Scribd
Upload
96 views

Equalizers

This document discusses two types of equalizers used to minimize intersymbol interference (ISI) in communication systems: zero-forcing equalizers and minimum mean square error (MMSE) equalizers. Zero-forcing equalizers use a transversal filter structure to force the equalizer output to have zero values at sampling instants, eliminating ISI. This is achieved by solving a set of equations to determine the filter tap gains. MMSE equalizers instead minimize the average squared error over a window of output samples, providing a more robust solution but with some residual ISI outside the window. Both approaches are explained through examples of designing a 3-tap equalizer for a given received pulse.

Uploaded by

20ELB355 MOHAMMAD ESAAM
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
96 views

Equalizers

This document discusses two types of equalizers used to minimize intersymbol interference (ISI) in communication systems: zero-forcing equalizers and minimum mean square error (MMSE) equalizers. Zero-forcing equalizers use a transversal filter structure to force the equalizer output to have zero values at sampling instants, eliminating ISI. This is achieved by solving a set of equations to determine the filter tap gains. MMSE equalizers instead minimize the average squared error over a window of output samples, providing a more robust solution but with some residual ISI outside the window. Both approaches are explained through examples of designing a 3-tap equalizer for a given received pulse.

Uploaded by

20ELB355 MOHAMMAD ESAAM
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Equalizers

Zero-Forcing Equalizer
• It is not necessary to eliminate or minimize ISI with neighboring pulses for all
t.
• All that is needed is to eliminate or minimize interference with neighboring
pulses at their respective sampling instants only.
• This is because the receiver decision is based on sample values only.
• This kind of (relaxed) equalization can be accomplished by equalizers using
the transversal filter structure.
• Transversal filter equalizers are easily adjustable to compensate against different
channels or even time-varying channels.
• The design goal is to force the equalizer output pulse to have zero ISI values at
the sampling (decision-making) instants.
• In other words, the equalizer output satisfies the Nyquist or the controlled ISI criterion.
• The time delay T between successive taps is chosen to be Tb , the interval
between pulses.
(a)
• Set the tap gains Co = 1 and Ck = 0 for all other values of k in the transversal filter in Fig. (a).
• The output of the filter will be the same as the input delayed by NTb.
• For single pulse pr(t) (Fig. b) at the input of the transversal filter, the filter output po(t) will
be pr(t - NTb).
• i.e., the filter output po(t) is pr(t) delayed by NTb . For convenience, this delay is
ignored as it is not relevant to the present discussion.
• This means that pr(t) is also the filter output po(t) for the above tap setting.
• For zero ISI, it is required that the output pulse po(t) satisfy the Nyquist's
criterion or the controlled ISI criterion.
• For the Nyquist criterion, the output pulse po(t) must have zero values at all
the multiples of Tb
• However, from Fig. (b) we see that the pulse amplitudes at integer multiples of
Tb are non-zero.
• By adjusting the tap gains (ck), we generate additional pulses of proper
amplitudes that will force the resulting output pulse to have desired values at
t = 0, ±Tb,…....
The output is
(1)
The samples of po(t) at t = kTb are

(2)

Or in simpler notation

(3)

Eqn. (3) is the convolution sum. Assuming that the ISI span is from –N to N symbols only,
then according to the Nyquist criteria, the output pulse po(t) must satisfy

(4)

Upon substituting (4) in (3), we get a set of 2N+1 simultaneous equations for 2N+1
variables {ck}.
The 2N+1 equations can be written in the matrix form

(5)

• 2N+1 × 2N+1 matrix Pr has identical entries along all the diagonal lines. Such a matrix is
known as the Toeplitz matrix.
• The Toeplitz matrix is fully determined by its first row and first column. It has some nice
properties and admits simpler algorithms for computing its inverse.
• The tap gain ck can be obtained by solving this set of equations by taking the inverse of the
matrix Pr .
MMSE Method
• An alternative approach is to minimize the mean square error between the
equalizer output response po[k] and the desired zero ISI response.
• This is known as the minimum mean square error (MMSE) method for
designing transversal filter equalizers.
• The MMSE method does not try to force the pulse samples to zero at 2N
points. Instead, the squared errors averaged over a set of output samples is
minimized.
• This method involves more simultaneous equations. Thus we must find the
equalizer tap values to minimize the average (mean) square error over a
larger window [-K, K]:
• The solution to this minimization problem can be better represented in matrix form as

where 𝑷𝑟† represents the Moore-Penrose pseudo-inverse of the non-square


matrix Pr of size (2K + 1) x (2N + 1).
The MMSE design often leads to a more robust equalizer for the reduction of
ISI.
Example:
For the received pulse pr(t), let

Design a 3-tap (N = 1) equalizer.


• Note that the equalizer determined from Eq. (5) can guarantee only the zero ISI condition
of Eq. (4). In other words, ISI is zero only for k = 0, ±I, ... , ±N.
• For outside this range, samples 𝑝𝑜 (𝑘𝑇𝑏 ) ≠ 0, indicating some residual ISI.
• The 3-tap ZFE of the previous example has residual ISI

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy