; % number of branch currents

nV = 2; % number of node voltages

% F = [Fconstitutive; Fconservative]
F = [
V1 - Z0*IS
V1 - Z0*I1 - S11*(V1+Z0*I1) - S12*(V2+Z0*I2)
V2 - Z0*I2 - S21*(V1+Z0*I1) - S22*(V2+Z0*I2)
V2 - Z0*IL
IS+I1
I2+IL
]

F =
V 1 − IS Z0
V 1 − I1 Z0 − S11 V 1 + I1 Z0 − S12 V 2 + I2 Z0
V 2 − I2 Z0 − S21 V 1 + I1 Z0 − S22 V 2 + I2 Z0
V 2 − IL Z0
I1 + IS
I2 + IL

6-237
6 RF Toolbox Examples

Jacobian Evaluation of Circuit

Use the jacobian function from the Symbolic Math Toolbox to compute the matrix of derivatives of
the function F with respect to the six unknowns (four branch currents and two node voltages)

J = jacobian(F,[IS; I1; I2; IL; V1; V2])

J =
−Z0 0 0 0 1 0
0 −Z0 − S11 Z0 −S12 Z0 0 1 − S11 −S12
0 −S21 Z0 −Z0 − S22 Z0 0 −S21 1 − S22
0 0 0 −Z0 0 1
1 1 0 0 0 0
0 0 1 1 0 0

Solve S-Parameters of Circuit

Create a two-column right-hand side vector, rhs, to represent the driving of each port.

syms rhs [nI+nV 2]

syms x v S

% Compute S-parameters of cascade

rhs(:,:) = 0;
rhs(nI+1,1) = 1/Z0; % rhs for driving input port
rhs(nI+nV,2) = 1/Z0 % rhs for driving output port

rhs =
0 0
0 0
0 0
0 0
1
0
Z0
1
0
Z0

To solve for the voltages, back solve the rhs with the Jacobian. The S-parameter matrix that MATLAB
outputs represents the two-port circuit shown in figure 1.

x = J \ rhs;
v = x(nI+[1 nV],:);
S = 2*v - eye(2)

S =
S11 S12
S21 S22

See Also

More About
• “Extract S-Parameters from Mutual Inductor” on page 6-240

6-238
 Extract S-Parameters from Circuit

• “De-Embedding S-Parameters” on page 6-110

• “Bisect S-Parameters of Cascaded Probes” on page 6-114

6-239
6 RF Toolbox Examples

Extract S-Parameters from Mutual Inductor

This example shows how to build a user-defined element from S-parameters and add it to an
rfbudget object for link budget analysis using the Symbolic Math Toolbox™. The user-defined
element in this example is a mutual inductor.

Consider a mutual inductor as shown in figure 1 with the inductors Laand Lb. This examples uses the
Symbolic Math Toolbox to extract the analytical S-parameters of the mutual inductor and write them
an RF Toolbox™ object. To extract S-parameters from a circuit, see “Extract S-Parameters from
Circuit” on page 6-236.

Figure 1: Mutual Inductor

One way to model a mutual inductor in the RF Toolbox is to draw the mutual inductor as an
equivalent of a two-port network of inductors in a T configuration. Such a mutual inductor is shown in
figure 2 with the mutual inductance M and the coupling coefficient k. Mutual inductance is given by
the equation M = k La × Lb relates M and k. Inductors in a T configuration can have negative values
when there is a strong coupling between the inductors or if the M is greater than Laor Lb.

6-240
 Extract S-Parameters from Mutual Inductor

Figure 2: T-Circuit Representation of Mutual Inductor

Represent Circuit in Node-Branch Form

As discussed in the “Extract S-Parameters from Circuit” on page 6-236 example, to extract S-
parameters from a circuit you need to drive one port while terminating the other. This is shown in
figure 3. Use constitutive and conservative equations to represent the circuit in node-branch form.
There are eight unknowns, five branch currents and three node voltages. Therefore there are eight
equations in the node-form, five constitutive equations for the branches and three conservative
equations obtained from the Kirchoff's Current Law for the nodes. The constitutive equation for a
resistor is derived from Ohm's Law,V = IR, and the constitutive equation for an inductor is given by
V = sLR, where s is a complex frequency.

Figure 3: Mutual Inductor Driven at Port 1 with Current Source

syms F
syms I [5 1]

6-241
6 RF Toolbox Examples

syms V [3 1]
syms Z0 La Lb M s

nI=5; % number of branch currents

nV=3; % number of node voltages

% F = [Fconstitutive; Fconservative]
F = [
V1 - Z0*I1
V1 - V2 - (La-M)*I3*s
V2 - M*I4*s
V2 - V3 + (Lb-M)*I5*s
V3 - I2*Z0
I1 + I3
I4 - I5 - I3
I2 + I5
]

F =
V 1 − I1 Z0
V 1 − V 2 − I3 s La − M
V 2 − I4 M s
V 2 − V 3 + I5 s Lb − M
V 3 − I2 Z0
I1 + I3
I4 − I3 − I5
I2 + I5

Compute Jacobian

Determine the Jacobian with respect to the unknowns, the five branch currents and three node
voltages.

J = jacobian(F,[I; V]);

Solve S-parameters for Further Analysis

As shown in the Extract S-Parameter from a Circuit example, create the right-hand side, rhs
vector to the drive and terminate ports.

syms rhs [nI+nV 2]

syms x v S t

% Compute S-parameters of cascade

rhs(:,:) = 0;
rhs(nI+1,1) = 1/Z0; % rhs for driving input port
rhs(nI+nV,2) = 1/Z0 % rhs for driving output port

rhs =

6-242
 Extract S-Parameters from Mutual Inductor

0 0
0 0
0 0
0 0
0 0
1
0
Z0
0 0
1
0
Z0

By backsolving rhs, solve for the voltages using Jacobian.

x = J \ rhs;
v = x(nI+[1 nV],:);
S = (2*v - eye(2));

Create Object for RF Toolbox

In order to create a sparameters object, the parameters must be determined at a set of frequencies.
To do so, define the variables for your mutual inductor. If you would like to test multiple values for
your variables and automatically update your sparameters object, use Numeric Sliders in the
Control drop-down under the Live Editor tab. Then, use the Symbolic Math Toolbox's
matlabFunction to automatically generate a function, mutualInductorS to compute the analytic
S-parameters at a set of frequencies. Finally, use the sparameters object to create a S-parameters
object.
matlabFunction(S,'file','mutualInductorS.m','Optimize',false);

La = ;

Lb = ;

Z0 = ;

k = ;
M = k*((La*Lb)^(1/2));

freq = linspace(1e9,2e9,10);
s = 2i*pi*freq;
s_param = zeros(2,2,10);
for index = 1:numel(freq)
s_param(:,:,index) = mutualInductorS(Lb,Lb,M,Z0,s(index));
end

Sobj = sparameters(s_param,freq);

Create Object for RF Budget

Use an rfwrite function to create a Touchstone® file from the sparameters object.
rfwrite(Sobj,'mutualInductor.s2p');

Create a nport object.

n = nport('mutualInductor.s2p');

6-243
6 RF Toolbox Examples

Provide the nport object as an input to rfbudget object.

b = rfbudget(n,2.1e9,-30,10e3);

Type this command at the MATLAB Command Window to open the mutual inductor as a S-parameter
element in the RF Budget Analyzer app.

show(b)

Using this method you can build your own components for RF budget analysis.

See Also

More About
• “Extract S-Parameters from Circuit” on page 6-236
• “De-Embedding S-Parameters” on page 6-110
• “Bisect S-Parameters of Cascaded Probes” on page 6-114

6-244
 Lossy Multiconductor Transmission Line Circuit

Lossy Multiconductor Transmission Line Circuit

This example shows how to analyze a lossy multiconductor transmission line circuit. The transmission
lines are modeled using RLGC distributed elements which are used in signal integrity analysis for
accurately capturing high-speed interconnect effects. You can analyze this circuit using RF Toolbox™
pwlresp function or using RF Blockset™ Circuit Envelope model. In this example you first compute
the time-domain response of the circuit excited by a periodic pulse signal and then you compare the
result to a circuit simulated using the RF Blockset Circuit Envelope simulation.

Figure 1: Interconnect circuit with lossy multiconductor transmission lines

Representation of Distributed Transmission Line Models with N-Port Elements

The circuit in this example contains two sets of lossy coupled transmission lines. The RLGC line
parameter matrices of the two lines are as follows [1].

Figure 2: RLGC line parameters for the 4-ports and 8-ports transmission lines

6-245
6 RF Toolbox Examples

Extract the S-parameters of each coupled transmission lines using rlgc2s function. Then represent
the S-parameters using an nport element.
freq = linspace(1e1,8e9,1001);
N = length(freq);
len1 = 0.1;
R1 = ones(2,2,N).*[75,15;15,75];
L1 = ones(2,2,N).*[494.6,63.3;63.3,494.6]*1e-9;
G1 = ones(2,2,N).*[0.1,-0.01;-0.01,0.1];
C1 = ones(2,2,N).*[62.8,-4.9;-4.9,62.8]*1e-12;

s_params1 = rlgc2s(R1,L1,G1,C1,len1,freq);
stlobj1 = sparameters(s_params1,freq);
nport_s1 = nport(stlobj1);

len2 = 0.1;
R2 = ones(4,4,N).*[50,10,1,0.0;10,50,10,1;1,10,50,10;0.0,1,10,50];
L2 = ones(4,4,N).*[494.6,63.3,7.8,0.0;63.3,494.6,63.3,7.8;7.8,63.3,494.6,63.3;0.0,7.8,63.3,494.6]
G2 = ones(4,4,N).*[0.1,-0.01,-0.001,0.0;-0.01,0.1,-0.01,-0.001;-0.001,-0.01,0.1,-0.01;0.0,-0.001,
C2 = ones(4,4,N).*[62.8,-4.9,-0.3,0.0;-4.9,62.8,-4.9,-0.3;-0.3,-4.9,62.8,-4.9;0.0,-0.3,-4.9,62.8]

s_params2 = rlgc2s(R2,L2,G2,C2,len2,freq);
stlobj2 = sparameters(s_params2,freq);
nport_s2 = nport(stlobj2);

Calculation of S-Parameter Object of Interconnect Circuit

Calculate the S-parameter object of the circuit given in Figure.1 by using the sparameters function.
ckt = circuit('interconnect');

add(ckt,[1 2],resistor(50))
add(ckt,[2 3 4 5],nport_s1,{'p1+' 'p2+' 'p3+' 'p4+'})
add(ckt,[3 0],resistor(75))
add(ckt,[4 0],capacitor(1e-12))
add(ckt,[5 0],resistor(100))
add(ckt,[4 6],resistor(25))
add(ckt,[7 0],resistor(50))
add(ckt,[4 8],resistor(25))
add(ckt,[9 0],resistor(100))
add(ckt,[6 7 8 9 10 11 12 13],nport_s2,{'p1+' 'p2+' 'p3+' 'p4+' 'p5+' 'p6+' 'p7+' 'p8+'})
add(ckt,[10 0],capacitor(2e-12))
add(ckt,[11 0],resistor(100))
add(ckt,[13 0],resistor(100))
add(ckt,[12 14],resistor(50))
add(ckt,[14 15],inductor(10e-9))
add(ckt,[15 0],capacitor(1e-12))

freqfit = linspace(1e1,8e9,1001)';
setports(ckt,[1 0],[15 0])
S_vout = sparameters(ckt,freqfit);

Generation of Rational Object of Multiconductor Transmission Line Circuit

Convert the transfer function of the S-parameter object of the circuit to a rational object over the
specified frequencies. Then calculate the frequency response using the freqresp function.
tfblkout = s2tf(S_vout,0,Inf,2);
fitblkout = rational(freqfit,tfblkout);

6-246
 Lossy Multiconductor Transmission Line Circuit

plot(freqfit/1e9,abs(tfblkout),'b',freqfit/1e9,abs(freqresp(fitblkout,freqfit)),'r:','LineWidth',
legend('Original data','Fitting result')
title('Frequency Response')
ylabel('Magnitude')
xlabel('Frequency (GHz)')

Computation of Transient Waveform Excited by Periodic Pulse Source

Apply a voltage source using a 1 V periodic pulse with a rise/fall time of 0.4 ns, a duration of 5 ns, and
a period of 1.6 ns. Then compute the time-domain response of the circuit by using the pwlresp
function.

SignalTime = [0,0.4,5.4,5.8]*1e-9;
SignalValue = [0,1,1,0];
Tsim = (0:1e-11:6e-8);
TP = 1.6e-8;
[WAVEOUT, tout] = pwlresp(fitblkout,SignalTime,SignalValue,Tsim,TP);

plot(tout,WAVEOUT, 'LineWidth',2)
legend('PWLRESP')
title('Output Waveform')
xlabel('Time (s)')
ylabel('Output (V)')
ylim([-0.05 0.38])

6-247
6 RF Toolbox Examples

Comparison of Time-Domain Response of Coupled Transmission Line Circuit Using RF

Blockset

A Simulink model is built to simulate the multiconductor transmission line circuit shown in Figure 1.
With the S-parameters calculated from previous session, the multiconductor transmission lines can
be represented by S-parameter blocks from RF Blockset Circuit Envelope Library. For instance, the 4-
port transmission lines are modeled by defining the S-parameter block mask as shown below, where
STL4p is a S-parameter object converted from reordering ports of S-parameter object stlobj1 with
snp2smp function. The output of the model is the time-domain response in the form of a passband
signal. For more information on passband signal implementation in the Circuit Envelope model, see
“Passband Signal Representation in Circuit Envelope” (RF Blockset).

6-248
 Lossy Multiconductor Transmission Line Circuit

To simulate the model:

1 Type open_system('simrf_coupled_pb') at the command prompt.

2 Select Simulation > Run.

After you finish simulating the model, compare the obtained transient waveform with that calculated
using RF Toolbox.

open_system('simrf_coupled_pb');
sim('simrf_coupled_pb');
plot(tout,WAVEOUT,CE_Data(:,1),CE_Data(:,2),'--','LineWidth',2)
legend('PWLRESP', 'Circuit Envelope')
title('Time-Domain Response of Lossy Coupled Transmission Lines')
xlabel('Time (s)')
ylabel('Output (V)')
ylim([-0.05 0.38])

6-249
6 RF Toolbox Examples

6-250
 Lossy Multiconductor Transmission Line Circuit

bdclose('simrf_coupled_pb');

References
[1] Tang, Tak K., and Michel Nakhla. “Analysis of High-Speed VLSI Interconnects Using the
Asymptotic Waveform Evaluation Technique.” IEEE Transactions on Computer-Aided Design
of Integrated Circuits and Systems 11, no. 3 (March 1992): 341–52. https://doi.org/
10.1109/43.124421.

See Also

Related Examples
• “Rational Fit S-Parameters” on page 6-214
• “Passband Signal Representation in Circuit Envelope” (RF Blockset)

6-251
6 RF Toolbox Examples

Richards-Kuroda Workflow for RF Filter Circuit

This example shows how to apply Richards-Kuroda workflow to an RF filter circuit.

Create a lowpass LC-Pi Chebyshev filter with the passband frequency of 1 GHz, passband attenuation
of 0.5 dB, and filter order of 5.

Fp = 1e9;
Ap = 0.5;
Ord = 5;
r = rffilter("FilterType","Chebyshev","ResponseType","Lowpass","Implementation","LC Pi","Passband
Fp,"PassbandAttenuation",Ap,"FilterOrder",Ord);

Plot the S21 parameter of the RF filter.

frequencies = linspace(0,5*Fp,1001);
rfplot(r, frequencies)

Convert the lumped-element filter to a distributed-element-based circuit at the operating frequency of

1 GHz using the richards function.

txCkt = richards(r,1e9)

txCkt =
circuit: Circuit element

6-252
 Richards-Kuroda Workflow for RF Filter Circuit

ElementNames: {'C_tx' 'L_tx' 'C_1_tx' 'L_1_tx' 'C_2_tx'}

Elements: [1x5 txlineElectricalLength]
Nodes: [0 1 2 3 4 5 6]
Name: 'unnamed'

Show the circuit properties in a table using the tableCircuitProperties function. You can find
the source code for this helper function in the Supporting Functions on page 6-260 section at the end
of this example.

tableCircuitProperties(txCkt,'Name','StubMode','Termination','Z0')

Name StubMode Termination Z0

__________ __________ ___________ ______

{'C_tx' } {'Shunt' } {'Open' } 29.312

{'L_tx' } {'Series'} {'Short'} 61.481
{'C_1_tx'} {'Shunt' } {'Open' } 19.679
{'L_1_tx'} {'Series'} {'Short'} 61.481
{'C_2_tx'} {'Shunt' } {'Open' } 29.312

Plot the S21 parameter of distributed-element-based filter circuit. The RF plot shows that distributed
and lumped filter behavior overlap close to the operating frequency and diverge significantly at
higher frequencies. This occurs due to the frequency-periodic nature of distributed elements.

rfplot(sparameters(txCkt, frequencies),2,1)

6-253
6 RF Toolbox Examples

The distributed-element-based circuit in txCkt circuit is not practical since all stubs essentially stem
from the same point in space. To separate the stubs and use only shunt stubs that are easier to
implement as microstrip lines, insert unit elements in a sequence and apply Kuroda's identities.

Add unit elements at the edges of the txCkt, operating at 1 GHz with the characteristic impedance
of 50 ohms. The edges of the circuit are port 1 of the first circuit element C_tx and port 2 of the last
circuit element C_2_tx.

txCkt_UE = insertUnitElement(txCkt,'C_tx',1,1e9,50);
txCkt_UE = insertUnitElement(txCkt_UE,'C_2_tx',2,1e9,50)

txCkt_UE =
circuit: Circuit element

ElementNames: {'C_tx_p1_elem_UE' 'C_tx' 'L_tx' 'C_1_tx' 'L_1_tx' 'C_2_tx' 'C_2_tx_p2_el

Elements: [1x7 txlineElectricalLength]
Nodes: [0 1 2 3 4 5 6 7 8]
Name: 'unnamed'

Show the circuit properties in a table.

tableCircuitProperties(txCkt_UE,'Name','StubMode','Termination','Z0')

Name StubMode Termination Z0

_____________________ ____________ _________________ ______

{'C_tx_p1_elem_UE' } {'NotAStub'} {'NotApplicable'} 50

{'C_tx' } {'Shunt' } {'Open' } 29.312
{'L_tx' } {'Series' } {'Short' } 61.481
{'C_1_tx' } {'Shunt' } {'Open' } 19.679
{'L_1_tx' } {'Series' } {'Short' } 61.481
{'C_2_tx' } {'Shunt' } {'Open' } 29.312
{'C_2_tx_p2_elem_UE'} {'NotAStub'} {'NotApplicable'} 50

Plot the S21 parameter of the new circuit txCkt_UE. The RF plot shows that the addition of unit
elements does not change the power magnitude behavior of the circuit, and thus this RF plot shows
the same characteristics of the S21 parameter as the distributed-element-based filter circuit.

rfplot(sparameters(txCkt_UE, frequencies),2,1)

6-254
 Richards-Kuroda Workflow for RF Filter Circuit

Apply Kuroda's identities to the first two and last two elements of the circuit. For more information,
see “Kuroda's Transformation”.

txCkt_Kur = kuroda(txCkt_UE,'C_tx_p1_elem_UE','C_tx');
txCkt_Kur = kuroda(txCkt_Kur,'C_2_tx','C_2_tx_p2_elem_UE')

txCkt_Kur =
circuit: Circuit element

ElementNames: {'Kuroda2_R2L_of_C_tx_p1_elem_UE' 'Kuroda2_R2L_of_C_tx' 'L_tx' 'C_1_tx' 'L_

Elements: [1x7 txlineElectricalLength]
Nodes: [0 1 2 3 4 5 6 7 8]
Name: 'unnamed'

Show the circuit properties in a table.

tableCircuitProperties(txCkt_Kur,'Name','StubMode','Termination','Z0')

Name StubMode Termination Z0

____________________________________ ____________ _________________ ______

{'Kuroda2_R2L_of_C_tx_p1_elem_UE' } {'Series' } {'Short' } 31.521

{'Kuroda2_R2L_of_C_tx' } {'NotAStub'} {'NotApplicable'} 18.479
{'L_tx' } {'Series' } {'Short' } 61.481
{'C_1_tx' } {'Shunt' } {'Open' } 19.679
{'L_1_tx' } {'Series' } {'Short' } 61.481

6-255
6 RF Toolbox Examples

{'Kuroda1_L2R_of_C_2_tx' } {'NotAStub'} {'NotApplicable'} 18.479

{'Kuroda1_L2R_of_C_2_tx_p2_elem_UE'} {'Series' } {'Short' } 31.521

Plot the S21 parameter of txCkt_Kur. The RF plot shows that, as expected, applying Kuroda's
identities does not change the behavior of the circuit (as opposed to adding a unit element, applying
Kuroda's identities retains both magnitude and phase behavior of the circuit).

rfplot(sparameters(txCkt_Kur, frequencies),2,1)

Add a unit elements at the edges of this circuit operating at 1 GHz with the characteristic impedance
of 50 ohms.

txCkt_UE2 = insertUnitElement(txCkt_Kur,'Kuroda2_R2L_of_C_tx_p1_elem_UE',1,1e9,50);
txCkt_UE2 = insertUnitElement(txCkt_UE2,'Kuroda1_L2R_of_C_2_tx_p2_elem_UE',2,1e9,50)

txCkt_UE2 =
circuit: Circuit element

ElementNames: {1x9 cell}

Elements: [1x9 txlineElectricalLength]
Nodes: [0 1 2 3 4 5 6 7 8 9 10]
Name: 'unnamed'

Show the circuit properties in a table.

tableCircuitProperties(txCkt_UE2,'Name','StubMode','Termination','Z0')

6-256
 Richards-Kuroda Workflow for RF Filter Circuit

Name StubMode Termination Z0

_______________________________________________ ____________ _________________ _____

{'Kuroda2_R2L_of_C_tx_p1_elem_UE_p1_elem_UE' } {'NotAStub'} {'NotApplicable'} 5

{'Kuroda2_R2L_of_C_tx_p1_elem_UE' } {'Series' } {'Short' } 31.52
{'Kuroda2_R2L_of_C_tx' } {'NotAStub'} {'NotApplicable'} 18.47
{'L_tx' } {'Series' } {'Short' } 61.48
{'C_1_tx' } {'Shunt' } {'Open' } 19.67
{'L_1_tx' } {'Series' } {'Short' } 61.48
{'Kuroda1_L2R_of_C_2_tx' } {'NotAStub'} {'NotApplicable'} 18.47
{'Kuroda1_L2R_of_C_2_tx_p2_elem_UE' } {'Series' } {'Short' } 31.52
{'Kuroda1_L2R_of_C_2_tx_p2_elem_UE_p2_elem_UE'} {'NotAStub'} {'NotApplicable'} 5

Plot the S21 parameter of txCkt_UE2.

rfplot(sparameters(txCkt_UE2, frequencies),2,1)

Apply Kuroda's identities to the first, second, second to last, and last pair of elements in the circuit.

txCkt_Kur2 = kuroda(txCkt_UE2,'Kuroda2_R2L_of_C_tx_p1_elem_UE_p1_elem_UE','Kuroda2_R2L_of_C_tx_p1
txCkt_Kur2 = kuroda(txCkt_Kur2,'Kuroda2_R2L_of_C_tx','L_tx');
txCkt_Kur2 = kuroda(txCkt_Kur2,'Kuroda1_L2R_of_C_2_tx_p2_elem_UE','Kuroda1_L2R_of_C_2_tx_p2_elem_
txCkt_Kur2 = kuroda(txCkt_Kur2,'L_1_tx','Kuroda1_L2R_of_C_2_tx')

txCkt_Kur2 =
circuit: Circuit element

ElementNames: {1x9 cell}

6-257
6 RF Toolbox Examples

Elements: [1x9 txlineElectricalLength]

Nodes: [0 1 2 3 4 5 6 7 8 9 10]
Name: 'unnamed'

Show the circuit properties in a table.

tableCircuitProperties(txCkt_Kur2,'Name','StubMode','Termination','Z0')

Name StubMode Terminat

______________________________________________________________ ____________ ___________

{'Kuroda1_R2L_of_Kuroda2_R2L_of_C_tx_p1_elem_UE_p1_elem_UE' } {'Shunt' } {'Open'

{'Kuroda1_R2L_of_Kuroda2_R2L_of_C_tx_p1_elem_UE' } {'NotAStub'} {'NotApplic
{'Kuroda1_R2L_of_Kuroda2_R2L_of_C_tx' } {'Shunt' } {'Open'
{'Kuroda1_R2L_of_L_tx' } {'NotAStub'} {'NotApplic
{'C_1_tx' } {'Shunt' } {'Open'
{'Kuroda2_L2R_of_L_1_tx' } {'NotAStub'} {'NotApplic
{'Kuroda2_L2R_of_Kuroda1_L2R_of_C_2_tx' } {'Shunt' } {'Open'
{'Kuroda2_L2R_of_Kuroda1_L2R_of_C_2_tx_p2_elem_UE' } {'NotAStub'} {'NotApplic
{'Kuroda2_L2R_of_Kuroda1_L2R_of_C_2_tx_p2_elem_UE_p2_elem_UE'} {'Shunt' } {'Open'

Plot the S21 parameter of txCkt_Kur2.

rfplot(sparameters(txCkt_Kur2, frequencies),2,1)

Create a microstrip transmission line. Then use the realize function to realize a circuit containing
the electrical-length-based transmission line txCkt_Kur2.

6-258
 Richards-Kuroda Workflow for RF Filter Circuit

txln = txlineMicrostrip('Height',0.0015748,'EpsilonR',4.6, 'LossTangent',0.026,'SigmaCond',596000

txMS = realize(txCkt_Kur2,txln)

txMS =
circuit: Circuit element

ElementNames: {1x9 cell}

Elements: [1x9 txlineMicrostrip]
Nodes: [0 1 2 3 4 5 6 7 8 9 10]
Name: 'unnamed'

Show the circuit properties in a table.

tableCircuitProperties(txMS,'Name','StubMode','Termination','LineLength','Width')

Name StubMode Ter

___________________________________________________________________ ____________ ______

{'txlMs_of_Kuroda1_R2L_of_Kuroda2_R2L_of_C_tx_p1_elem_UE_p1_elem_'} {'Shunt' } {'Open

{'txlMs_of_Kuroda1_R2L_of_Kuroda2_R2L_of_C_tx_p1_elem_UE' } {'NotAStub'} {'NotA
{'txlMs_of_Kuroda1_R2L_of_Kuroda2_R2L_of_C_tx' } {'Shunt' } {'Open
{'txlMs_of_Kuroda1_R2L_of_L_tx' } {'NotAStub'} {'NotA
{'txlMs_of_C_1_tx' } {'Shunt' } {'Open
{'txlMs_of_Kuroda2_L2R_of_L_1_tx' } {'NotAStub'} {'NotA
{'txlMs_of_Kuroda2_L2R_of_Kuroda1_L2R_of_C_2_tx' } {'Shunt' } {'Open
{'txlMs_of_Kuroda2_L2R_of_Kuroda1_L2R_of_C_2_tx_p2_elem_UE' } {'NotAStub'} {'NotA
{'txlMs_of_Kuroda2_L2R_of_Kuroda1_L2R_of_C_2_tx_p2_elem_UE_p2_ele'} {'Shunt' } {'Open

Plot the S21 parameter of txMS. The S21 parameter of the microstrip-based circuit deviates from the
S21 of txCkt_Kur2. This is due to the practical realization considerations. Remove losses to improve
the agreement between the two plots.

rfplot(sparameters(txMS, frequencies),2,1)
set(gca,'YLim',[-20 0]);

6-259
6 RF Toolbox Examples

Supporting Functions

tableCircuitProperties:

type("tableCircuitProperties.m")

function tableCircuitProperties(ckt,varargin)
c = cell(numel(ckt.Elements),nargin-1);
for col = 1:nargin-1
c(:,col) = {ckt.Elements.(varargin{col})};
end
disp(cell2table(c, 'VariableNames',varargin));
end

See Also
richards | kuroda | insertUnitElement | realize | txlineElectricalLength

6-260
 Design RF Chain Using RF Antenna Object

Design RF Chain Using RF Antenna Object

This example shows how to design an RF chain using an rfantenna object.

RF Antenna Transmitter and Receiver Elements

Create an RF antenna transmitter element. The properties of the RF antenna transmitter are as
shown.

Txant = rfantenna

Txant =
rfantenna: Antenna element

Name: 'Antenna'
Type: 'Transmitter'
Gain: 0
Z: 50

Create an RF antenna receiver element. The properties of the RF antenna receiver are as shown.

Rxant = rfantenna("Type","Receiver")

Rxant =
rfantenna: Antenna element

Name: 'Antenna'
Type: "Receiver"
Gain: 0
Z: 50
TxEIRP: -30
PathLoss: 0

The TxEIRP property of the RF antenna receiver element is the effective isotropic radiated power
(EIRP) from the transmitter to the receiver. This is essentially a power value falling on the receiver.
The EIRP also takes into account the path loss. Path loss is specified in the PathLoss property of the
receiver. Transmitter EIRP is calculated using this equation:

TxEIRP = Path Loss + Gain

Use RF Antenna Receiver Element in RF Budget Chain

Calculate the RF budget of a series of RF elements by tying this command at the command line at the
input frequency of 2.1 GHz, available input power of –30 dBm, and bandwidth of 10 MHz. The RF
antenna receiver element must be the first element in the RF chain.

a = amplifier;
m = modulator;
rfb = rfbudget([Rxant a m],2.1e9,-30,100e6)

6-261
6 RF Toolbox Examples

MATLAB displays the warning message because the available input power is replaced by TxEIRP, the
EIRP of the of the receiver.

Open the RF Budget Analyzer app to visualize RF budget chain using the show command at the
command line.

show(rfb)

6-262
 Design RF Chain Using RF Antenna Object

The Available Input Power option is dimmed in the app when you add a receiver element. By
default, the TxEIRP property has the same value as the Available Input Power option.

Use RF Antenna Transmitter Element in RF Budget Chain

Calculate the RF budget of a series of RF elements by using the RF antenna transmitter as one of the
elements. The RF antenna transmitter element must be the last element in the RF chain.

a = amplifier;
m = modulator;
rfb = rfbudget([a m Txant],2.1e9,-30,100e6)

6-263
6 RF Toolbox Examples

Open the RF Budget Analyzer app to visualize RF budget chain using the show command at the
command line.

show(rfb)

6-264
 Design RF Chain Using RF Antenna Object

Design RF Receiver Antenna Using Antenna Designer App

By default, the Antenna Source property of the RF antenna receiver element in the app set to
Isotropic Receiver. Set the Antenna Source to Antenna Designer, which gives you the
option of designing the antenna in the Antenna Designer app of the Antenna Toolbox. To use the
Antenna Designer app or the Antenna Object, you need Antenna Toolbox™ license.

6-265
6 RF Toolbox Examples

Click Create Antenna in the Element Parameters pane. The Antenna Designer app opens. Select
New to explore the antenna library. This example uses a dipoleFolded antenna element with a
center frequency of 2.1 GHz. Select the Folded element from the Antenna Gallery, set the Design
Frequency to 2.1 MHz, and click Accept.

Select Update Element to update the Antenna element in the RF Budget Analyzer app. Select OK
in the Confirm Update dialog box.

6-266
 Design RF Chain Using RF Antenna Object

After importing the design to the RF Budget Analyzer app, select Apply.

6-267
6 RF Toolbox Examples

Design RF Receiver Antenna Using Antenna Object

Create a folded dipole antenna at 2.1 GHz in the MATLAB workspace.

antenna = design(dipoleFolded,2.1e9)

Set the Antenna Source to Antenna Object in the RF Budget Analyzer app to design an antenna
using antenna object.

6-268
 Design RF Chain Using RF Antenna Object

Select the antenna in the Antenna Object field and click Apply.

6-269
6 RF Toolbox Examples

Export Your RF Transmitter and Receiver Chain

Export the RF transmitter chain to RF Blockset™ by selecting the Export to Blockset option in the
Export drop-down list.

6-270
 Design RF Chain Using RF Antenna Object

Now export the RF receiver chain to RF Blockset by selecting the Export to Blockset option in the
Export drop-down list.

6-271
6 RF Toolbox Examples

The exported RF Blockset model shows that the RF transmitter chain does not have TxEIRP and
PathLoss as input properties, and similarly, the RF receiver chain does not have EIRP and
Directivity as its outputs.

When you export an RF receiver, the exported RF Blockset model only shows the values of the Gain
and Impedance parameters, whereas, in case of an RF transmitter, all the transmit antenna
parameters of an rfantenna object are exported to the model.

Copyright 2020 The MathWorks, Inc.

See Also
RF Budget Analyzer

Related Examples
• “Superheterodyne Receiver Using RF Budget Analyzer App” on page 6-2
• “RF Budget Harmonic Balance Analysis of Low-IF Receiver, IP2 and NF” on page 6-225

6-272
 Circuit Envelope Simulation at MATLAB Command Line

Circuit Envelope Simulation at MATLAB Command Line

You can perform circuit envelope simulation on your RF system either in Simulink® or in MATLAB®.
This example shows how to perform circuit envelope simulation at the MATLAB command line using
the rfsystem System object™. The circuit envelope simulation technique allows you to perform
time-varying harmonic balance analysis of an RF system around multiple carrier frequencies. You can
use this technique for simulating RF systems operating with narrowband signals. These narrowband
signals are simulated around carrier frequencies that are either harmonic multiples of fundamental
carrier frequencies or intermodulation frequencies that result from mixing products and nonlinear
effects. This added level of fidelity allows you to predict out-of-band spectral regrowth or take into
account the impact of interferers. For more information, see “Circuit Envelope Basics” (RF Blockset).

In this example, you first design your RF system using an rfbudget object or the RF Budget
Analyzer app and then you will export your system to an rfsystem System object. You then provide
provide a stimulus to this System object, set its sample rate, and view the associated RF Blockset™
model. You can also modify this model to increase the level of modeling fidelity.

Design an RF system. This example uses an RF transmitter designed in the “RF Transmitter System
Analysis” example. Type rfBudgetAnalyzer("rftxr.mat") command at the command line to
visualize the RF system in the RF Budget Analyzer app.

6-273
6 RF Toolbox Examples

In the app, select Export and then select RF system to export this RF transmitter system to a
rfsystem System object. This imported rfsystem System object is stored in a variable, rfs. For
this example, the rfs variable is saved in rfsystem_txmr.mat data file using open_system
(rfs) and save_system(rfs,'rfsystem_txmr') commands.

Load the rfs variable from the rfsystem_txmr.mat data file.

S = load('rfsystem_txmr.mat','rfs');
rfs = S.rfs;

Specify an input time-domain signal for the RF system.

in = [1e-3*ones(8,1); zeros(8,1)] .* ones(1,10);
in = in(:);

Calculate the output time-domain signal of the RF system.

out = rfs(in);
out = [out; rfs(in)];

Specify the sample time of the RF system.

t = rfs.SampleTime*(0:length(out)-1);

Plot the simulated output.

plot(t,[in; in],'-o',t,abs(out),'-+')
grid on

6-274
 Circuit Envelope Simulation at MATLAB Command Line

Release system resources and turn off fast restart.

release(rfs)

Open an RF Blockset model of the RF system using the open_system object function.

open_system(rfs)

You can also expand your system to a multiple input single output (MISO) or a single input multiple
output (SIMO) system. For more information, see “Model MISO Receiver and SIMO Transmitter
Systems”.

See Also
open_system | load_system

Related Examples
• “RF Receiver Modeling for LTE Reception” (RF Blockset)

6-275
6 RF Toolbox Examples

Design Matching Network Using Lumped Components from

Modelithics Library

This example shows you how to:

• Design a matching network for a reference antenna using ideal components from RF Toolbox™
and real-world components from Modelithics SELECT+ Library™. Ideal components do not take
parasitic effects into account. To include these effects in your simulation, you need to build the
matching network with real-world components.
• Analyze the ideal and the real-world matching network.
• Compare the performance of the ideal and the real-world matching network with respect to a
reference antenna.

The presence of parasitics changes the response of the matching network with real-world lumped
components. The non-ideal behavior results from variations in the material properties of the substrate
and packing material, solder and pad properties, and orientation.

Video Walkthrough

For a walkthrough of the example, play the video.

Create Inset-Fed Patch Antenna

The first step before designing the matching network for the reference antenna is to design the
reference antenna itself. This example uses an inset-fed patch antenna as a reference antenna. Create
simple inset-fed patch antenna using the design function from Antenna Toolbox™ implemented on a
PCB. Set the antenna dimensions and use an FR4 substrate in this design. And the set the substrate
thickness to 0.0014 m.

6-276
 Design Matching Network Using Lumped Components from Modelithics Library

antennaObject = design(patchMicrostripInsetfed, 2400*1e6);

antennaObject.Length = 0.0265;
antennaObject.Width = 0.0265;
antennaObject.Height = 0.0014;
antennaObject.Substrate.Name = 'FR4';
antennaObject.Substrate.EpsilonR = 4.8;
antennaObject.Substrate.LossTangent = 0.026;
antennaObject.Substrate.Thickness = 0.0014;
antennaObject.FeedOffset = [-0.02835, 0];
antennaObject.StripLineWidth = 0.0016223;
antennaObject.NotchLength = 0.0037853;
antennaObject.NotchWidth = 0.002839;
antennaObject.GroundPlaneLength = 0.0567;
antennaObject.GroundPlaneWidth = 0.0567;

Visualize Inset-Fed Patch Antenna

Use the show function to visualize the structure of this patch antenna.

figure;
show(antennaObject)

Analyze Inset-Fed Patch Antenna

Perform full-wave analysis on the inset-fed patch antenna over 2.3–2.5 GHz and save the S-parameter
results to a MAT file. Load the MAT file to view the impedance and reflection coefficient response.

6-277
6 RF Toolbox Examples

load InsetPatchAntenna.mat
plotFrequency = 2400*1e6;
freqRange = linspace(2.3e9, 2.5e9, 11);

Plot the impedance of the inset-fed patch antenna.

figure;
impedance(antennaObject, freqRange);

Plot the reflection coefficient response of the inset-fed patch antenna.

figure;
s = sparameters(antennaObject, freqRange);
rfplot(s)

6-278
 Design Matching Network Using Lumped Components from Modelithics Library

Build Rational Model

Build a rational model for the antenna S-parameter data. This allows you to refine the frequency
points in the analysis range and not simulate the antenna again during full-wave analysis.

s_rat = rational(s);
[resp,~] = freqresp(s_rat,freqRange);
hold on
plot(freqRange/1e9,20*log10(abs(resp)))
title('Antenna Reflection Coefficient')
legend('Full-Wave','Rational Model')

6-279
6 RF Toolbox Examples

Design Matching Network

Load the S-parameter data from the workspace to the Matching Network Designer app. Set the
matching frequency to 2.35 GHz and the network topology to L-Topology. Once the app generates the
networks, select series C, shunt L from the network list and export the network. The app saves the
network to MAT file.

6-280
 Design Matching Network Using Lumped Components from Modelithics Library

To build the matching network, convert the S-parameter data to an nport object and add it to a
circuit. Assign ports to the circuit for RF analysis.
ant = nport(s);
load mnapp_LTopo_CserLsh.mat
ckt_lumped = circuit;
add(ckt_lumped,[1 2 0 0],mnckt)
add(ckt_lumped,[2 0],ant)
setports(ckt_lumped,[1 0])

Perform S-Parameters Frequency Sweep

Select 100 points in the analysis frequency range and analyze the matching network response with
the antenna S-parameters as the load. Overlay the antenna-only port reflection coefficient over this
response. This shifts the antenna response in the lower band to 2.35 GHz.
freqRange = linspace(2.3e9, 2.5e9, 100);
lumped_s = sparameters(ckt_lumped,freqRange);
[resp,freq] = freqresp(s_rat,freqRange);
figure
rfplot(lumped_s,1,1)
hold on
plot(freqRange/1e9,20*log10(abs(resp)))
legend('ANT+Matching N/W','ANT','Location','Best')
title('Reflection Coefficient of Antenna and Matching Network')

6-281
6 RF Toolbox Examples

Build Matching Network with Non-Ideal Lumped Component Models

Select the non-ideal lumped components from Modelithics Select+ Library. You must have the
Modelithics Select+ Library license to run the following code. For more information, see Modelithics
SELECT+ Library for MATLAB.

Create RF Component using Modelithics SELECT+ Library

Set up the Modelithics Select+ Library by specifying the full path to the library.

mdlxSetup('C:\mdlx_library\SELECT')

Create the Modelithics library object.

mdlx = mdlxLibrary;

Search the library for a 1.6117 pF capacitor mounted on a 59 mil FR4 substrate. Note that
Modelithics components are substrate scalable, and that 59 mil is chosen to match the antenna
substrate thickness of 1.4mm.

search(mdlx,'FR4Standard59mil',Type='Capacitors',Value=1.6117e-12)

6-282
 Design Matching Network Using Lumped Components from Modelithics Library

Search the library for a 2.6611 nH inductor mounted on a 59 mil FR4 substrate.

search(mdlx,'FR4Standard59mil',Type='Inductors',Value=2.6611e-9)

6-283
6 RF Toolbox Examples

Create an array of Modelithics capacitors.

cList = search(mdlx,'FR4Standard59mil',Type='Capacitors',Value=1.6117e-12);

Create an array of Modelithics inductors.

lList = search(mdlx,'FR4Standard59mil',Type='Inductors',Value=2.6611e-9);

Create Matching Network with Modelithics Components

Create a matching network with L-Topology, series C, and shunt L using the first element in the array
of Modelithics capacitors and inductors. Most Modelithics lumped components have two ports.
mdlxckt = circuit;
add(mdlxckt,[1 2 0 0],cList(1));
add(mdlxckt,[2 0 0 0],lList(1));
setports(mdlxckt,[1 0],[2 0]);

As with the matching network with ideal lumped components, add the matching network with the
Modelithics lumped components and the S-parameters of the inset-fed patch antenna to a circuit.
Assign ports to the circuit for RF analysis.
mdlxckt_lumped = circuit;
add(mdlxckt_lumped,[1 2 0 0],mdlxckt)
add(mdlxckt_lumped,[2 0],nport(s))
setports(mdlxckt_lumped,[1 0])

6-284
 Design Matching Network Using Lumped Components from Modelithics Library

Analyze Matching Network and Antenna

Analyze the matching network response with the antenna S-parameters as the load.
mdlxlumped_s = sparameters(mdlxckt_lumped,freqRange);

Compare Ideal and Non-Ideal Reflection Coefficients

Overlay the reflection coefficient from the antenna S-parameters with the ones using ideal matching
network and the one with real-world components from the Modelithics Select+ Library.
figure
rfplot(mdlxlumped_s,1,1)
hold on
plot(freqRange/1e9,20*log10(abs(resp)))
rfplot(lumped_s,1,1)
legend('ANT+Modelithics Components','ANT','ANT+Ideal Components','Location','Best')
title('Reflection Coefficient of Antenna and Ideal/Non-ideal Matching Network')

Inspect Component Data Sheets

The two L and C components are available only in a limited set of discrete values. Use the
datasheet function to inspect their respective data sheets to identify the values that are closer to
the nominal values identified before.
datasheet(cList(1));
datasheet(lList(1));

6-285
6 RF Toolbox Examples

Discrete component values from the data sheets are shown in this figure.

Build Matching Network Using Discrete Component Values

Build a new matching network, and try a few of the discrete values close to the nominal values. The
best working values for a matching network operating at 2.35 GHz are 1.3 pF and 1.8 nH. The
discrete values are slightly different from the nominal values due to parasitic effects.

c = clone(cList(1));
c.Value = 1.3e-12;
l = clone(lList(1));
l.Value = 1.8e-9;
mdlxcktD = circuit;
add(mdlxcktD,[1 2 0 0],c);
add(mdlxcktD,[2 0 0 0],l);
add(mdlxcktD,[2 0],nport(s))
setports(mdlxcktD,[1 0]);
mdlxlumpeddiscrete_s = sparameters(mdlxcktD,freqRange);

Overlay the reflection coefficient from the ideal matching network, the one with real-world
components from the Modelithics Select+ Library using nominal values, and the one with real-world
components from the Modelithics Select+ Library using discrete values.

figure
rfplot(mdlxlumpeddiscrete_s,1,1)
hold on
rfplot(mdlxlumped_s,1,1);
rfplot(lumped_s,1,1)
legend('ANT+Modelithics Components Discrete Values', ...
'ANT+Modelithics Components Nominal Values', ...
'ANT+Ideal Components','Location','Best')
title('Reflection Coefficient of Antenna and Ideal/Non-ideal Matching Network')

The resulting matching network now works at the target frequency.

6-286
 Design Matching Network Using Lumped Components from Modelithics Library

See Also
mdlxLibrary | search | mdlxPart | mdlxSetup | datasheet

Related Examples
• “Impedance Matching of Small Monopole Antenna” on page 6-166

6-287
6 RF Toolbox Examples

Perform Frequency Planning to Find Spur-Free IF Bandwidths

This example shows you how find spur-free IF bandwidths of a mixer. In this example, you wiil:

• Perform frequency planning to identify spur-free intermediate frequencies (IF) bandwidths.

• Visualize the mixer spurs to observe the effect of interference when selecting the desired IF.
• Export the cascaded mixer chain to the RF Budget Analyzer app and vary the IF and RF
bandwidths to analyze the budget output.

Frequency planning in mixers involves the careful selection and arrangement of input and output
frequencies to ensure optimal performance. This process aims to minimize interference, spurious
signals, and other undesirable effects that can arise due to the mixing of frequencies within the mixer
circuit. By strategically planning the which frequencies to use within the mixer, engineers can
mitigate issues and optimize the desired output signals.

Define Mixer Configuration

This example considers a dual-conversion receiver operating in the 5275 to 5525 MHz band, as
depicted in the figure. To find a spur-free IF for this receiver, the receiver must be down-converted to
dual IF bands. To find an IF center frequency that is spur-free for all IF bands, specify the RF center
frequency, the RF bandwidth, and the IF bandwidth that correspond with the input RF signal.

Define the RF center frequency for mixer 1.

RFCF1 = 5400e6;

Define the RF bandwidth (RFBW) and IF bandwidth (IFBW) of the mixer 1.

RFBW1 = 250e6;
IFBW1 = 50e6;

6-288
 Perform Frequency Planning to Find Spur-Free IF Bandwidths

Define the RFBW and IFBW of the mixer 2.

RFBW2 = 150e6;
IFBW2 = 25e6;

Specify the IMT measured data for each RF band.

IMT1 = [99 0 21 17 26;

11 0 29 29 63;
60 48 70 86 41;
90 89 74 68 87;
99 99 95 99 99];

IMT2 = [99 1 9 12 15;

20 0 26 31 48;
55 70 51 70 53;
85 90 60 70 94;
96 95 94 93 92];

Assign the IMT data to mixers by using the mixerIMT object.

mixer1 = mixerIMT("ConverterType","Down","IMT",IMT1);
mixer2 = mixerIMT("ConverterType","Down","IMT",IMT2);

Perform Frequency Planning for First Mixer Element

Plot frequency planning output of the first mixer with 250 MHz as the RFBW and 50 MHz as the
IFBW. The plot displays the spur-free regions or frequency bands from which you can select the IF.

rfplot(mixer1,'Frequency',RFCF1,'RFBW',RFBW1,'IFBW',IFBW1)
xlim([0.1 2]*1e9) % X-limit is set to 2 GHz

6-289
6 RF Toolbox Examples

Select an IF frequency from the frequency planning graph that falls within the spur-free zones. This
example presents two cases: one where you select a signal that does not fall in the spur-free zones
and visualize the interference effect on the desired signal, and another where you select a signal from
the in spur-free zones and visualize the interference free desired signal.

Case 1: Selected Signal Lies Outside Spur-Free Zone

Consider an IF of 1350 MHz that falls that does not fall in the spur-free zones. This IF value implies
an LO1 of as 4050 MHz assuming a low-side injection.

Given this data, plot the mixer spurs using rfplot function.

mixer1.LO = 4050e6;
rfplot(mixer1,'Frequency',RFCF1)
ylim([-100 0])

6-290
 Perform Frequency Planning to Find Spur-Free IF Bandwidths

This plot shows that selecting an IF that does not fall in the spur-free zone results in interference
from spurs affecting the desired signal.

Case 2: Selected Signal Lies Within Spur-Free Zone

Considering a low-side injection mixer, this example selects an IF of 1600 MHz for mixer 1, resulting
in an LO frequency of 3800 MHz.

Plot the mixer spur graph for mixer 1 with LO frequency set to 3800 MHz. This plot locates the
signals and spurs in power (dBm) versus frequency (GHz).

mixer1.LO = 3800e6;
rfplot(mixer1,'Frequency',RFCF1)
ylim([-100 0])

6-291
6 RF Toolbox Examples

This plot shows that the desired signal is free from interference.

Perform Frequency Planning for Second Mixer Element

Perform frequency planning for the second mixer with an RFBW of 150 MHz and an IFBW of 25 MHz.
The RF center frequency of the second mixer is the desired IF of mixer 1, which is 1600 MHz.

rfplot(mixer2,'Frequency',1600e6,'RFBW',RFBW2,'IFBW',IFBW2)
xlim([0.1 1]*1e9)

6-292
 Perform Frequency Planning to Find Spur-Free IF Bandwidths

Provide the IF of the first mixer as an input to the second mixer. Based on the frequency planning
plot, select 70 MHz as IF2. This results in 1670 MHz as LO2, and the mixer is down converted.

Visualize the mixer spurs of mixer 2 using the rfplot function.

mixer1.LO = 1670e6;
rfplot(mixer2,'Frequency',1600e6)
ylim([-100 0])

6-293
6 RF Toolbox Examples

The plot shows that the desired signal is free of interference.

Perform Frequency Planning Using RF Budget Analyzer App

Create the budget object using the rfbudget function. Cascade the mixers in series.

b = rfbudget([mixer1,mixer2],RFCF1,-30,RFBW1);

You can use the RF Budget Analyzer app to perform frequency planning. To export the budget object
b to the app, type the show(b) command in the command line.

show(b)

In the app, set the Input Frequency to 5.4 GHz and set the Signal Bandwidth to 250 MHz. The
Signal Bandwidth represents the RFBW of the first mixer. In the Element Parameters pane, click the
Plot Frequency Plan button. In the plot tab, set the IFBW to 50e6. If you do not specify IFBW, the
app assigns IFBW to Signal Bandwidth.

6-294
 Perform Frequency Planning to Find Spur-Free IF Bandwidths

To visualize the spurs of the mixer 1, in the Element Parameters pane, click the Plot Spurs button.

6-295
6 RF Toolbox Examples

To set the RFBW of the second mixer, adjust the Signal Bandwidth to 150 MHz in the System
Parameter section. Then, select the second mixer element from the Cascade pane, and in the
Element Parameters pane, click the Plot Frequency Plan button. In the plot, set the IFBW to 25
MHz and click Enter.

6-296
 Perform Frequency Planning to Find Spur-Free IF Bandwidths

See Also
mixerIMT | rfplot

Related Examples
• “Visualize Mixer Spurs” on page 6-95

6-297
6 RF Toolbox Examples

Characterize High-Frequency Data and Interconnects Using

IEEE P370 Standard

This example shows you how to use the IEEE P370 source code, its associated data files, and RF
Toolbox™ functions to implement the following recommended practices for de-embedding the effect
of fixtures when measuring a device under test (DUT).

• S-parameters quality check: Perform quality checks on the S-parameters of the measured data.
• Self de-embedding of 2X-Thru: Verify that the behavior of the self-de-embedded fixtures is
transparent.
• Compare the TDR of the fixture model to the FIX-DUT-FIX: Verify the assumption that the
fixture model is identical to the fixture attached to the DUT.
• Determine the similarity of S-parameters: Qualify and quantify the accuracy of the de-
embedded DUT against the expected DUT.

For more information on the standard practices, see [1].

Introduction

De-embedding is the mathematical process of removing the unwanted effects of test fixtures that
connect test equipment to the DUT in a measurement setup. De-embedding characterizes the effects
of fixtures and uses the calculated calibration data to compensate for undesired distortions.

This example uses the following abbreviations:

• 2X-Thru: Connected replicas of the left and right fixtures, used for characterizing the effects of
the fixtures
• FIX-DUT-FIX: Composite structure of the test fixtures and the DUT during measurement

Setup

RF Toolbox ships IEEE P370 source files. Therefore, copy the contents from the +ieee370 package
into a new folder in the MATLAB® working directory. Measured data files are included with this
example and do not need to be copied.
ieee370Src = fullfile(matlabroot,"toolbox","rf","thirdparty","+ieee370");
ieee370Dir = fullfile(pwd,"ieee370");
copyfile(ieee370Src,ieee370Dir,"f")
addpath(ieee370Dir)

Extract the S-parameters of the 2x-Thru for de-embedding.

sFIX = sparameters("case_01_2xThru.s2p");
sdataFIX = sFIX.Parameters;
freqFIX = sFIX.Frequencies;
nptsFIX = length(freqFIX);
portNumFIX = size(sdataFIX,1);

Extract the S-parameters of the FIX-DUT-FIX for de-embedding.

sFDF = sparameters("case_01_F-DUT1-F.s2p");
sdataFDF = sFDF.Parameters;
freqFDF = sFDF.Frequencies;

6-298
 Characterize High-Frequency Data and Interconnects Using IEEE P370 Standard

nptsFDF = length(freqFDF);
portNumFDF = size(sdataFDF,1);

Define the S-parameters of the DUT, which you use as a reference baseline to quantify the accuracy
of the calculated DUT at the end of this example.

sDUT = sparameters("case_01_DUT1.s2p");
sdataDUT = sDUT.Parameters;
freqDUT = sDUT.Frequencies;

Suppress Warnings

The third-party IEEE P370 source files issue warnings. Suppress the warnings for the sake of keeping
this example simple.

warnStruct = warning("off","MATLAB:colon:operandsNotRealScalar");
origWarnStatus = warnStruct.state;

S-Parameters Quality Check

Quantify the causality, reciprocity, and passivity of the 2X-Thru and FIX-DUT-FIX S-parameters to
gauge the quality of the measurement. Perform this quality check using the frequency domain and
time domain functions provided by IEEE P370.

Use IEEE P370 Frequency Domain Function

To get the causality, reciprocity, and passivity metrics of the S-parameters, call the function
ieee370QualityCheckFrequencyDomain, which internally calls the IEEE P370 source function
qualityCheckFrequencyDomain.

First, call the function on the measured 2X-Thru data.

[causalityMetricFIX_FD,reciprocityMetricFIX_FD,passivityMetricFIX_FD] = ...
ieee370QualityCheckFrequencyDomain(sdataFIX,nptsFIX,portNumFIX);

Next, call the function on the measured FIX-DUT-FIX data.

[causalityMetricFDF_FD,reciprocityMetricFDF_FD,passivityMetricFDF_FD] = ...
ieee370QualityCheckFrequencyDomain(sdataFDF,nptsFDF,portNumFDF);

Now, verify that the values of the metric indicate that the measured data pass the quality check. If the
measurements fail, you must measure the structures again after recalibrating the measurement
setup.

structure = ["2X-Thru" "FIX-DUT-FIX"];

for idx = 1:numel(structure)
currStructure = structure(idx);
switch currStructure
case "2X-Thru"
caus = causalityMetricFIX_FD;
reci = reciprocityMetricFIX_FD;
pass = passivityMetricFIX_FD;
case "FIX-DUT-FIX"
caus = causalityMetricFDF_FD;
reci = reciprocityMetricFDF_FD;
pass = passivityMetricFDF_FD;
end

6-299
6 RF Toolbox Examples

reportCausality_FD(caus,currStructure)
reportReciprocity_FD(reci,currStructure)
reportPassivity_FD(pass,currStructure)
end

Causality for 2X-Thru is good (Frequency domain check).

Reciprocity for 2X-Thru is good (Frequency domain check).

Passivity for 2X-Thru is good (Frequency domain check).

Causality for FIX-DUT-FIX is good (Frequency domain check).

Reciprocity for FIX-DUT-FIX is good (Frequency domain check).

Passivity for FIX-DUT-FIX is good (Frequency domain check).

Use IEEE P370 Time Domain Function

Call the IEEE P370 qualityCheck function to perform the S-parameters check using a time domain
method that constructs new S-parameters by enforcing passivity, causality, and reciprocity on the
original data.

Set up variables to be used by qualityCheck.

portNum = 2;
dataRate = 2; % gbps
risePer = 0.1; % fractional rise time of UI
samplePerUI = 36;
pulseShape = 1; % 1 is Gaussian; 2 is rectangular with Butterworth filter; 3 is rectangular with
extrapolationMethod = 2; % 1 is constant extrapolation; 2 is zero padding; 3 is repeating

Now, call qualityCheck on the 2X-Thru data, which produces plots that display the time response of
the original and the constructed S-parameters. Ideally, the time responses of the original and
constructed S-parameters must be identical, as the quality metric is based on quantifying the
difference between the two. In the following plots, the results are visually similar, indicating that the
data is most likely of good quality.

[causalityMetricFIX_TD,reciprocityMetricFIX_TD,passivityMetricFIX_TD] = ...
qualityCheck(freqFIX,sdataFIX,portNum, ...
dataRate,samplePerUI,risePer, ...
pulseShape,extrapolationMethod, ...
1);

causality_time_domain_difference_mv = 2×2

0.0002 0.0036
0.0036 0.0002

passivity_time_domain_difference_mv = 2×2
10-15 ×

0.1998 0.6784
0.5711 0.3210

6-300
Characterize High-Frequency Data and Interconnects Using IEEE P370 Standard

6-301
6 RF Toolbox Examples

6-302
Characterize High-Frequency Data and Interconnects Using IEEE P370 Standard

6-303
6 RF Toolbox Examples

Next, call qualityCheck function on the FIX-DUT-FIX data. Once again, observe that the time
responses of the original and constructed S-parameters are similar, indicating that the data is most
likely of high quality.

[causalityMetricFDF_TD,reciprocityMetricFDF_TD,passivityMetricFDF_TD] = ...
qualityCheck(freqFDF,sdataFDF,portNum, ...
dataRate,samplePerUI,risePer, ...
pulseShape,extrapolationMethod, ...
5);

causality_time_domain_difference_mv = 2×2

0.0002 0.0043
0.0043 0.0002

passivity_time_domain_difference_mv = 2×2
10-15 ×

0.1866 0.6576
0.5685 0.3194

6-304
Characterize High-Frequency Data and Interconnects Using IEEE P370 Standard

6-305
6 RF Toolbox Examples

6-306
Characterize High-Frequency Data and Interconnects Using IEEE P370 Standard

6-307
6 RF Toolbox Examples

Although the data seems to be of good quality based on visual inspection of the plots, quantitatively
verify that the error is within the ranges suggested by IEEE P370.

for idx = 1:numel(structure)

currStructure = structure(idx);
switch currStructure
case "2X-Thru"
caus = causalityMetricFIX_TD;
reci = reciprocityMetricFIX_TD;
pass = passivityMetricFIX_TD;
case "FIX-DUT-FIX"
caus = causalityMetricFDF_TD;
reci = reciprocityMetricFDF_TD;
pass = passivityMetricFDF_TD;
end

reportCausality_TD(caus,currStructure)
reportReciprocity_TD(reci,currStructure)
reportPassivity_TD(pass,currStructure)
end

Causality for 2X-Thru is good (Time domain check).

Reciprocity for 2X-Thru is good (Time domain check).

Passivity for 2X-Thru is good (Time domain check).

Causality for FIX-DUT-FIX is good (Time domain check).

6-308
 Characterize High-Frequency Data and Interconnects Using IEEE P370 Standard

Reciprocity for FIX-DUT-FIX is good (Time domain check).

Passivity for FIX-DUT-FIX is good (Time domain check).

Cross-Validate with RF Toolbox

Cross-validate with RF Toolbox functions iscausal, ispassive, and passivity that the S-
parameters are causal and passive. The plots from the passivity function illustrate that the data is
less than unity, indicating passivity.

for idx = 1:numel(structure)

currStructure = structure(idx);
switch currStructure
case "2X-Thru"
S = sFIX;
case "FIX-DUT-FIX"
S = sFDF;
end

if iscausal(S)
disp(currStructure + " is causal according to iscausal function.")
else
disp(currStructure + "is not causal according to iscausal function.")
end

if ispassive(S)
disp(currStructure + " is passive according to ispassive function.")
else
disp(currStructure + "is not passive according to ispassive function.")
end
figure
passivity(S)
end

2X-Thru is causal according to iscausal function.

2X-Thru is passive according to ispassive function.

6-309
6 RF Toolbox Examples

FIX-DUT-FIX is causal according to iscausal function.

FIX-DUT-FIX is passive according to ispassive function.

6-310
 Characterize High-Frequency Data and Interconnects Using IEEE P370 Standard

Self De-embedding of 2X-Thru

To verify the assumptions of the de-embedding algorithm are appropriate for the fixtures, de-embed
the 2X-Thru from itself to check that the result is an electrically transparent interconnect [1].

Self De-embed

Call the IEEEP3702xThru function to extract the error boxes representing the left and right halves
of a 2-port 2X-Thru. You can also use the IEEE function IEEEP370mmZc2xthru to extract error boxes
from mixed mode S-parameters.

[sFIX1,sFIX2] = IEEEP3702xThru(sFIX);

Calculate the de-embeded 2X-Thru.

sFIX_deembed = deembedsparams(sFIX,sFIX1,sFIX2);

Check De-embeded 2X-Thru

Plot the magnitude and phase of the self-de-embedded 2X-Thru to observe that the following criteria
are met:

• Absolute value of the magnitude of residual insertion loss is < 0.1 dB.
• Phase of self-de-embedded 2X-Thru is < 1 deg.

sFIX21 = squeeze(sFIX_deembed.Parameters(2,1,:));

6-311
6 RF Toolbox Examples

figure
hAx = subplot(2,1,1);
plot(hAx,freqFIX,db(sFIX21));
xlabel(hAx,"Frequency (Hz)"),ylabel(hAx,"Magnitude (dB)")
legend("dB (S21)")
title(hAx,"S21 self de-embedding amplitude")

hAx = subplot(2,1,2);
plot(hAx,freqFIX,angle(sFIX21));
xlabel(hAx,"Frequency (Hz)"),ylabel(hAx,"Angle (degrees)")
legend("deg (S21)")
title(hAx,"S21 self de-embedding angle")

These plots show that the magnitude is 0 and the angle is less than 1e-16, suggesting that the self-de-
embedding is successful.

Compare TDR of Fixture Model and FIX-DUT-FIX

Verify assumption that the fixture model is identical to the fixture attached to the DUT by comparing
their time-domain reflectometer (TDR) responses.

Calculate TDR

Visualize the TDR responses of the 2X-Thru and the FIX-DUT-FIX by calling the IEEE P370 function
compare_s2x_fdf_tdr.

compare_s2x_fdf_tdr(sFIX,sFDF)

6-312
Characterize High-Frequency Data and Interconnects Using IEEE P370 Standard

6-313
6 RF Toolbox Examples

Use the TDR response calculated and displayed in the previous steps to validate the behavior of the
fixtures on the 2X-Thru and FIX-DUT-FIX. To source the data from the plots, use the helper function
getTdrData.

[zdataFIX,zdataFDF,timeFIX,timeFDF,hFigs] = getTdrData;

Compare TDRs

Inspect the TDR plots to set the start of the fixture to the time of the midpoint of the 2X-Thru. The
start and midpoint times for port 1 and port 2 are identical, indicating that you can use the same
computations for both ports.

assert(isequal(timeFIX,timeFDF))
t = timeFIX;

tStart = 0.07;
tEnd = 0.56;
[~,idxStart] = min(abs(t-tStart));
[~,idxEnd] = min(abs(t-tEnd));

zdataFIX1 = zdataFIX(:,1);zdataFIX2 = zdataFIX(:,2);

zdataFDF1 = zdataFDF(:,1);zdataFDF2 = zdataFDF(:,1);

hFig = hFigs(1); hAx = hFig.CurrentAxes;

text(hAx,tStart,zdataFIX1(idxStart),"\leftarrow START","FontWeight","bold")
text(hAx,tEnd,zdataFIX1(idxEnd),sprintf("MIDPOINT\n") + "\downarrow","FontWeight","bold","Vertica

6-314
 Characterize High-Frequency Data and Interconnects Using IEEE P370 Standard

hFig = hFigs(2); hAx = hFig.CurrentAxes;

text(hAx,tStart,zdataFIX2(idxStart),"\leftarrow START","FontWeight","bold")
text(hAx,tEnd,zdataFIX2(idxEnd),sprintf("MIDPOINT\n") + "\downarrow","FontWeight","bold","Vertica

6-315
6 RF Toolbox Examples

Measure the point-wise difference between the 2X-Thru impedance and the FIX-DUT-FIX impedance
from the start to the midpoint.

absDiff1 = abs(zdataFIX1(idxStart:idxEnd)-zdataFDF1(idxStart:idxEnd));
absDiff2 = abs(zdataFIX2(idxStart:idxEnd)-zdataFDF2(idxStart:idxEnd));

Assuming the most stringent requirements, verify that the difference from the reference impedance is
within ±2.5%.

assert(sFIX.Impedance==sFDF.Impedance)
Z = sFIX.Impedance;

assert(max(absDiff1./Z) < 0.025)

assert(max(absDiff2./Z) < 0.025)

Determine Similarity of Expected and De-embedded S-parameters of DUT

De-embed the DUT from the fixtures. Compare the de-embedded result against the expected original
DUT.

sDUT_deembed = deembedsparams(sFDF,sFIX1,sFIX2);
sdataDUT_deembed = sDUT_deembed.Parameters;

Perform Qualitative Measure of Accuracy

Visually compare the de-embedded DUT against the original DUT. Ideally, the de-embedded results
and the original results must be identical.

6-316
 Characterize High-Frequency Data and Interconnects Using IEEE P370 Standard

figure
plotComparisonDUT(sdataDUT_deembed,sdataDUT,freqFDF,freqDUT)
sgtitle("Qualitative Results (IEEE P370)")

Perform Quantitative Measure of Accuracy

Calculate the absolute error between the de-embedded DUT and the original DUT and report it in dB.

figure
plotErrorDUT(sdataDUT_deembed,sdataDUT,freqFDF)
sgtitle("Quantitative Results (IEEE P370)")

6-317
6 RF Toolbox Examples

Improved Algorithm for Extracting Error Boxes

The de-embedded results from the Determine the similarity of S-parameters section begin to deviate
from the expected results at frequencies around 40 GHz, resulting in errors worse than -40 dB. This
deviation is an artifact caused by ripples originating from the sinc due to windowing performed
during the extraction of the left and right fixtures. The effect is accentuated at higher frequencies
due to the boundary discontinuity resulting from the circular nature of the DFT [2].

You can overcome these inherent limitations of windowing and the DFT by utilizing a continuous time
technique [3]. By representing the signal as a Laplace domain partial fraction, the time gating
operations are not reliant on the periodic assumptions of the DFT and thus avoid the boundary
discontinuity. In fact, because the gate width is infinite in this context, you can completely eliminate
the ripples, as the windowing by modeling a Heaviside step function.

Once again implement steps in the Self De-embed section by using the enhanced version of the IEEE
P370 function IEEEP3702xThru_ModifiedTimeGate to conduct de-embedding with higher
accuracy.

[sFIX1_improved,sFIX2_improved] = IEEEP3702xThru_ModifiedTimeGate(sFIX);
sFIX_deembed_improved = deembedsparams(sFIX,sFIX1_improved,sFIX2_improved);

sDUT_deembed_improved = deembedsparams(sFDF,sFIX1_improved,sFIX2_improved);
sdataDUT_deembed_improved = sDUT_deembed_improved.Parameters;

Visually compare the de-embedded DUT against the original DUT. The actual de-embedded results
are substantially closer to the expected original data, especially at higher frequencies.

6-318
 Characterize High-Frequency Data and Interconnects Using IEEE P370 Standard

figure
plotComparisonDUT(sdataDUT_deembed_improved,sdataDUT,freqFDF,freqDUT,"southwest")
sgtitle("Qualitative Results (IEEE P370 Modified)")

Plot the error. Observe that the highest error is below -40 dB, which is better than the maximum
error of the unmodified IEEE P370 algorithm.

figure
plotErrorDUT(sdataDUT_deembed_improved,sdataDUT,freqFDF)
sgtitle("Quantitative Results (IEEE P370 Modified)")

6-319
6 RF Toolbox Examples

Clean up

Clean up the warnings that you suppressed earlier in the example.

warnStruct = warning(origWarnStatus,'MATLAB:colon:operandsNotRealScalar');

References

[1] "IEEE Standard for Electrical Characterization of Printed Circuit Board and Related Interconnects
at Frequencies up to 50 GHz". IEEE Std 370-2020, vol., no., pp.1-147. 8 Jan. 2021. doi: 10.1109/
IEEESTD.2021.9316329.

[2] Oppenheim, Alan V., Ronald W. Schafer, and John R. Buck. Discrete-Time Signal Processing. 2nd
ed. Upper Saddle River, N.J: Prentice Hall, 1999.

[3] Reeves T. "Time Domain Gating of Microwave Component Responses Using Analog Techniques".
EDICON, 2016.

See Also

Related Examples
• “Rational Fit S-Parameters” on page 6-214

6-320
 Design Input Matching Network for Dipole Antenna

Design Input Matching Network for Dipole Antenna

This example shows how to implement an input matching network for a dipole antenna. You can
implement the input matching network in three steps:

• First, load the precalculated S-parameter matrix for a dipole antenna and the circuit element
values of a bandpass filter.
• Second, implement a netlist representation of a filter-antenna network using circuit objects.
• Finally, plot the input response and impedance of the network and observe the effect of the
matching network.

Load Precalculated Data for Antenna and Bandpass Filter

Load the precalculated S-parameter matrix for a dipole antenna resonating at 850 MHz.

dipole_Sparam = sparameters('Dipole_ant.s1p');

Load the precalculated circuit element values of a bandpass filter.

load bpf_elem_dip_ant_match

Plot the S-parameters of the dipole antenna and observe the resonance at 850 MHz.

rfplot(dipole_Sparam)

6-321
6 RF Toolbox Examples

Set up the frequency sweep parameters.

bw = 200e6; % bandwidth
f0 = 850e6; % center frequency
freq_sweep = f0-2*bw:bw/10000:f0+2*bw;

Implement Netlist Representation of Filter-Antenna Network

Implement a netlist representation of the filter-antenna network using circuit objects. The input
matching network is a bandpass filter and its inductor and capacitor values are derived from
bpf_elem_dip_ant_match.mat file. The matching network is shown in this figure.

ant = nport(dipole_Sparam);
ant_filt_ckt_1 = circuit('tuned_filt');
add(ant_filt_ckt_1,[1,2],inductor(m01));
add(ant_filt_ckt_1,[2,0],capacitor(c0));
add(ant_filt_ckt_1,[2,3],inductor(ind_bpf(1)));
add(ant_filt_ckt_1,[3,4],capacitor(cap_bpf(1)));
add(ant_filt_ckt_1,[4,0],inductor(ind_bpf(2)));
add(ant_filt_ckt_1,[4,0],capacitor(cap_bpf(2)));
add(ant_filt_ckt_1,[4,5],inductor(ind_bpf(3)));
add(ant_filt_ckt_1,[5,6],capacitor(cap_bpf(3)));
add(ant_filt_ckt_1,[6,0],ant);

Set up the input port for the network analysis.

setports(ant_filt_ckt_1,[1 0],{'in'});
S_ant_filt_1 = sparameters(ant_filt_ckt_1,freq_sweep);

Plot Response of Network

Plot the input response of the network and observe the S11 at the resonance frequency. The
operating region of the dipole antenna is widened when it is attached to an input matching network.

figure
rfplot(S_ant_filt_1)
ylim([-40 0])

6-322
 Design Input Matching Network for Dipole Antenna

Plot the impedance at the operating region of the dipole antenna.

Z_in_ant = (1+rfparam(dipole_Sparam,1,1))./(1-rfparam(dipole_Sparam,1,1))*50;
figure;
freq = dipole_Sparam.Frequencies;
plot(freq/1e9,real(Z_in_ant))
hold on
plot(freq/1e9,imag(Z_in_ant))
grid on
xlabel('Frequency (GHz)')
ylabel('Impedance (ohms)')
legend('Real{Zin}', 'Imag{Zin}' )

6-323
6 RF Toolbox Examples

See Also
Objects
circuit | capacitor | inductor

Apps
Matching Network Designer

Related Examples
• “Impedance Matching of Small Monopole Antenna” on page 6-166

6-324
 Model RF Impairments in 5G NR Downlink Waveform

Model RF Impairments in 5G NR Downlink Waveform

This example shows how to model RF impairments introduced by a superheterodyne transmitter, such
as phase noise, in-phase and quadrature (I/Q) imbalance, filter effects, and memoryless nonlinearity,
on a 5G NR downlink waveform. It also shows how to measure the corresponding error vector
magnitude (EVM).

Introduction

The example comprises these steps:

1 Generate a 5G NR downlink waveform defined by the 3GPP NR standard.
2 Model RF impairments as components of a superheterodyne transmitter.
3 Calculate the EVM of the 5G NR downlink waveform that is applied to the superheterodyne
transmitter.

This figure shows the processing chain implemented in this example. For a more detailed discussion
of the EVM measurement workflow, see the “EVM Measurement of 5G NR Downlink Waveforms with
RF Impairments” (5G Toolbox) example.

Simulation Parameters

Simulation Parameters for 5G NR Downlink Waveform Generation

For RF testing of base stations, the 3GPP 5G NR standard specifies a series of NR test model (NR-TM)
waveforms. For testing user equipment (UE), the standard outlines a set of fixed reference channel
(FRC) waveforms. The NR-TMs and FRCs for frequency range 1 (FR1) are detailed in TS 38.141-1,
whereas those for frequency range 2 (FR2) are described in TS 38.141-2.

Each FRC or NR-TM waveform is defined by a combination of these parameters:

6-325
6 RF Toolbox Examples

• NR-TM/FRC name
• Channel bandwidth
• Subcarrier spacing
• Duplexing mode

Define simulation parameters for an NR downlink waveform defined by the 3GPP 5G NR standard.

rc = ; % Reference channel (NR-TM or FRC)

% Select the NR waveform parameters

bw = ; % Channel bandwidth

scs = ; % Subcarrier spacing

dm = ; % Duplexing mode

For TMs, the generated waveform can contain more than one physical downlink shared channel
(PDSCH). The chosen PDSCH to be analyzed is determined by the radio network temporary identifier
(RNTI). By default, this example considers these RNTIs for EVM calculation with respect to the
following TMs:

• NR-FR1-TM2: RNTI = 2 (64QAM EVM)

• NR-FR1-TM2a: RNTI = 2 (256QAM EVM)
• NR-FR1-TM2b: RNTI = 2 (1024QAM EVM)
• NR-FR1-TM3.1: RNTI = 0 and 2 (64QAM EVM)
• NR-FR1-TM3.1a: RNTI = 0 and 2 (256QAM EVM)
• NR-FR1-TM3.1b: RNTI = 0 and 2 (1024QAM EVM)
• NR-FR1-TM3.2: RNTI = 1 (16QAM EVM)
• NR-FR1-TM3.3: RNTI = 1 (QPSK EVM)
• NR-FR2-TM2: RNTI = 2 (64QAM EVM)
• NR-FR2-TM2a: RNTI = 2 (256QAM EVM)
• NR-FR2-TM3.1: RNTI = 0 and 2 (64QAM EVM)
• NR-FR2-TM3.1a: RNTI = 0 and 2 (256QAM EVM)

As per the specifications (TS 38.141-1, TS 38.141-2), these TMs are not designed to perform EVM
measurements for conformance tests: NR-FR1-TM1.1, NR-FR1-TM1.2, NR-FR2-TM1.1. However, TMs
are easy to generate and parameterize, so you can use them outside the scope of the test for which
they were designed for, using their EVM measurements as a general methodology for assessing signal
quality. If you generate these TMs, the algorithm used in the example measures the EVM for the
following RNTIs:

• NR-FR1-TM1.1: RNTI = 0 (QPSK EVM)

• NR-FR1-TM1.2: RNTI = 2 (QPSK EVM)
• NR-FR2-TM1.1: RNTI = 0 (QPSK EVM)

For PDSCH FRCs and physical downlink control channel (PDCCH), by default, RNTI 0 is considered
for EVM calculation.

6-326
 Model RF Impairments in 5G NR Downlink Waveform

Calculate the PDSCH EVM for the RNTIs. To override the default PDSCH RNTIs, specify the
targetRNTIs vector.
targetRNTIs = [];

Simulation Parameters for EVM Measurement

To print EVM statistics, select displayEVM.

displayEVM = ;

To plot EVM statistics, clear plotEVM.

plotEVM = ;

To measure EVM as defined in TS 38.104, Annex B(FR1) / Annex C(FR2), select evm3GPP.

evm3GPP = ;

Simulation Parameters for RF Impairments Modeling

This example considers the most common impairments that distort the waveform when it passes
through an RF transmitter or receiver: phase noise, I/Q imbalance, filter effects, and memoryless
nonlinearity. To enable or disable impairments, toggle the phaseNoiseOn, IQImbalanceOn, and
nonLinearityModelOn flags. Define the distortion parameter values.
%#ok<*UNRCH>
% Set phase noise

phaseNoiseOn = ;
minFrequencyOffset = 1e4;

% Set I/Q imbalance

IQImbalanceOn = ;
amplitudeImbalance = 0.2;
phaseImbalance = 0.5;

% Set filtering

filterOn = ;

% Set amplifier gain

gain = 25; % Always on

% Set nonlinearity

nonLinearityModelOn = ;
rappMagnitudeSmooth = 1.15;
rappPhaseSaturation = 0.88;
rappPhaseSmooth = 3.43;

Furthermore, specify the modeling of white noise.

nfOn = ;
nf = 7;

6-327
6 RF Toolbox Examples

Generate 5G NR Downlink Waveform

Before generating the 5G NR downlink waveform, specify a higher waveform sample rate to model
wideband filter effects. You can increase the sample rate by multiplying the nominal sample rate with
the oversampling factor, OSR. To use the nominal sample rate, set OSR to 1.

OSR = 5; % oversampling factor

% Create waveform generator object

tmwavegen = hNRReferenceWaveformGenerator(rc,bw,scs,dm);

% Waveform bandwidth
bandwidth = tmwavegen.Config.ChannelBandwidth*1e6;

if OSR > 1
% The |Config| property in |tmwavegen| specifies the configuration of
% the standard-defined reference waveform. It is a read-only property.
% To customize the waveform, make the |Config| property writable.
tmwavegen = makeConfigWritable(tmwavegen);

% Increase the waveform sample rate by multiplying the nominal sample

% rate with |OSR|
nominalSampleRate = getNominalSampleRate(tmwavegen.Config);
tmwavegen.Config.SampleRate = nominalSampleRate*OSR;
else
filterOn = false;
end

Now, generate the waveform and get the waveform sample rate.

[txWaveform,tmwaveinfo,resourcesinfo] = generateWaveform(tmwavegen,tmwavegen.Config.NumSubframes)
sr = tmwaveinfo.Info.SamplingRate; % waveform sample rate

Normalize the waveform to fit the dynamic range of the nonlinearity.

gain = 25;
peakPower = 10^(-gain/20);
txWaveform = peakPower.*txWaveform/max(abs(txWaveform),[],'all');

The waveform consists of one frame for frequency division duplex (FDD) and two for time division
duplex (TDD). Repeat the signal twice. Remove the first half of the resulting waveform to avoid the
transient introduced by the phase noise model.

txWaveform = repmat(txWaveform,2,1);

Transmitter with RF Impairments

This section shows how to model RF impairments as components of a superheterodyne transmitter.

Using RF Toolbox™ Idealized Baseband library elements, add phase noise, I/Q imbalance, filter
effects, and memoryless nonlinearity to model the effects of upconverting the waveform to the carrier
frequency. There are two ways to model RF impairments: components-based and isolated
impairments modeling. This example models impairments using component-based impairments
modeling as shown in this figure. For more information on how to model the impairments as isolated
constituents, see the “EVM Measurement of 5G NR Downlink Waveforms with RF Impairments” (5G
Toolbox) example.

6-328
 Model RF Impairments in 5G NR Downlink Waveform

Build the superheterodyne transmitter architecture by characterizing these RF components:

• IQ modulator
• Bandpass filter (BPF)
• Power amplifier

% Define frequency values

switch tmwavegen.Config.FrequencyRange
case "FR1" % carrier frequency for FR1
fc = 4e9;
case "FR2" % carrier frequency for FR2
fc = 30e9;
end

% Build transmitter
RFmixer = rf.Mixer('Model','iqmod');
RFfilter = rf.Filter('ResponseType','Bandpass', ...
'UseFilterOrder',false, ...
'PassFreq_bp',fc+[-(sr/2-(sr/2*0.6)) sr/2-(sr/2*0.6)], ...
'PassAtten',0.7,...
'StopFreq_bp',fc+[-(sr/2-(sr/2*0.5)) sr/2-(sr/2*0.5)], ...
'StopAtten',60, ...
'SampleRate',sr, ...
'RF',fc);
RFHPA = rf.Amplifier('Gain',gain);

Add phase noise distortion. The phase noise characteristic is generated with the multipole zero model
described in TR 38.803 Section 6.1.10. The figure shows the phase noise characteristic.

if phaseNoiseOn
[level,frequencyOffset] = getPhaseNoisePSD(fc,sr,minFrequencyOffset);

RFmixer.IncludePhaseNoise = true;
RFmixer.PhaseNoiseLevel = level;
RFmixer.PhaseNoiseFrequencyOffset = frequencyOffset;
RFMixer.NoiseSeedSource = 'user';
RFMixer.NoiseSeed = 2137;

6-329
6 RF Toolbox Examples

visualizePhaseNoise(RFmixer)
end

Add I/Q imbalance by applying a 0.2 dB amplitude imbalance and a 0.5 degree phase imbalance to
the mixer. You can also increase the amplitude and phase imbalances by setting GainImbalance and
PhaseImbalance to higher values.

if IQImbalanceOn
RFMixer.GainImbalance = amplitudeImbalance;
RFMixer.PhaseImbalance = phaseImbalance;
end

Filter the waveform by using a BPF and plot its magnitude response.

If the current passband and stopband frequencies result in high EVM values for a different waveform
bandwidth and OSR, use a wider filter by increasing the parameters PassFreq_bp and StopFreq_bp
in the Idealized Baseband filter block. To use a narrower filter, reduce PassFreq_bp and
StopFreq_bp. You can also modify the passband ripple and the stopband attenuation. This figure
shows the magnitude response of the BPF.

if filterOn
RFfilter.FrequencyPoints = linspace(fc-sr,fc+sr,2001);

visualize(RFfilter)
end

6-330
 Model RF Impairments in 5G NR Downlink Waveform

Introduce nonlinear distortion using the Rapp model in the power amplifier. Set the parameters for
the Rapp model to match the characteristics of the memoryless model from TR 38.803 Annex A.1.
These figures show the nonlinearity introduced by the Rapp model.

if nonLinearityModelOn
RFHPA.Model = 'modified-rapp';

RFHPA.MagnitudeGainDB = gain;
RFHPA.Vsat = gain;
RFHPA.MagnitudeSmooth = rappMagnitudeSmooth;

RFHPA.PhaseSaturation = rappPhaseSaturation;
RFHPA.PhaseSmooth = rappPhaseSmooth;

visualize(RFHPA)
plotNonLinearCharacteristic(RFHPA)
end

6-331
6 RF Toolbox Examples

6-332
 Model RF Impairments in 5G NR Downlink Waveform

Finally, introduce white noise in elements of the transmitter.

if nfOn
% Speciy NF in IQ modulator
RFmixer.IncludeNoise = true;
RFmixer.NF = nf;

% Specify NF in HPA
RFHPA.IncludeNoise = true;
RFHPA.NF = nf;
end

Simulate the impairments of the transmitter on the waveform.

out_iq = RFmixer(txWaveform);
release(RFmixer)

if filterOn
out_filt = RFfilter(out_iq);
release(RFfilter)
else
out_filt = out_iq;
end

rxWaveform = RFHPA(out_filt);
release(RFHPA)

Plot the spectrum of the waveform before and after adding the RF impairments.

6-333
6 RF Toolbox Examples

scope = spectrumAnalyzer('SampleRate',sr,...
'ChannelNames',{'Before impairments','After impairments'},...
'Title', 'Waveform before and after impairments');
scope([(1/peakPower).*txWaveform,rxWaveform]);
release(scope);

The signal was previously repeated twice. Remove the first half of this signal. This avoids any
transient introduced by the impairment models.

nFrames = 1; % because dm = "FDD"

rxWaveform(1:nFrames*tmwaveinfo.Info.SamplesPerSubframe*10,:) = [];

Measure the EVM of the 5G NR Downlink Waveform

Calculate the EVM of the resulting signal. In this example, you plot the RMS and peak EVMs per
orthogonal frequency division multiplexing (OFDM) symbol, slot, and subcarrier and calculate the
overall EVM (RMS EVM averaged over the complete waveform). Annex B and Annex C of TS 38.104
define alternative methods for computing the EVM in FR1 and FR2, respectively. For detailed steps
on the decoding and analysis of the waveform, see the description of the helper function
hNRDownlinkEVM in the “EVM Measurement of 5G NR Downlink Waveforms with RF Impairments”
(5G Toolbox) example.

cfg = struct();
cfg.Evm3GPP = evm3GPP;
cfg.TargetRNTIs = targetRNTIs;
cfg.PlotEVM = plotEVM;

6-334
 Model RF Impairments in 5G NR Downlink Waveform

cfg.DisplayEVM = displayEVM;
cfg.IQImbalance = IQImbalanceOn;

% Compute and display EVM measurements

[evmInfo,eqSym,refSym] = hNRDownlinkEVM(tmwavegen.Config,rxWaveform,cfg);

EVM stats for BWP idx : 1

Low edge PDSCH RMS EVM, Peak EVM, slot 0: 1.213 3.862%
Low edge DM-RS RMS EVM, Peak EVM, slot 0: 1.159 3.001%
High edge PDSCH RMS EVM, Peak EVM, slot 0: 1.217 3.818%
High edge DM-RS RMS EVM, Peak EVM, slot 0: 1.159 3.001%
Low edge PDSCH RMS EVM, Peak EVM, slot 1: 1.143 3.861%
Low edge DM-RS RMS EVM, Peak EVM, slot 1: 1.081 2.481%
High edge PDSCH RMS EVM, Peak EVM, slot 1: 1.137 3.918%
High edge DM-RS RMS EVM, Peak EVM, slot 1: 1.081 2.481%
Low edge PDSCH RMS EVM, Peak EVM, slot 2: 1.276 4.235%
Low edge DM-RS RMS EVM, Peak EVM, slot 2: 1.128 2.524%
High edge PDSCH RMS EVM, Peak EVM, slot 2: 1.275 4.245%
High edge DM-RS RMS EVM, Peak EVM, slot 2: 1.128 2.524%
Low edge PDSCH RMS EVM, Peak EVM, slot 3: 1.232 4.316%
Low edge DM-RS RMS EVM, Peak EVM, slot 3: 1.144 2.827%
High edge PDSCH RMS EVM, Peak EVM, slot 3: 1.231 4.306%
High edge DM-RS RMS EVM, Peak EVM, slot 3: 1.144 2.827%
Low edge PDSCH RMS EVM, Peak EVM, slot 4: 1.268 3.995%
Low edge DM-RS RMS EVM, Peak EVM, slot 4: 1.413 3.297%
High edge PDSCH RMS EVM, Peak EVM, slot 4: 1.273 4.037%
High edge DM-RS RMS EVM, Peak EVM, slot 4: 1.413 3.297%
Low edge PDSCH RMS EVM, Peak EVM, slot 5: 1.155 4.288%
Low edge DM-RS RMS EVM, Peak EVM, slot 5: 1.105 2.447%
High edge PDSCH RMS EVM, Peak EVM, slot 5: 1.159 4.004%
High edge DM-RS RMS EVM, Peak EVM, slot 5: 1.105 2.447%
Low edge PDSCH RMS EVM, Peak EVM, slot 6: 1.203 4.308%
Low edge DM-RS RMS EVM, Peak EVM, slot 6: 1.080 3.127%
High edge PDSCH RMS EVM, Peak EVM, slot 6: 1.201 4.337%
High edge DM-RS RMS EVM, Peak EVM, slot 6: 1.080 3.127%
Low edge PDSCH RMS EVM, Peak EVM, slot 7: 1.168 4.306%
Low edge DM-RS RMS EVM, Peak EVM, slot 7: 1.234 3.643%
High edge PDSCH RMS EVM, Peak EVM, slot 7: 1.177 4.308%
High edge DM-RS RMS EVM, Peak EVM, slot 7: 1.234 3.643%
Low edge PDSCH RMS EVM, Peak EVM, slot 8: 1.236 4.373%
Low edge DM-RS RMS EVM, Peak EVM, slot 8: 1.143 2.694%
High edge PDSCH RMS EVM, Peak EVM, slot 8: 1.231 4.403%
High edge DM-RS RMS EVM, Peak EVM, slot 8: 1.143 2.694%
Low edge PDSCH RMS EVM, Peak EVM, slot 9: 1.111 3.595%
Low edge DM-RS RMS EVM, Peak EVM, slot 9: 1.160 2.889%
High edge PDSCH RMS EVM, Peak EVM, slot 9: 1.109 3.568%
High edge DM-RS RMS EVM, Peak EVM, slot 9: 1.160 2.889%
Low edge PDSCH RMS EVM, Peak EVM, slot 10: 1.174 3.874%
Low edge DM-RS RMS EVM, Peak EVM, slot 10: 1.149 2.510%
High edge PDSCH RMS EVM, Peak EVM, slot 10: 1.170 3.880%
High edge DM-RS RMS EVM, Peak EVM, slot 10: 1.149 2.510%
Low edge PDSCH RMS EVM, Peak EVM, slot 11: 1.484 4.588%
Low edge DM-RS RMS EVM, Peak EVM, slot 11: 1.607 3.240%
High edge PDSCH RMS EVM, Peak EVM, slot 11: 1.484 4.721%
High edge DM-RS RMS EVM, Peak EVM, slot 11: 1.607 3.240%
Low edge PDSCH RMS EVM, Peak EVM, slot 12: 1.269 4.018%
Low edge DM-RS RMS EVM, Peak EVM, slot 12: 1.263 2.728%
High edge PDSCH RMS EVM, Peak EVM, slot 12: 1.265 4.164%

6-335
6 RF Toolbox Examples

High edge DM-RS RMS EVM, Peak EVM, slot 12: 1.263 2.728%
Low edge PDSCH RMS EVM, Peak EVM, slot 13: 1.150 3.748%
Low edge DM-RS RMS EVM, Peak EVM, slot 13: 1.077 2.706%
High edge PDSCH RMS EVM, Peak EVM, slot 13: 1.151 3.838%
High edge DM-RS RMS EVM, Peak EVM, slot 13: 1.077 2.706%
Low edge PDSCH RMS EVM, Peak EVM, slot 14: 1.151 3.486%
Low edge DM-RS RMS EVM, Peak EVM, slot 14: 1.144 2.532%
High edge PDSCH RMS EVM, Peak EVM, slot 14: 1.149 3.531%
High edge DM-RS RMS EVM, Peak EVM, slot 14: 1.144 2.532%
Low edge PDSCH RMS EVM, Peak EVM, slot 15: 1.378 4.363%
Low edge DM-RS RMS EVM, Peak EVM, slot 15: 1.405 3.458%
High edge PDSCH RMS EVM, Peak EVM, slot 15: 1.381 4.337%
High edge DM-RS RMS EVM, Peak EVM, slot 15: 1.405 3.458%
Low edge PDSCH RMS EVM, Peak EVM, slot 16: 1.172 3.867%
Low edge DM-RS RMS EVM, Peak EVM, slot 16: 1.067 2.887%
High edge PDSCH RMS EVM, Peak EVM, slot 16: 1.171 3.940%
High edge DM-RS RMS EVM, Peak EVM, slot 16: 1.067 2.887%
Low edge PDSCH RMS EVM, Peak EVM, slot 17: 1.122 3.716%
Low edge DM-RS RMS EVM, Peak EVM, slot 17: 1.075 2.712%
High edge PDSCH RMS EVM, Peak EVM, slot 17: 1.133 3.512%
High edge DM-RS RMS EVM, Peak EVM, slot 17: 1.075 2.712%
Low edge PDSCH RMS EVM, Peak EVM, slot 18: 1.124 3.887%
Low edge DM-RS RMS EVM, Peak EVM, slot 18: 1.198 3.045%
High edge PDSCH RMS EVM, Peak EVM, slot 18: 1.122 3.906%
High edge DM-RS RMS EVM, Peak EVM, slot 18: 1.198 3.045%
PDCCH RMS EVM, Peak EVM, slot 0: 0.436 0.951%
PDCCH DM-RS RMS EVM, Peak EVM, slot 0: 0.240 0.374%
PDCCH RMS EVM, Peak EVM, slot 1: 0.541 1.029%
PDCCH DM-RS RMS EVM, Peak EVM, slot 1: 0.281 0.603%
PDCCH RMS EVM, Peak EVM, slot 2: 0.512 1.030%
PDCCH DM-RS RMS EVM, Peak EVM, slot 2: 0.263 0.446%
PDCCH RMS EVM, Peak EVM, slot 3: 0.502 1.220%
PDCCH DM-RS RMS EVM, Peak EVM, slot 3: 0.234 0.347%
PDCCH RMS EVM, Peak EVM, slot 4: 0.441 0.951%
PDCCH DM-RS RMS EVM, Peak EVM, slot 4: 0.228 0.390%
PDCCH RMS EVM, Peak EVM, slot 5: 0.454 1.007%
PDCCH DM-RS RMS EVM, Peak EVM, slot 5: 0.321 0.460%
PDCCH RMS EVM, Peak EVM, slot 6: 0.564 1.002%
PDCCH DM-RS RMS EVM, Peak EVM, slot 6: 0.253 0.407%
PDCCH RMS EVM, Peak EVM, slot 7: 0.497 1.084%
PDCCH DM-RS RMS EVM, Peak EVM, slot 7: 0.321 0.622%
PDCCH RMS EVM, Peak EVM, slot 8: 0.442 0.904%
PDCCH DM-RS RMS EVM, Peak EVM, slot 8: 0.280 0.433%
PDCCH RMS EVM, Peak EVM, slot 9: 0.487 1.015%
PDCCH DM-RS RMS EVM, Peak EVM, slot 9: 0.281 0.548%
PDCCH RMS EVM, Peak EVM, slot 10: 0.444 0.870%
PDCCH DM-RS RMS EVM, Peak EVM, slot 10: 0.218 0.467%
PDCCH RMS EVM, Peak EVM, slot 11: 0.510 1.126%
PDCCH DM-RS RMS EVM, Peak EVM, slot 11: 0.267 0.385%
PDCCH RMS EVM, Peak EVM, slot 12: 0.497 1.062%
PDCCH DM-RS RMS EVM, Peak EVM, slot 12: 0.303 0.560%
PDCCH RMS EVM, Peak EVM, slot 13: 0.417 0.906%
PDCCH DM-RS RMS EVM, Peak EVM, slot 13: 0.212 0.437%
PDCCH RMS EVM, Peak EVM, slot 14: 0.494 0.995%
PDCCH DM-RS RMS EVM, Peak EVM, slot 14: 0.196 0.417%
PDCCH RMS EVM, Peak EVM, slot 15: 0.514 1.137%
PDCCH DM-RS RMS EVM, Peak EVM, slot 15: 0.212 0.412%
PDCCH RMS EVM, Peak EVM, slot 16: 0.483 0.998%

6-336
 Model RF Impairments in 5G NR Downlink Waveform

PDCCH DM-RS RMS EVM, Peak EVM, slot 16: 0.228 0.327%
PDCCH RMS EVM, Peak EVM, slot 17: 0.476 1.380%
PDCCH DM-RS RMS EVM, Peak EVM, slot 17: 0.266 0.402%
PDCCH RMS EVM, Peak EVM, slot 18: 0.424 0.923%
PDCCH DM-RS RMS EVM, Peak EVM, slot 18: 0.202 0.384%

6-337
6 RF Toolbox Examples

6-338
Model RF Impairments in 5G NR Downlink Waveform

6-339
6 RF Toolbox Examples

6-340
Model RF Impairments in 5G NR Downlink Waveform

6-341
6 RF Toolbox Examples

Averaged overall PDSCH RMS EVM: 1.216%

Overall PDSCH Peak EVM = 4.7207%
Averaged overall PDCCH RMS EVM: 0.482%
Overall PDCCH Peak EVM = 1.38%

As per TS 38.104, the minimum requirement for the RMS EVM for QPSK/BPSK, 16QAM, 64QAM and
256QAM modulation must not exceed the EVM level of 17.5%, 12.5%, 8%, and 3.5%, respectively.
This example's default waveform NR-FR1-TM3.2 utilizes 16QAM, and the printed EVM measurements
indicate that the minimum requirement is met due to the PDSCH RMS EVM remaining below 12.5%.

The EVM per OFDM symbol, per slot, per subcarrier, and resource grid plots help identify specific
issues affecting the waveform's integrity:

• The OFDM symbol plot can be analyzed to pinpoint precise moments when the signal integrity
decreases to help identify adaptive modulation schemes that can apply real-time correction
mechanisms.
• The slot plot can give gain insight into broader fluctuations over time to address interference
patterns or network conditions.
• The subcarrier plot can be examined to identify frequency selective interference and fading to
improve spectral efficiency.
• The resource grid can be investigated to optimize scheduling algorithms and resource allocation
strategies to minimize overall network errors.

In this example, the EVM across all plots must remain stable because the RF impairments modeled
are temporally and spectrally independent.

6-342
 Model RF Impairments in 5G NR Downlink Waveform

The constellation diagram helps diagnose random versus systematic issues. A widely dispersed
contellation indicates high levels of noise or uncorrelated interference, whereas consistent deviations
point to high non-linearity and I/Q imbalance.

References

[1] TR 38.803 V14.3.0. "Study on new radio access technology: Radio Frequency (RF) and co-
existence aspects." 3rd Generation Partnership Project; Technical Specification Group Radio Access
Network.

[2] TS 38.141-1. "NR; Base Station (BS) conformance testing Part 1: Conducted conformance testing."
3rd Generation Partnership Project; Technical Specification Group Radio Access Network.

[3] TS 38.141-2. "NR; Base Station (BS) conformance testing Part 2: Conducted conformance testing."
3rd Generation Partnership Project; Technical Specification Group Radio Access Network.

[4] TS 38.104. "NR; Base Station (BS) radio transmission and reception." 3rd Generation Partnership
Project; Technical Specification Group Radio Access Network.

Local Functions

function [sampleRate] = getNominalSampleRate(cfgObj)

% Obtain nominal sample rate for the input parameter CFGOBJ. CFGOBJ is
% an object of type 'nrDLCarrierConfig'
sampleRate = nr5g.internal.wavegen.maxSampleRate(cfgObj);
end

function [PN_dBcPerHz,fVec] = getPhaseNoisePSD(fc,sr,minFrequencyOffset)

% Compute phase noise PSD for specified frequency offset fVec (in Hz) as
% defined in TR 38.803

fVec = logspace(log10(minFrequencyOffset),log10(sr/2)-0.001,20);
[~,idx] = min(abs(fc - [29.55, 45, 70]*1e9));
validMdl = {'29.55','45','70'};
mdl = validMdl{idx};
switch mdl
case '29.55'
% Parameter set from TR 38.803 for 29.55 GHz Fc
fcBase = 29.55e9;
fz = [3e3 550e3 280e6];
fp = [1 1.6e6 30e6];
alphaz = [2.37 2.7 2.53];
alphap = [3.3 3.3 1];
PSD0 = 32;
case '45'
% Parameter set from TR 38.803 for 45 GHz Fc
fcBase = 45e9;
fz = [3e3 451e3 458e6];
fp = [1 1.54e6 30e6];
alphaz = [2.37 2.7 2.53];
alphap = [3.3 3.3 1];
PSD0 = 35.65;
otherwise
% Parameter set from TR 38.803 for 70 GHz Fc
fcBase = 70e9;
fz = [3e3 396e3 754e6];
fp = [1 1.55e6 30e6];

6-343
6 RF Toolbox Examples

alphaz = [2.37 2.7 2.53];

alphap = [3.3 3.3 1];
PSD0 = 39.49;
end

% Compute numerator
num = ones(size(fVec));
for ii = 1:numel(fz)
num = num.*(1 + (fVec./fz(ii)).^alphaz(ii));
end

% Compute denominator
den = ones(size(fVec));
for ii = 1:numel(fp)
den = den.*(1 + (fVec./fp(ii)).^alphap(ii));
end

% Compute phase noise and apply a shift for carrier frequencies

% different from the base frequency
PN_dBcPerHz = 10*log10(num./den) + PSD0 + 20*log10(fc/fcBase);
end

function plotNonLinearCharacteristic(RFHPA)
% Plot the nonlinear characteristic of the power amplifier (PA) impairment.

% Input samples
x = complex((1/sqrt(2))*(-1+2*rand(1000,1)),(1/sqrt(2))*(-1+2*rand(1000,1)));

% Nonlinearity
yRapp = RFHPA(x);

% Release object to feed it a different number of samples

release(RFHPA);

% Plot characteristic
figure;
plot(10*log10(abs(x).^2),10*log10(abs(x).^2));
hold on;
grid on
plot(10*log10(abs(x).^2),10*log10(abs(yRapp).^2)-RFHPA.Gain,'.');
xlabel('Input Power (dBW)');
ylabel('Gainless Output Power (dBW)');
title('Nonlinearity Impairment')
legend('Linear characteristic', 'Rapp nonlinearity','Location','Northwest');
end

See Also
rf.Amplifier | rf.Mixer | rf.PAmemory | rf.Sparameter | rf.Filter

6-344