Propagation in free space

The ability of EM wave to propagate depends on the environment

parameters. (Vacuum, free space, obstructions, area topology, etc)

According Maxwells equations EMW radiate in all direction in the
speed of light (3.3 sec to travel 1km)

So we need to calculate the free space losses, which follow the
following equation
( )
2
4
log 10

d
L
fs
=

Where

d is the distance to travel
is the wavelength of the EMW

In logarithmic form

Lfs=32.44 +20log(f) +20 log (d)

Then

Lfs=Lo +10 log (d)

Where is the slope of attenuation as a function of distance to travel.

Now we have to calculate the Power at the end path of the
transmitted wave (Mobile). Considering a pure free space
propagation using isotropic antenna as a reference antenna. We
can find the following parameters

Pt: transmitting power
Pr: receiving power
Gt: transmitting antenna gain (linear)
Gr: Receiving antenna gain (linear)
D: Distance between transmitting and receiving antenna
: Frequency wave length

Then the receiving power will equal to the transmitting power
multiplied by the gain of the trans and receiving antenna multiplied by
the frequency divided by the distance between the transmit and the
receiving antenna



Taking the logarithmic for both sides

Pr(dbm)=Pt(dbm) +10 log (Gt) +10 log(Gr) 20 log (D) 20 log
f(MHz) X
Where X : 32.45 db if distance in Km
36.58 db if distance in miles
Example
Calculate the Pr when
Pt: 4.3 watt
Gt is 14 dbi
Gr is 3 dbi
D is 10 km
Freq is 850 Mhz
Wave length
Repweat when Freq is 6 Ghz













( )
|
|
|
.
|

\
|

=
2
4
Pr

D
Gr Gt
Pt


Propagation over flat earth

Now we have omni or directed antenna different heights of
transmitting and receiving antenna
Then we have more factors
Ht: transmitting antenna height
Hr: Receiving antenna height
Then the receiving power will be

( )
2
2
Pr
|
.
|

\
|

=
D
Hr Ht
Gr Gt Pt

There are two radios waves are reaching the mobile
The earth is super conductor no losses

In log form

Pr(dbm) =Pt(dbm) +10log(Gt) +10 log(Gr) +20 log(Ht) +20 log(Hr)
40 log(D)

Example
Calculate the previos example with
Ht is 150 ft
Hr is 6 ft
Gt is 12 dbi
Gr is 2 dbi
D is 4 miles
Pt is 4.3 watt










Knife edge model

As practical exercise for the rf propagation in urban and dense urban
areas the rf wave propagates depending on a lot of diffractions; so
Knife edge model is created so that to include the factor of diffraction

We have new parameters
D1: distance between the tx antenna and obstruction
Df2: distance between the obstruction and the rx antenna
H: difference between (the los between the top of the obstacle and
the tx antenna) and (the los between the tx antenna and the rx
antenna)

Steps for calculating the knife model
Calculate the Fresnel-Kirchoff diffraction parameter

( )
|
|
.
|

\
|

+
=
2 1
2 1
2
d d
d d
h

h=Height of the obstacle height of the rx antenna

Example
Calculate the additional losses due to the presence of the Knife-edge
obstruction
Repeat if freq is 6 Ghz













Multiple KED: Epstein Paterson model

Loss from each obstruction is calculated separately
Total loss is the summation of all the calculated losses

Example
Calculate the total losses for the last fig
D1 Is 0.6 km
D2 is 0.75 km
D3 is 1.2 km
D4 is 1.3 km
Tx height is 40 m
Rx height is 2 m
H1 is 46 m
H2 is 48 m
H3 is 25 m
























Foose factor

The attenuation of the obstruction varies as a function of the actual
height and shape
Foose factor is a empirical correction for the conservative behavior of
the Esptein Paterson model

f os
F n RSL =


nos is number of obstacles
Ff the foose factor correction (Db/obstruction)

Example
Assume the foose factor is 3.9 db/obs
Find the losses due to the last example
























Classical propagation model

Propagation models are essentially curve fitting exercises.
Propagation tests are conducted at different freqs, antenna heights
and locations over different periods and distances
The received signal is analyzed using mathematical tools and is fitted
to an appropriate curve. Formulae to match these curves are then
generated and used as models.
Frequently used Propagation models
Okumara
Okumara - Hata
Cost 231 Hata
Walfisch-Ikegame
Cost 231- Walfisch-Ikegame
Xlos (Motorola Proprietary model)























Okumara

In the early 1960s, a J apanese scientist by name Okumara
conducted extensive propagation test for mobile systems at different
frequencies. The tests were conducted at 200, 453,922,1310,140,
1920 MHz. The tests were also conducted for different BTS and
Mobile antenna heights, at each freq, over varying distances between
the BTS and the mobile

The Okumara test are valid for
150-2000MHz
1-100Kms
BTS heights of 30-200m
MS antenna heights, typically 1.5m

These results of Okumara tests were graphically represented and
were not easy for computer-based analysis.

Fig 1 Okumara test results

Hata model

Hata took Okumara data and derived a set of empirical equations to
calculate the path loss in various environments. He also suggested
correction factors to be used in Quasi-open and suburban areas

The general path loss equation is given by:

( ) ( ) ( ) { }
o BTS m BTS p
Q d h h a h f Q Q L + + + = ) log( ) log( 55 . 6 9 . 44 log 82 . 13 log
2 1

Where

=
p
L path loss
= f Freq in MHz
= d Distance between BTS and the mobile (1-20Km)
=
BTS
H
Base station height in meters (30 to 100 m)
=
m
H
Mobile height (1-10m)
( ) = hm a Correction required if mobile height is more than 1.5 meter
and is given by
( ) ( ) { } ( ) { } 8 . 0 log 56 . 1 7 . 0 log 1 . 1 = f h f hm a
m
For urban areas
( ) ( ) { } 97 . 4 75 . 11 log 2 . 3
2
=
m m
h h a
For dense urban areas
55 . 69
1
= Q
For freq (150-1000 Mhz)
3 . 46
1
= Q
For freq (1500- 2000 MHz)

16 . 26
2
= Q
For freq (150-1000 Mhz)
9 . 33
2
= Q
For freq (1500- 2000 MHz)

0 =
o
Q
For Urban area
3 =
o
Q
For Dense urban area


Corrections to Okumara-Hata Model

The formulae we used for the Okumara model were the urban
environment. The model gets modified if the environment is semi
urban, suburban or open area.

Corrections
4 . 5 ) 28 / ( log 2
2
) (
= f l L
Basic p sub
94 . 40 ) log( 33 . 18 )) (log( 78 . 4
2
) (
+ = f f L L
basic p open
94 . 35 ) log( 33 . 18 ) (log( 78 . 4
2
) (
+ = f f L L
Basic p Quasi

Example
Use Okumara model to find the RSL at 2.3 miles from the BTS
operating ate 870 MHz using the following data

Radiation center line of the BTS =40m
Height of the MS =3m
Terrain elevation at the location of the BTS =340m
Average height of the terrain =312m
Transmitting power =19.5 w
BTS gain =10 db
Ms gain =0 db
Area =Urban area

Solution
The free space loss is calculated by the following equation
Lfs=32.45 +20log(2.3*1.609)+20log(870)-10=92.61db

The basic median attenuation is determined from figure
Amu =24 db

The effective height of the BTS transmitting antenna
hte =40+340-312=68 m

Correction for the base station antenna height is determined from
figure
Htu=-12 db

The total path loss is equal to
Pl=92.61 +24+12=128.61

RSL=10log(19.5*1000) 128.61=-85.7dbm


Example
Use the pervious data to get the RSl by the Hata equations

Solution
The mobile antenna height gain can be obtained as:
a(hm)=(1.1log(870)-0.7)*3 (1.56log(870)-0.8)=3.81 db
Qo=0 urban area
The effective height of the BTS transmitter is given as
hte=68m
RSL= 10log(19.5*1000) + 10 +69.55 26.16log(870) + 3.81 +
13log(68) (44.9 6.55log(68)) log(2.3*1.609) =-84.61 dbm

If you make comparison between the two solutions there is a
negligible amount.

Example
Calculate the cell radius of the site has the following parameters
Freq=900 MHz
BTS height=30m
Mobile Height=3m
BTS ERP=55dbm
RSL at the mobile =-75 dbm
Attenuation slope=3.5

RSL=ERP Lps
-75=55-Lps
Lps=130 dbm
Lps=Lo +10log(d)
Lo=69.55 +26.16log(900) 13.82 log(30) a(hm)
a(hm)=1.1(log(900)-0.7)*3 (1.56log(900) 0.8)
Lo=122.57 db

From RSL =ERP (Lo +10log(d))

Then 130 122.57 =10log(d)
Where =3.5
Then d =1.629 Km



































Cost 231-Hata model

The cost 231-sub group on propagation models proposed an
improved propagation model for urban areas to be applied above
1500 MHz
Hence this model can be used for planning DCS 1800 systems.
Like Hata model, the cost 231 Hata model is also based on the
measurements of Okumara.
The general path loss equation for Cost 231-hata model is identical to
the Hata model, excepting that the constants Q1 and Q2 have
different values.

The general path loss equation is given by:
{ }
0
2
2 1
) log( ) log( 55 . 6 9 . 44 ) ( ) log( 82 . 13 ) log( Q d h h a h f Q Q L
BTS m BTS p
+ + + =


3 . 46
1
= Q For freq from 1500 to 2000 MHz
9 . 33
2
= Q For 1500 to 2000 MHz
0
0
= Q For urban
3
0
= Q For dense Urban



















Walfisch-Ikegami Model

Line of sight propagation:
This is useful for dense urban environment. The model is essentially
based on Okumara studies but takes into account several factors like
building density, average height of buildings, street widths and so on.

The simplest model is to assume that the antenna height for the BTS
is generally below rooftop so that the signals are guided through a
street canyon. It is assumed the there a LOS between the MS and
the BTS.








The path loss is given by:
) log( 26 ) log( 20 6 . 42 d f L
LOS
+ + =


This can be written as LOS equation as:
) log( 10 d L L
o LOS
+ =


Where, Lo=42.6 +20 log (f) and is 2.6

The value of attenuation slope is 2.0 in free space. In dense urban
environment, under LOS conditions, signals are actually guided
between, thereby causing a wave-guide effect. (Which is similar to
free space propagation).


























COST 231 Walfisch-Ikegami Model
NON LOS model
Here, we assume that the BTS antenna is above roof level for any
building within the cell and that there is no LOS between the BTS and
the MS

We define the following parameters

W Is the distance between street MS and Building
m
h
MS height
B
h
BTS antenna height
r
h
Roof height
B
h
Difference between rooftop and the BTS antenna height
m
h
Difference between MS and rooftop



For the sake of simplicity, we assume that the environment has
buildings of uniform height. For MS on the street, the signal
undergoes diffraction from rooftop and also multiple diffraction due to
the surrounding buildings.


The general form is
LMD RFT FS p
L L L L + + =


Where

=
Fs
L
Free space loss=32.44 +20log(f) +20log(d)
=
RFT
L
Roof top diffraction loss
=
MDB
L
Additional loss due to multiple diffraction due to
surrounding buildings

) ( ) log( 20 ) log( 10 ) ( 10 9 . 16 L h f W lof L
m RFT
+ + + =

where
m r m
h h h =

L()=Losses due to elevation angle

357 . 0 10 ) ( + = L
For 0<<35

) 35 ( 075 . 0 5 . 2 ) ( + = L
For 35<<55

) 55 ( 114 . 0 4 ) ( + = L
For 55<<90

The losses due to multiple diffraction and scattering components due
to buildings are given by:
) log( 9 ) log( ) log( W f K d K K K L
f d a o MBD
+ + + =


where
) 1 log( 18
B o
h K + =

) ( 8 . 0 54
B a
h K =

) ( 15 18
r
b
d
h
h
K

=

)
`

|
.
|

\
|
+ = 1
925
7 . 0 4
f
K
f For suburban areas
)
`

|
.
|

\
|
+ = 1
925
5 . 1 4
f
K
f For urban areas

B
h
BTS antenna height

r B B
h h h =


For simplified calculations, we can assume Ka=54 and Kd=18.



Fresnel Zone

In Multipath environment, the signal suffers from unpredictable deep
fades. To facilitate an estimation of the Multipath effects, we have
studied various propagation models
In all cases we have seen that the path loss can be characterized by
the straight-line equation

) log( 10 d L L
o P
+ =


From the above equation we can find that there is a certain
combination of parameters for which the path loss is not sensitive to
. But it is not possible to achieve this in reality

However, there exists a Fresnel zone point within which LOS
conditions are met and beyond which Multipath effects come into play

Such a point is called Break Point.



To explain the concept of break point we use a 2 ray model.
We are going to define the following parameters

hb BTS antenna height
hm MS height
d distance between the BTS and MS
d1 reflected path
d2 direct path



2 1
d d d =

( ) { } ( ) { }
2
1
2
2
2
1
2
2
d h h d h h d
m B m B
+ + + =


Using binomial series we can simplify the above equation as
d h h d
m B
/ 2 =

When
d h h
m B
/ ) (
is <<1

Fresnel Zones are a set on concentric circles, which are loci of all
points having the same signal strength. The Fresnel zone are /2
apart from each other. The diffraction is maximum when the
difference between the direct ray and the diffracted is /2.



Then we can write
2 / / 2 = = d h h d
m B
Or
/ 4
m B o
h h D =

Now Do is the break point
The path loss slope is similar to LOS path loss within the break point.
Diffractions and Multipath phenomena usually happen beyond this
point.