Recommendation ITU-R P.

2040-1
(07/2015)

Effects of building materials and structures

on radiowave propagation above
about 100 MHz

P Series
Radiowave propagation
ii Rec. ITU-R P.2040-1

Foreword
The role of the Radiocommunication Sector is to ensure the rational, equitable, efficient and economical use of the
radio-frequency spectrum by all radiocommunication services, including satellite services, and carry out studies without
limit of frequency range on the basis of which Recommendations are adopted.
The regulatory and policy functions of the Radiocommunication Sector are performed by World and Regional
Radiocommunication Conferences and Radiocommunication Assemblies supported by Study Groups.

Policy on Intellectual Property Right (IPR)

ITU-R policy on IPR is described in the Common Patent Policy for ITU-T/ITU-R/ISO/IEC referenced in Annex 1 of
Resolution ITU-R 1. Forms to be used for the submission of patent statements and licensing declarations by patent
holders are available from http://www.itu.int/ITU-R/go/patents/en where the Guidelines for Implementation of the
Common Patent Policy for ITU-T/ITU-R/ISO/IEC and the ITU-R patent information database can also be found.

Series of ITU-R Recommendations

(Also available online at http://www.itu.int/publ/R-REC/en)

Series Title

BO Satellite delivery
BR Recording for production, archival and play-out; film for television
BS Broadcasting service (sound)
BT Broadcasting service (television)
F Fixed service
M Mobile, radiodetermination, amateur and related satellite services
P Radiowave propagation
RA Radio astronomy
RS Remote sensing systems
S Fixed-satellite service
SA Space applications and meteorology
SF Frequency sharing and coordination between fixed-satellite and fixed service systems
SM Spectrum management
SNG Satellite news gathering
TF Time signals and frequency standards emissions
V Vocabulary and related subjects

Note: This ITU-R Recommendation was approved in English under the procedure detailed in Resolution ITU-R 1.

Electronic Publication
Geneva, 2015

 ITU 2015
All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without written permission of ITU.
 Rec. ITU-R P.2040-1 1

RECOMMENDATION ITU-R P.2040-1

Effects of building materials and structures on radiowave

propagation above about 100 MHz
(Question ITU-R 211/3)
(2013-2015)

Scope
This Recommendation provides guidance on the effects of building materials and structures on radio-wave
propagation.

The ITU Radiocommunication Assembly,

considering
a) that electrical properties of materials and their structures strongly affect radiowave
propagation;
b) that it is necessary to understand the losses of radiowaves caused by building materials and
structures;
c) that there is a need to give guidance to engineers to avoid interference from outdoor to
indoor and indoor to outdoor systems;
d) that there is a need to provide users with a unified source for computing effects of building
materials and structures,
noting
a) that Recommendation ITU-R P.526 provides guidance on diffraction effects, including
those due to building materials and structures;
b) that Recommendation ITU-R P.527 provides information on the electrical properties of the
surface of the Earth;
c) that Recommendation ITU-R P.679 provides guidance on planning broadcasting-satellite
systems;
d) that Recommendation ITU-R P.1238 provides guidance on indoor propagation over the
frequency range 900 MHz to 100 GHz;
e) that Recommendation ITU-R P.1406 provides information on various aspects of
propagation relating to terrestrial land mobile and broadcasting services in the VHF and UHF
bands;
f) that Recommendation ITU-R P.1407 provides information on various aspects of multi-path
propagation;
g) that Recommendation ITU-R P.1411 provides propagation methods for short paths in
outdoor situations, in the frequency range from about 300 MHz to 100 GHz;
h) that Recommendation ITU-R P.1812 provides a propagation prediction method for
terrestrial point-to-area services in the frequency range 30 MHz to 3 GHz,
2 Rec. ITU-R P.2040-1

recommends
that the information and methods in Annex 1 and Annex 2 should be used as a guide for the
assessment of the effects of building material properties and structures on radiowave propagation,
and in developing deterministic models of propagation involving the built environment.

Annex 1 describes basic principles, and provides expressions to evaluate reflection from and
transmission through building materials and structures. It also includes a model for electrical
properties as a function of frequency, and a table of parameters for relevant materials.
Annex 2 gives definitions for various types of propagation loss associated with buildings, and
provides guidance on measuring building entry losses.
Examples of building-entry loss measurements may be found in Report ITU-R P.2346.

Annex 1

1 Introduction
This Annex provides guidance on the effects of building material electrical properties and structures
on radio-wave propagation.
Section 2 describes fundamental principles concerning the interaction of radio waves with building
materials, defines various parameters in use for these purposes, and gives basic expressions for
reflection from and transmission through single material interfaces and single and multiple layer
slabs, typical of building construction.
Section 3 defines a model for electrical properties, and a table of parameters for various building
materials.

2 Basic principles and theory

Radio waves that interact with a building will produce losses that depend on the electrical properties
of the building materials and material structure. In this section, theoretical effects of material
electrical properties and structure on radio-wave propagation will be discussed.

2.1 Theory of material electrical properties

2.1.1 Introduction
This section describes the development of simple frequency-dependent formulae for the permittivity
and conductivity of common building materials. The formulae are based on curve fitting to a
number of published measurement results, mainly in the frequency range 1-100 GHz. The aim is to
find a simple parameterization for use in indoor-outdoor ray trace modelling.
The characterization of the electrical properties of materials is presented in a number of different
ways in the literature. These are described in § 2.1.2 in order that the measured data can be reduced
to a common format.
 Rec. ITU-R P.2040-1 3

2.1.2 Method
2.1.2.1 Definitions of electrical constants
The following treatment deals only with non-ionized, non-magnetic materials, and throughout we
therefore set the free charge density, f, to zero and the permeability of the material, , to the
permeability of free space 0.
The fundamental quantities of interest are the electrical permittivity, , and the conductivity, .
There are many ways of quantifying these parameters in the literature, so we first make explicit
these different representations and the relations between them.

2.1.2.2 Derivation
The starting point is the wave equation derived from Maxwell’s equations. Under the above

assumptions, the wave equation for the electric field E is:
 
 2E J f
 E –  0 2   0
2
(1)
t t

where:

E : (vector) electric field intensity (V/m)
Jf : current density of free charges (A/m2)
: dielectric permittivity (F/m)
0 : permeability of free space (N/A2) = 4  107 by definition.
 
In a conductor, J f is related to E through Ohm’s Law by:

 
J f  E (2)

where:
: conductivity (S/m).
Combining equations (1) and (2) gives:
 
 2E E
 E –  0 2   0
2
(3)
t t

Writing E in exponential notation:

   
E  E0e j t – k  r  (4)

where:
  
E0 : value of E for t = r = 0 (V/m)

k : (vector) wave number (m−1) magnitude = 2/ where  is the wavelength in m
: angular frequency (s1) = 2f where f is the frequency in s1
𝑟⃗ : (vector) spatial distance (m).
4 Rec. ITU-R P.2040-1

and substituting in equation (3) gives

k 2 –  02  j 0  0 (5)


where k is the magnitude of k .
Equation (5) shows that the electric field intensity propagates as an attenuated sinusoidal wave.

2.1.2.3 Non-conducting dielectric

In a non-conducting dielectric ( = 0) the field is unattenuated and from equation (5) the velocity of
propagation, v (= /k), is:
1
v (6)
 0

 is conventionally written in terms of the relative permittivity and the permittivity of free space:
   0 (7)
where
: relative dielectric permittivity of the medium concerned
0: dielectric permittivity of free space = 8.854  10−12 (F/m).
Thus the velocity of propagation in a medium of relative permittivity  can be written:
c (8)
v


where c is the velocity of light in free space (= 1/ 00 ). In other words,  is the refractive
index of the dielectric medium.

2.1.2.4 Conducting dielectric

When   0, the wave attenuates as it propagates. It is convenient in this case to define a complex
relative permittivity which may be derived as follows. Equation (5) can be rearranged, with the
substitution c 2 1 /  00  , to give:

c2 
  j (9a)
 2  0
2
Since equation (8) gives c 2   , this can be interpreted as a complex relative permittivity given by


  ' j (9b)
 0

This shows that the relative permittivity defined for a pure dielectric, becomes the real part ' of the
more general, complex relative permittivity  defined for a conducting dielectric.
There are no universally accepted symbols for these terms. In this Recommendation, relative
permittivity is written in the form:

    j (10)
 Rec. ITU-R P.2040-1 5

where ' and '' are the real and imaginary parts. Using equation (9b), the imaginary part is given
by:


  (11)
 0

Note that the sign of the imaginary part of  is arbitrary, and reflects the sign convention in
equation (4). In practical units, equation (11) gives a conversion from " to :
  0.05563 fGHz (12)

Another formulation of the imaginary part of  is in terms of the loss tangent, defined as:

 (13)
tan  


and so:

tan   (14)


From equation (10) this gives:

  ' (1  j tan ) (15)

and in practical units:

  0.05563  tan  f GHz (16)

Another term sometimes encountered is the Q of the medium. This is defined as:


Q (17)


and is the ratio of the displacement current density D / t to the conduction current density Jf. For
non-conductors, Q  . From equation (14):

Q 1/ tan  (18)

Yet another term encountered is the complex refractive index n which is defined to be  . Writing
n in terms of its real and imaginary parts:

n  n  jn   (19)
6 Rec. ITU-R P.2040-1

', " and  are given from equations (10) and (12) by:
  ( n) 2 – (n) 2
  2 nn (20)
  0.1113 n n f GHz

2.1.2.5 Attenuation rate

A conducting dielectric will attenuate electromagnetic waves as they propagate. To quantify this,
substitute equation (5) in equation (4) and simplify using equation (14):
 
  
E  E0 exp j t – ' 1 – j tan k0  r  (21)

where:

k0 : (vector) wave number (m1) in free space.
The imaginary part under the square root sign leads to an exponential decrease of the electric field
with distance:
 
E  E0 exp  r /  

(22)

In a practical calculation using complex variables, the attenuation distance, , at which the field
amplitude falls by 1/e, can be evaluated as:
1 (23a)


Im k0  
where the function “Im” returns the imaginary part of its argument. Analytically it can be shown
that:

1 2 cos  (23b)

k0 ' 1  cos 

which can be evaluated by calculating tan  from ' and  and inverting to obtain cos . More
direct evaluation is possible in the two limits of   0 (dielectric limit) and    (good
conductor limit). By choosing the appropriate approximation of the term under the square root sign
in equation (21) these limits are:

1 2 (24)
 dielectric 
k0 ' tan 

and:
1 2 (25)
 conductor 
k0 ' tan 

Equations (24) and (25) are accurate to about 3% for tan  < 0.5 (dielectric) tan  > 15 (conductor).
conductor is usually referred to as the “skin depth”.
For practical purposes the attenuation rate is a more useful quantity than the attenuation distance,
and is related to it simply by
 Rec. ITU-R P.2040-1 7

20 log10 e
A  8.686 /  (26)

where:
A: attenuation rate in dB/m (with  in m).
Substituting equations (24) and (25) in equation (26) and converting to practical units gives:

 (27a)
Adielectric  1636


Aconductor  545.8  fGHz (27b)

2.1.3 Frequency dependence of material properties

In the literature, the real part of the dielectric constant, ', is always given, but often the frequency
is not specified. In practice for many materials, the value of ' is constant from DC up to around
5-10 GHz after which it begins to fall with frequency.
The value of  is usually a strong function of frequency in the band of interest, increasing with
frequency. This may be one reason why the imaginary part of the dielectric constant, or the loss
tangent, is often specified in the literature: equations (12) and (16) show that these terms remove a
linear frequency dependence compared to the frequency dependence of .
For each material a simple regression model for the frequency dependence of  can be obtained by
fitting to measured values of  at a number of frequencies.

2.1.4 Models of material properties frequency dependence

In order to derive the frequency dependence of material properties, the values of the electrical
constants of the materials can be characterized in terms of the measurement frequency, real part (')
and imaginary part ('') of the relative permittivity, loss tangent (tan ) and conductivity ().
Expressions in § 2.1.2.4 permit conversions between these quantities.
For the conductivity, there is usually statistically significant evidence for an increase with
frequency. In this case the trend has been modelled using:

d
  c f GHz (28)

where c and d are constants characterizing the material. This is a straight line on a log()–log(f)
graph. The trend line is the best fit to all available data.
For the relative permittivity one can assume similar frequency dependency:

b
  a f GHz (29)

where a and b are constants characterizing the material. However in almost all cases there is no
evidence of a trend of relative permittivity with frequency. In these cases a constant value can be
used at all frequencies. The constant value is the mean of all the values plotted. Some examples are
given in Table 3.
8 Rec. ITU-R P.2040-1

2.2 Effects of material structure on radiowave propagation

2.2.1 Plane wave reflection and transmission at a single planar interface
This section considers a plane wave incident upon a planar interface between two homogeneous and
isotropic media of differing electric properties. The media extend sufficiently far from the interface
such that the effect of any other interface is negligible. This may not be the case with typical
building geometries. For example, propagation losses due to a wall may be influenced by multiple
internal reflections. Methods for calculating reflection and transmission coefficients of single-layer
and multi-layer slabs are given in § 2.2.2.
A plane wave is useful for analysis purposes, but the concept is largely theoretical. In practice a
wave may approximate but not be exactly planar. The point is important here because a truly plane
wave does not experience free-space (spreading) loss. The following methods take no account of
free-space losses, only the effect of the media interface.

2.2.1.1 Oblique incidence on a plane media interface

Figure 1 illustrates a sinusoidal plane wave incident obliquely to a plane interface separating two
uniform non-magnetic dielectric media having relative permittivities 1 and 2. Values for  can be
calculated from the real part of the permittivity, ', and conductivity, , using equations (10) and
(11). Table 3 provides parameters from which these can be calculated as functions of frequency.
There are three important theorems for this case that follow from geometrical considerations.
1) The vector wave numbers of the reflected and transmitted (refracted) waves lie in the plane
of incidence, i.e. the plane defined by wave number ki of the incident wave and the normal
to the interface. This is taken to be the x-z plane in Fig. 1.
2) The angles of incidence and reflection are equal (both in Fig. 1).
3) The angle of refraction, is related to the angle of incidence by Snell’s law.

1 1
sin 1  sin 2 (30)
c1 c2

where c1  c/ 1 and c2  c/ 2 are the respective wave speeds in the two media, and 1 and 2
represent the complex relative permittivities of the two media.
These theorems ensure that the exponential space-time factors, exp j  t – k  r  , for the three

waves ( k  k1, k1´ , k2 , respectively) are identical at all points in the interface.
 Rec. ITU-R P.2040-1 9

FIGURE 1
Reflection and refraction of plane waves at plane interface
TE polarisation TM polarisation
x x
Medium 1 1 2 Medium 2 Medium 1 1 2 Medium 2

Sr Sr
Hr
kr kr Hr
kt Et kt
Er Et
St Er St
1 1 Ht
2 Ht z 2 z
1 1

ki ki
Ei
Si Si
Ei
Hi
Hi

P.2040-01

Two polarizations of the incident wave are shown in Fig. 1.

a) On the left the incident electric vector Ei is perpendicular to the plane of incidence.
This is known as transverse electric (TE) polarisation. Other terms are perpendicular
polarisation, s-polarisation, and -polarisation.
b) On the right the incident electric vector Ei is parallel to the plane of incidence.
This is known as transverse magnetic (TM) polarisation. Other terms are parallel
polarisation, p-polarization, and -polarization.
In the following descriptions, polarization will be designated by TE or TM.
An arbitrarily or circularly polarised wave can be resolved into its TE and TM components for
calculation purposes, which can then be re-combined.
E-field reflection and transmission coefficients are defined as the ratios of reflected and transmitted
(refracted) vectors respectively to the corresponding incident vector as they exist at the interface. In
general such coefficients are complex. The following expressions take no account of free-space or
other losses prior or subsequent to the interaction of a wave with the interface.
The requirement that electric and magnetic vectors are continuous in the plane of the interface give
the following expressions for electric field coefficients. Reflection and transmission coefficients are
denoted by R and T respectively. The subscripts indicate the vectors concerned, and whether the
polarization is TE or TM. Each of equations (31a) to (32b) are in two parts, according to whether
total internal reflection occurs. Total internal reflection is only possible when a wave is incident
upon a medium with lower refractive index.
E-field reflection coefficient for TE polarisation:

 1 cos 1   2 cos  2 1
 sin 1  1
E   cos 1   2 cos  2 2 (31a)
ReTE  r  1
Ei  1
 1 sin 1  1
 2

E-field reflection coefficient for TM polarisation:

10 Rec. ITU-R P.2040-1

  2 cos 1  1 cos  2 1
 sin 1  1
E r   2 cos 1  1 cos  2 2 (31b)
ReTM  
Ei  1
 1 sin 1  1
 2

E-field transmission coefficient for TE polarisation:

 2 1 cos 1 1
 sin 1  1
E   cos 1   2 cos  2 2 (32a)
TeTE  t  1
Ei  1
 0 sin 1  1
 2

E-field transmission coefficient for TM polarisation:

 2 1 cos 1 1
 sin 1  1
E   cos 1  1 cos  2 2 (32b)
TeTM  t  2
Ei  1
 0 sin 1  1
 2

where 1 and 2 are the complex relative permittivities of medium 1 and 2 respectively. These can
be evaluated using equation (9b) with values of ' and  obtained from § 3 and Table 3.
The cos2 terms in equations (31a) to (32b) can be evaluated in terms of 1 using equation (30) as:


cos  2  1  1 sin 2 1 (33)
2

At 1 = 0 the incidence plane is not uniquely defined. In this case all directions of propagation are
normal to the interface, and the coefficient amplitudes from the expression for each polarisation is
the same. In the case of reflection there is an apparent sign change. This arises purely from how the
polarizations are defined; it is not a physical discontinuity.

2.2.1.2 Calculation examples

Figure 2 gives examples of reflection and transmission coefficient amplitudes for a wave in air
incident upon concrete at 1 GHz calculated over a range of incidence angles for both polarizations
using equations (31a) to (32b), taking the properties of concrete from Table 3.
 Rec. ITU-R P.2040-1 11

FIGURE 2
Reflection and transmission coefficients for air/concrete interface at 1 GHz
1 1

Transmission coefficient amplitude

0.9 0.9
Reflection coefficient amplitude 0.8 0.8
0.7 0.7
TE
0.6 0.6
0.5 0.5
TM
0.4 0.4
0.3 0.3
TE
0.2 0.2
0.1
TM 0.1
0 0
0 10 20 30 40 50 60 70 80 90 0 10 20 30 40 50 60 70 80 90
Angle of incidence (degrees) Angle of incidence (degrees)

P.2040-02

2.2.1.3 Substitutions available in coefficient values

It can be useful to note the following substitutions for E-vector coefficients, where the subscripts
denote the medium, 1 or 2, in which the wave is incident on an interface:
a) For either polarisation, R1   R2 , and thus R12  R22
b) For either polarisation, T1T2  1  R 2 , where according to a) R can be either R1 or R2.

2.2.1.4 Coefficients for power flux-densities

Coefficients for power flux densities can be obtained from the E-vector coefficients:

S 2
RsTE  r  ReTE (34a)
Si

S 2
RsTM  r  ReTM (34b)
Si

St 2 (35a)
TsTE   TeTE 2
Si 1

St 2 (35b)
TsTM   TeTM 2
Si 1

The change in signal level in decibels due to reflection or transmission is thus given by 10 log RS  or
10 log TS 
where RS and TS stand for either reflection or transmission S-vector coefficient in
equations (34a) to (35b).
Conservation of energy at the media interface requires that for a given incident wavefront area,
the sum of the reflected and transmitted power flux equals the incident power flux. To illustrate this,
account must be taken of the change in wavefront width upon refraction. For either polarization:

cos  2
RS  TS 1 (36)
cos 1

cos  2
where adjusts for the change in wavefront width.
cos 1
12 Rec. ITU-R P.2040-1

2.2.1.5 Simplified expressions for incident wave in air

When medium 1 is air, equations (31a) to (32b) can be simplified to:

cos     sin 2 
ReTE  (37a)
cos     sin 2 

 cos     sin 2 
ReTM  (37b)
 cos     sin 
2

2 cos 
TeTE  (38a)
cos     sin 2 

2  cos 
TeTM  (38b)
 cos     sin  2

where  is the angle of incidence and  is the relative permittivity of the medium upon which the
wave is incident.
Total internal reflection at the interface is not possible in equations (37a) to (38b) since it can be
assumed that the wave is incident upon a medium with a higher refractive index than air.

2.2.2 Plane wave reflection and transmission for a single- or multi-layer slab
2.2.2.1 General method for a multi-layer slab
Figure 3 illustrates a plane wave incident upon a slab consisting of N layers, each with smooth,
planar and parallel surfaces, where N can be 1 or more. The relative permittivity of layer n is n,
and its width dn metres. It is assumed that the slab is in air, and for calculation purposes this is
designated as layers 0 and N + 1, with relative permittivity 1 and width 0.

FIGURE 3
Plane wave incident on single- or multi-layer slab

Air Slab Air

Layers: 0 1 2 n N N+1

Relative permittivities:0 = 1 1 2 n N N + 1 = 1

Layer widths, metres: d0 = 0 d1 d2 dn dN dN + 1 = 0

St
Sr
N + 1 = 

 n



Si

P.2040-03

The incidence and reflection angles are 0, and the wave will emerge from layer N at N+1 = 0.
The direction of propagation in layer n is n. A complete ray path through the layers is not shown in
 Rec. ITU-R P.2040-1 13

Fig. 3. For a single incident ray Si the departing rays Sr and St are spatially distributed due to
multiple internal reflections in the layers.
Reflection and transmission coefficients for the slab can be calculated by the following recursive
method.
First initialise:

AN 1 1 BN 1  0 FN 1 1 GN 1  0 (39a)-(39d)

Then for n = N, N-1, etc. … 0:

An  0.5 exp  j k n d n cos n An 1 1  Yn 1   Bn 1 1  Yn 1  (40a)

Bn  0.5 exp  j kn d n cos n An 1 1  Yn 1   Bn 1 1  Yn 1  (40b)

Fn  0.5 exp  j k n d n cos n Fn 1 1  Wn 1   Gn 1 1  Wn 1  (40c)

Gn  0.5 exp  j k n d n cos n Fn 1 1  Wn 1   Gn 1 1  Wn 1  (40d)

where

cos  n 1 n (41a)
Wn 1 
cos  n n 1

cos  n 1 n 1 (41b)
Yn 1 
cos  n n

sin 0
sin  n  (41c)
n

2
kn  n (41d)


and  is the free-space wavelength in metres.

The notional width d0 = 0 results in the exponential terms in equations (40a) to (40d) for n = 0
evaluating to 1. The relative permittivities for N + 1 are included in Fig. 3 only for consistency, they
are not used in the calculation.
Having evaluated equations (40a) to (40d) for, in order, n = N to n = 0, the E-field reflection and
transmission coefficients are given by:

B0 G0 1 1
RTE  RTM  TTE  TTM  (42a)-(42d)
A0 F0 A0 F0

where the subscripts TE and TM denote transverse-electric and transverse-magnetic incident

polarization respectively.
Attachment 1 provides an alternative formulation for the multi-layer slab method.
14 Rec. ITU-R P.2040-1

2.2.2.2 Simplified method for a single-layer slab

For a slab consisting of a single layer, that is, for which N = 1, the foregoing method can be
simplified to:

R 1  exp( j 2q) 
R (Reflection coefficient) (43a)
1  R2 exp( j 2q)

(1  R2 ) exp(  jq)

T (Transmission coefficient) (43b)
1  R2 exp(  j 2q)

where:

2 d
q   sin 2  (44)


d is the thickness of the building material, and R represents ReTE or ReTM, as given by equations
(37a) or (37b) respectively, depending on the polarization of the incident E-field.

2.2.2.3 Calculation examples

Figures 4 to 7 show examples of results from equations (42a)-(42d) for a single concrete slab at
1 GHz with four incidence angles. The same results may be obtained from equations (43a) and
(43b). The electrical properties for concrete are taken from Table 3.

FIGURE 4
Reflection coefficient for a concrete slab at 1 GHz, TE polarisation
0
Reflection coefficient amplitude (dB)

–5

– 10

– 15 85°
60°
– 20 30°
0°

– 25
Polarisation TE
– 30
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Slab thickness (m)

P.2040-04
 Rec. ITU-R P.2040-1 15

FIGURE 5
Reflection coefficient for a concrete slab at 1 GHz, TM polarisation
0

R eflection coefficient amplitude (dB)

–5

– 10

– 15 85°
0°
30°
– 20
60°
– 25
Polarisation TM
– 30
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Slab thickness (m)

P.2040-05

FIGURE 6
Transmission coefficient for a concrete slab at 1 GHz, TE polarisation
0
Polarisation TE
Transmission coefficient amplitude (dB)

–5

– 10
0°
– 15 30°
60°
– 20

– 25
85°

– 30
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Slab thickness (m)

P.2040-06

FIGURE 7
Transmission coefficient for a concrete slab at 1 GHz, TM polarisation
0
Polarisation TM
60°
Transmission coefficient amplitude (dB)

–5 30°
0°
– 10

– 15 85°

– 20

– 25

– 30
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Slab thickness (m)

P.2040-07

It will be noted in Figs 5 and 7 that the coefficients for TM polarization for 85 degrees incidence
have anomalous values compared to the ordering of the other three angles. This is the effect of the
minimum in reflection coefficient visible in Fig. 2 for TM polarization, known as the
pseudo-Brewster angle.
16 Rec. ITU-R P.2040-1

2.2.3 Waveguide propagation in buildings

2.2.3.1 Theory on frequency characteristics of attenuation constant in waveguide
A waveguide may comprise of a hollow space surrounded by lossy dielectric materials. In the case
of a building structure, a corridor, underground mall, or tunnel can be considered as a waveguide.
The radiowave power that propagates in a waveguide is attenuated according to the distance.
It is well known that a waveguide has frequency characteristics such as the cut-off frequency that
varies according to the shape. In this section, a formula is presented to derive the attenuation
constant for the frequency characteristics in a waveguide.
The cross-section of a square waveguide structure is shown in Fig. 8. In this case, the intrinsic
constants of the lossy dielectric material are different for the sidewalls and for the ceiling and the
floor.

FIGURE 8
Cross-section of waveguide and material constants

h2
r 

y r 

x
b z 

h1


a

P.2040-08

In Fig. 8, a is the width and b is the height of the waveguide (m), h1 and h2 are the root mean square
roughness of the Gaussian distribution of the surface level, and  is the tilt of the root mean square
(rad). The complex permittivity values for materials ri* are calculated as follows.

  
ri*  ri  j  ri"  i , i  1,2 (45)
 0 

where ri is the relative dielectric constant and i is the conductivity. The quantity ri″ is the loss
tangent of the materials,  is the angular frequency and 0 is the permittivity of free space.
The basic attenuation constant is formulated as follows.
 Rec. ITU-R P.2040-1 17

    *2 
*
  r1 
2   r1 1   1
Lbasic, h  K h  Re
3 *

3 *   Im 
  
  a  r1  1 b  r 2  1  2  a  r1  1 b  r 2  1 
4 * 4 *
 
   (dB/m) (46)
     2
*
  r 2*

2  1 r 2   1
Lbasic, v  K v  Re
  a 3
 *
 1

b 3
 *
 1




2 
Im 4 *

 a  
r1  
1
 4 *
b  r2 
 1

 

r1 r2
 

Kh and Kv are constant values that are dependent on the section shape. The constant values
dependent on the section shape are given in Table 1.

TABLE 1
Constant values for various cross-section shapes
Shape Circle Ellipse Square Arch-backed
Kh 5.09 4.45 4.34 5.13
Kv 5.09 4.40 4.34 5.09

The formulas mentioned above are valid based on equation (47) representing the condition of
constraint.

a  r1 1
 
 r1 (m) (47)
  b  r 2 1

Unique characteristics in square shape case

The attenuation constant due to roughness, which is regarded as local variations in the level of the
surface relative to the mean level of the surface of a wall, is given by:

 h  2  h  2 
Lroughness,h  K h    12    22  
2

 a   b  
(dB/m) (48)

2  h1 
2
 h2  
2
Lroughness,v  K v    2    2  
 a   b  

The attenuation constant due to the wall tilt is given by:

22
Ltilt,h  K h

(dB/m) (49)
22
Ltilt,v  Kv


Therefore, the total attenuation constant in a square shape case is the sum of the above losses:
18 Rec. ITU-R P.2040-1

Lh  Lbasic,  Lroughness ,h  Ltilt,h

h
(dB/m) (50)
Lv  Lbasic,  Lroughness ,v  Ltilt,v
v

2.2.3.2 Applicability of waveguide theory

The waveguide theory shows good agreement with the measured propagation characteristics in the
corridor in the frequency range of 200 MHz to 12 GHz in case there is no pedestrian traffic in the
corridor.
Effect of pedestrian traffic on waveguide
Figure 9 shows a comparison of the theoretical and measured attenuation constant values during the
day (when pedestrian traffic is present), and during the night (when the corridor is empty).
Theoretical values are calculated based on the parameters given in Table 2.

TABLE 2
Parameters used in underground calculation

Width Height Tilt Roughness Material constant

(m) (m) (degrees) h1 h2 r1 r2  
Underground 6.4 3.0 0.35 0.4 0.2 15 10 0.5 0.1

FIGURE 9
Attenuation constant comparison for day and night
100 100
70 70 3m Vertical pol.
3m Horizontal pol.
Atte nuation constant ( dB/100 m )

Atte nuation constant ( dB/100 m )

40 40 6.4 m
6.4 m Theory
Theory
20 20

10 10
7 7

4 4
Measured Measured
day time day time
2 2
night time night time

1 1
0.2 0.4 0.7 1 2 4 7 10 0.2 0.4 0.7 1 2 4 7 10 15
Frequency (GHz) Frequency (GHz)

P.2040-09

Figure 9 shows that the waveguide theory is applicable to realistic propagation characteristics in the
corridor in the frequency range of 200 MHz to 12 GHz at night. However, the waveguide theory is
not applicable to realistic propagation characteristics during daytime, because the received power is
attenuated by pedestrian traffic.
Therefore the waveguide theory is applicable to situations where there is no influence from
shadowing obstacles.
 Rec. ITU-R P.2040-1 19

2.3 Theory and results for frequency selective surface materials

2.3.1 Frequency selective surfaces
The power of scattering waves varies with roughness of surfaces. In this section, a theory for
calculating scattered fields from the surface having round convexity array is described. First, for
parameterizing the roughness of the surface, the rough surface is defined by using a round
convexity array formed by locating circular cylinders periodically.
Second, the reflection coefficient of the scattered fields is defined by using the lattice sums
characterizing a periodic arrangement of scatterers and the T-matrix for a circular cylinder array.
Third, a numerical result that shows the frequency-depending characteristic of the reflection from
the round convexity’s surface is shown. Finally, a measurement result is shown to explain that the
power of scattering waves varies with the frequency of an incident wave when there is a round
convexity array on the surface of a building.

2.3.2 Theory for wave propagation around the surface of round convexity array
By making periodical round convexity array on a surface of a building, as shown in Fig. 10,
reflection/scattering waves can be controlled larger than those from the flat surface. The theory to
calculate the scattered waves from the periodic arrays of circular cylinders can be used to define the
propagation waves around a convexity array of a surface.

FIGURE 10
The surface of round convexity array

P.2040-10

FIGURE 11
Geometry of a periodic array of circular cylinders
y

0 in


r d x

P.2040-11

When the identical circular cylinders are situated periodically in an x axis as shown in Fig. 11, the
power reflection coefficient R for the -th propagating mode with k > 0 is given as:

𝑘𝜈
𝑅𝜈 = k in
|𝒑𝑇𝜈 ∙ 𝒂sc
0 |
2
(51)
0 sin φ
20 Rec. ITU-R P.2040-1

where k0 = 20, 0 is the wavelength of the waves indenting in angle in. In equation (51),
𝒑𝑇𝜈 and 𝒂sc
0 are obtained as follows:

2(𝑗)𝑚 (𝑘𝑥𝜈 +𝑗𝑘𝜈 )𝑚

(𝑚 ≥ 0)
ℎ𝑘𝜈 𝑘0𝑚
𝒑𝜈 = [2(−𝑗)|𝑚| (𝑘 |𝑚| ] (52)
𝑥𝜈 −𝑗𝑘𝜈 )
|𝑚| (𝑚 < 0)
ℎ𝑘𝜈 𝑘0

𝒂sc ̅ ̅ ̅ −1 ⋅ 𝑻
0 = (𝑰 − 𝑻 ⋅ 𝑳)
̅ ⋅ 𝒂𝑖𝑛 (53)

where 𝑰̅ is the unit matrix, 𝑘𝑥𝜈 = −𝑘0 cos in + 2𝜈π/ℎ, 𝑘𝜈 = √𝑘02 − 𝑘𝑥𝜈 2 and h is the periodic

space between each round convex. 𝑳̅ is a square matrix whose elements are defined in terms of the
following lattice sums:

(1) in (1) in
𝐿𝑚𝑛 = ∑∞
𝑙=0 𝐻𝑚−𝑛 (𝑘0 𝑙ℎ)𝑒
𝑗𝑘0 𝑙ℎ
+ (−1)𝑚−𝑛 ∑∞
𝑙=0 𝐻𝑚−𝑛 (𝑘0 𝑙ℎ)𝑒
−𝑗𝑘0 𝑙ℎ
(54)
(2) ̅ is the T-matrix for the scattered
where 𝐻𝑚 is the m-th order Hankel function of the first kind. 𝑻
fields and is given by the following diagonal matrix for the incident electric field 𝐸𝑧𝑖𝑛 and the
incident magnetic field 𝐻𝑧𝑖𝑛 , respectively.
′ ′
𝐸 √ε𝑟 𝐽𝑚 (𝑘𝑑)𝐽𝑚 (𝑘0 𝑑)−𝐽𝑚 (𝑘𝑑)𝐽𝑚 (𝑘0 𝑑)
𝑇𝑚𝑛 =− ′ (1) ′(1) δ𝑚𝑛 (55a)
√ε𝑟 𝐽𝑚 (𝑘𝑑)𝐻𝑚 (𝑘0 𝑑)−𝐽𝑚 (𝑘𝑑)𝐻𝑚 (𝑘0 𝑑)
′ (𝑘𝑑)𝐽 (𝑘 𝑑)− ε 𝐽 (𝑘𝑑)𝐽 ′ (𝑘 𝑑)
𝐽𝑚
H 𝑚 0 √ 𝑟 𝑚 𝑚 0
𝑇𝑚𝑛 =− ′ (𝑘𝑑)𝐻(1) ′(1) δ𝑚𝑛 (55b)
𝐽𝑚 𝑚 (𝑘0 𝑑)−√ε𝑟 𝐽𝑚 (𝑘𝑑)𝐻𝑚 (𝑘0 𝑑)

where ε𝑟 is the relative permittivity of the dielectric cylinder, 𝐽𝑚 is the m-th order Bessel function,
the prime demotes the derivative with respect to the argument, and δ𝑚𝑛 denotes the Kronecker’s
delta. 𝒂𝑖𝑛 denotes a column vector whose elements represent unknown amplitudes of the incident
field.

in
𝒂𝑖𝑛 = [(𝑗)𝑛 𝑒 −𝑗𝑛 ] (56)

2.3.3 Calculation results

The calculation result of a power reflection coefficient is shown in Fig. 12. The result is calculated
by using equation (51) in the case that the electric field 𝐸𝑧𝑖𝑛 is transmitted in the angle in=90° to the
dielectric round convexities whose diameter and permittivity are d = 0.3h and r = 2.0, respectively.
In the result, there is the frequency band that the incident wave is reflected almost completely by the
surface even if its material is a lossless dielectric substance.
 Rec. ITU-R P.2040-1 21

FIGURE 12
Power reflection coefficient R0 as functions of the normalized wavelength h/0
at normal incidence electric field 𝑬𝒊𝒏
𝒛

R0
0.5

0
0.4 0.6 0.8 1
h/0

P.2040-12

2.3.4 Measurement
The measurements of the scattered waves from the building having the round convexity array were
carried out. Figure 13 shows the comparison of the scattered waves from the building between the
flat surface and the surface with round convexity arrays. The scattered waves from the building
were measured in various reflected angles r between 30° to 90°, when the electric field is
transmitted in the angle in. The incident angle and reflection angle are defined as shown in Fig. 14.

FIGURE 13
Geometry of a periodic array of circular cylinders

Round convexities ( h/ 0 = 1.37)

0 Round convexities ( h/ 0 = 0.73)
Flat surface
E z (dB)

–20
r

–40
40 60 80
r
 (degree)

P.2040-13
22 Rec. ITU-R P.2040-1

FIGURE 14
A plane figure of the compositional diagram
for measurements
y
in
 r


P.2040-14

The measurement results show that the power of the scattered field from the surface having a round
convexity array becomes larger than that from the flat surface, and can be controlled by the period
between and diameter of each round convexity. Note that the relative permittivity and the
conductivity of the building material were estimated as r = 6.0 and σ = 0.1 S/m, respectively.

3 Compilations of electrical properties of materials

Representative data on material electrical properties can be hard to find, as characteristics are
expressed using different combination of parameters, and the relative permittivity may be quoted at
frequencies that are not close to that of interest. A table of representative material properties has
therefore been compiled using the curve-fitting approach described in § 2.1.4.
Data from eight sets of material electrical properties (a total of more than 90 separate
characteristics) given in the open literature have been collated, converted to a standard format and
grouped into material categories.
For each group, simple expressions for the frequency-dependent values of the real part of the
relative permittivity, ', and the conductivity, , were derived. These are:

  a f b (57)

and:

c f d (58)

where f is frequency in GHz and  is in S/m. (' is dimensionless.) The values of a, b, c and d are
given in Table 3. Where the value of b or d is zero the corresponding value of  or  is a or c
respectively, and independent of frequency.
If required, the imaginary part of the relative permittivity " can be obtained from the conductivity
and frequency:

 17.98  / f (59)

Parameters for air, metal and three conditions of ground are included in Table 3 for completeness.
 Rec. ITU-R P.2040-1 23

TABLE 3
Material properties
Material class Real part of relative Conductivity Frequency range
permittivity S/m
a b c d GHz
Vacuum (≈ air) 1 0 0 0 0.001-100
Concrete 5.31 0 0.0326 0.8095 1-100
Brick 3.75 0 0.038 0 1-10
Plasterboard 2.94 0 0.0116 0.7076 1-100
Wood 1.99 0 0.0047 1.0718 0.001-100
Glass 6.27 0 0.0043 1.1925 0.1-100
Ceiling board 1.50 0 0.0005 1.1634 1-100
Chipboard 2.58 0 0.0217 0.7800 1-100
Floorboard 3.66 0 0.0044 1.3515 50-100
Metal 1 0 107 0 1-100
Very dry ground 3 0 0.00015 2.52 1-10 only
Medium dry ground 15 −0.1 0.035 1.63 1-10 only
Wet ground 30 −0.4 0.15 1.30 1-10 only

The frequency ranges given in Table 3 are not hard limits but are indicative of the measurements
used to derive the models. The exceptions are the three ground types where the 1-10 GHz frequency
limits must not be exceeded. Typical values of relative permittivity and conductivity for different
types of ground, as function of frequency in the range 0.01 MHz to 100 GHz, are given in
Recommendation ITU-R P.527.
The loss tangents of all the dielectric materials in Table 3 are less than 0.5 over the frequency
ranges specified. The dielectric limit approximations for the attenuation rate given in equations (24)
and (27) can therefore be used to estimate the attenuation of an electromagnetic wave through the
materials.

Attachment 1 (to Annex 1)

Alternative method to obtain reflection and transmission coefficients

for building materials represented by N dielectric slabs
based on ABCD matrix formulation

An alternative formulation of the method in § 2.2.2.1 is given below to obtain the reflection, R, and
transmission, T, coefficients for a building material represented by N dielectric slabs based on the
ABCD matrix formulation, as illustrated in Fig. 5. The regions on both sides of the building
material are assumed to be free space. This alternative method produces exactly the same results as
that given in § 2.2.2.1.
B Z 0  CZ 0 (60a)
R
2 A  B Z 0  CZ 0
24 Rec. ITU-R P.2040-1

T (60b)
T
2 A  B Z 0  CZ 0

Where A, B, and C are the elements of the ABCD matrix given, using matrix multiplication, by:
 A B   A1 B1   Am Bm   AN BN 
C D  C D ...C Dm  C N
...
DN 
(61a)
   1 1  m

where:
Am  cos(mdm ) (61b)

Bm  jZ m sin(mdm ) (61c)
j sin(md m )
Cm  (61d)
Zm

Dm  Am (61e)
1/ 2
 sin 2 0 
 m  k m cos ( m )  k m 1 –  (61f)
 m 

2
k0  (61g)

k m  k 0 m (61h)

 is the free-space wavelength, k0 is the free-space wave number, m and km are the complex
relative permittivity and wave number in the m-th slab, m is the propagation constant in the
direction perpendicular to the slab plane, and dm is the width of the m-th slab.
The wave impedances Z are given, according to incidence polarization, by:

120 
Zm  TE polarisation (62a)
m cos m

120  cos m
Zm  TM polarisation (62b)
m

where:
0  N 1 1 (63a)
0   N 1   (63b)
Z0  Z N 1 (63c)
 Rec. ITU-R P.2040-1 25

Annex 2

1 Introduction
This Annex provides definitions of terms relating to building loss, and gives guidance on
recommended measurement practices.
Report ITU-R P.2346 contains a compilation of the results of measurements of building entry loss.

2 Description of scenarios involving the outdoor-indoor interface

2.1 Outside-inside propagation: issues concerning entry-loss reference field

A difficulty with defining the entry loss reference field is that the presence of the building will
modify signal strengths outside it. Figure A2.1 illustrates, in somewhat simplified form, the issues
involved. The three sections of the figure show:
a) A relatively isolated outdoor point receives a direct and ground-reflected ray. In fact both of
these rays, in an urban environment, may well arrive from a distant source via diffraction
over a building to the left of the figure. For propagation at small angles to the horizontal,
there will be fairly simply and mainly vertical lobing, that is, maxima and minima when the
point is moved vertically.
b) Without moving the point, a building is placed just behind it. It now receives two additional
rays reflected from the building, one of which is also ground-reflected.
The lobing pattern will now have fine structure in both the vertical and horizontal
directions.
c) The point is now moved inside the building. For the purposes of illustration the frequency
is assumed to be high enough such that only rays entering a window are significant. At a
lower frequency, where penetration through the wall is significant,
the ray pattern would change.

FIGURE A2.1
Simplified ray diagrams for outdoor and indoor points

a)
Outdoor point

b)
Outdoor point in
front of building

c)
Indoor point

P.2040-A2-0 1

Although multipath propagation causes lobing, the power-sum of multiple rays approximates to the
spatially-averaged field. In general, therefore, the presence of a building behind a receiver can be
expected to increase the received signal strength. Inside the building, particularly close to the
illuminated external wall, a larger number of rays is likely to be received, although many will be
attenuated by transmission, reflection or diffraction. It is thus possible to have a stronger signal
inside than outside.
26 Rec. ITU-R P.2040-1

2.2 Propagation losses in the built environment

Figure A2.2 shows the different kinds of building losses encountered in an outdoor-indoor and
indoor-outdoor scenario. The definitions are given in the next sections.

FIGURE A2.2
Different kinds of propagation loss involving buildings

Indoor
Indoor

Entry loss Exit

loss
Losses
within
building Shadowing loss

P.2040-A2-02

3 Definitions

3.1 Definition of building entry loss

Building entry loss is the additional loss due to a terminal being inside a building.

3.2 Definition of building shadowing loss

The building shadowing loss is the difference between the median of the location variability of the
signal level outside the illuminated face of a building and the signal level outside the opposite face
of the building at the same height above ground, with multipath fading spatially averaged for both
signals. It can be considered as the transmission loss through a building.

3.3 Definition of (e.g. wall) penetration

Signals outside a building enter an enclosed building by penetration mostly through walls.
Wall penetration can also refer to the penetration through partitions inside buildings.
Inside buildings, wall penetration loss is the difference between the median of the location
variability of the signal level on one side of a wall, and the signal level on the opposite side of the
wall at the same height above ground, with multipath fading spatially averaged for both signals.
It can be considered as the transmission loss through a wall.

3.4 Definition of aperture penetration

Aperture penetration is the penetration of signals from one side of a wall to the other side through
openings on the walls like windows.

3.5 Definition of building exit loss

From reciprocity, the numerical value of building exit loss will be the same as the building entry
loss. In the remainder of this text the terms are used interchangeably.
 Rec. ITU-R P.2040-1 27

4 Measurement of building entry loss

4.1 Introduction
Building entry loss can be measured as the difference, expressed in dB, between the spatial median
of the signal level outside the illuminated face of a building and the spatial median of the signal
level inside the building at the same height above ground, shown as “h” in Fig. A2.3 below
(i.e., loss = spatial median external field – spatial median internal field, where measurements are in
decibel units). The purpose of the outside measurement is to approximate the field strength which
would exist at the indoor location if the building did not exist. Where the distance between the
outside and inside measurements is a significant portion of the overall path, the additional free
space loss should be allowed for.
The outdoor field should be measured as close as possible to the building while ensuring that
near-field effects are avoided and antenna characteristics are unaffected. Measurements made with
directional and omnidirectional antennas may be expected to give different results; antenna
characteristics should, in any case, be carefully described. Where it is not possible to measure the
outdoor field incident on the building a predicted value should be used and this should be clearly
stated.
Measurements should normally be performed with a line of sight (LoS) between the outdoor
terminal and one face of the building under test.

FIGURE A2.3
Location of reference and measurement antennas for building entry loss measurement

Distance sufficiently large to minimise variation of FSPL across

indoor measurement area, while maintening LOS between
outdoor terminal antenna and face of building

Indoor measurement
antenna Outdoor terminal
antenna
Outdoor reference
antenna

P.2040-A2-03

The area chosen for spatial averaging inside the building will depend on the particular application,
and should be clearly stated; room-averages have been found to represent a practical and useful
basis for discretisation.

4.2 Parameters to be recorded

The following parameters should be recorded when performing measurements of building entry
loss.
It is assumed that each measurement set will consist of a number of samples, with the results being
expressed as a tabulated cumulative distribution function of loss.
Researchers are asked to provide as much additional detail as possible; in particular, interior and
exterior photographs should be supplied wherever possible.
28 Rec. ITU-R P.2040-1

TABLE A2.1
Measurement parameters
Parameter Units or classification Notes
Frequency MHz
Bandwidth of test signal MHz 0 MHz if CW source used
Surrounding environment Open/suburban/urban/dense urban Required to estimate
importance of coupling via
energy scattered from other
buildings
LoS to building? Yes/No Should normally be LoS to
minimise measurement error
Averaging Spectral / spatial / other Free-format field to allow user
to describe form of averaging
(if any) used
Penetration depth 1 = indoor terminal in room/space with
external wall facing outdoor terminal
2 = indoor terminal in room/space with
no external wall
3 = indoor terminal in room/space with
other external wall
Floor on which Ground floor = 0
measurements made
Area within which samples Square metres
taken
Number of samples Sufficient number of samples
should be taken to provide for
statistical confidence in the
results
Reference 1 = measured median signal Measurement preferred where
2 = predicted free space path loss possible
Distance of outdoor metres
terminal from building
Elevation angle of path degrees
Minimum azimuth with degrees
respect to normal to
building face
Maximum azimuth with degrees
respect to normal to
building face
 Rec. ITU-R P.2040-1 29

TABLE A2.2
Building parameters
Parameter Units or classification Notes
Width metres Approximate footprint for irregular
building
Length metres
Height metres
Total number of floors
Thickness of exterior walls metres
Thickness of interior walls metres
Thickness of floors metres
Proportion of building elevation %
area composed of
windows/apertures
Window elements 0 = unknown
1 = single
2 = double
3 = triple
9 = other
Window coating 0 = unknown
1 = none
2 = metallised glass
3 = internal wire mesh
4 = metal blinds/shutters
9 = other
Metallic thermal insulation fitted? 0 = unknown
1 = no
2 = yes
9 = other
Floor material 0 = unknown
1 = wood
2 = metal
3 = concrete
9 = other
Primary exterior wall material 0 = unknown Material forming the greatest
1 = stone proportion of the exterior walls
2 = brick
3 = brick with cavity
4 = lightweight block
5 = wooden
6 = concrete
7 = glass
8 = metal
9 = other
Secondary exterior wall material 0 = unknown
1 = stone
2 = brick
3 = brick with cavity
4 = lightweight block
5 = wood
6 = concrete
7 = glass
8 = metal
9 = other
30 Rec. ITU-R P.2040-1

TABLE A2.2 (end)

Parameter Units or classification Notes
Internal walls 0 = no interior walls
1 = stone
2 = brick
3 = lightweight block
4 = wood
5 = concrete
6 = plasterboard (wooden stud)
7 = plasterboard (metal stud)
8 = metallised plasterboard
9 = other
Roof materials 0 = unknown
1 = concrete tiles
2 = slate tiles
3 = wooden shingles
4 = sheet metal
5 = wood with roofing felt
9 = other

______________