TLT-6407 Section 3

TAMPERE UNIVERSITY OF TECHNOLOGY
Institute of Communications Engineering
TLT-6406 RADIO NETWORK PLANNING
Section 3
GSM coverage planning
Location probability
 Components of fading
 Slow fading is caused the obstacles near the mobile stations
 All three mobiles (A, B, C) are suffering slow fading because they
are the shadows of the buildings
 Fast fading occurs when the signal is travelling through several paths having
different lengths to the receiver
 Mobile B suffers fast fading due to reflections and mobile C due to
reflections and diffraction
A B C
TLT-6406 Radio Network Planning 2

 An example of a fading signal. Thicker line shows the slow fading
component of a signal and thinner line the original received signal.
-30
-35
-40
-45
Signal level [dBm]
-50
-55
-60
-65
-70
-75

 Fast fading component in a signal can be extracted by using
averaging window of 32 wave lengths
 x is the random variable (here the slow fading signal),  is the
mean value of x and  is the standard deviation (STD) of x
 typically  =7-8 dB @ 900 MHz [1]
( x )2
1 
px  e 2 2
2   2

Point location probability
 Example 1. Distribution of the received signals at a certain point.
The average field strength at the given distance is –82 dBm and standard deviation is 8 dB.
Average () = -82 dBm
Deviation () = 8 dB
 -  = -82 dBm – 8 dB = -90 dBm
 +  = -82 dBm + 8 dB = -74 dBm
 68% of samples fall between –90 dBm and –74 dBm.
Probability
-110 -106 -102 -98 -94 -90 -86 -82 -78 -74 -70 -66 -62 -58 -54
Sig na l level [d Bm ]

 Cumulative distribution of example: 50% point is at –82 dBm.
 Slow fading margin as a function of location probability. =8 dB.

 Example 2. Probability to receive a certain signal level at a certain
point.
Location probability of 99% corresponds to slow fading margin of 18 dB if
standard deviation is 8 dB.
Slow fading margin of 18 dB is calculated by subtracting –100 dBm from
–82 dBm.

 Slow fading margin as a function of location probability. =8 dB.
15.0
10.0
5.0
Slow Fading Margin
0.0
20 % 30 % 40 % 50 % 60 % 70 % 80 % 90 %
-5.0
-10.0
-15.0
Lo c a tio n p ro b a b ility [%]

Area location probability (single cell)
 Point location probability 50% => 74 % => Area location probability 74% =>
90 %
 Description of the slow fading margin. Ph is point location probability, x0 the
minimum accepted signal level (e.g. sensitivity level) and µ is the mean value of
x0.
Location probability Ph = 50%

µ-x0 = 0 dB
R
Location probability Ph = 74%

µ-x0 = 5.1 dB

Auseful defines the useful service area within a circle of radius R where the signal
level received by the mobile station exceeds a given threshold x0. If px is the
probability that the received signal r exceeds x0 in an area dA, then the useful area
can be written as [3]
1
2  x0
Auseful   p dA
 R
where
 (r  )2
1  1 1 x 
px0  
x0 2   2
e 2 2
dr    erf  0
2 2

  2 

In macrocellular environment the mean value of the received signal strength, , can
be written as
d
 (d )  Fcell edge  10 log10  
 R
where Fcell edge is field strength on the cell edge in dBµV/m and d is the distance
from base station to some point inside the cell coverage area, R is the distance
from the base station to the cell edge, and  describes how the average received
signal strength depends on the distance from base station (propagation
exponent). The value of  varies normally from 2.5 to 4.5. In propagation
models  is also used as a key parameter. For example, in Okumura-Hata model
the value of  is expressed as shown below. As it can be seen  depends on the
height of the base station antenna hb. [4]
  44.9  655
.  log10 (hb )

Substituting , probability px can be presented as
  r 
 x0  Fcell edge  10    log10   
p x0    erf 
1 1  R
2 2   2 
 
 
If making the following substitutions
( x0  Fcell edge )  e 
a b  10    log10  
 2   2

Point location probability is defined in one point on the cell edge. Location
probability over a cell area takes into account the fact that signal level gets
higher closer to the transmitter.
1  a  b  1  
 2ab 1 
    
Auseful  1  erf (a)  e  b2 
 1  erf   
2
   b   
Location probability on
the cell edge
Location probability over

the cell area

 Examples of required slow fading margins are calculated for point
location probability and area location probability with standard
deviations of 6 dB, 7 dB and 8 dB
 In case of area location probability slope is assumed to be 33.8
dB/dec
Deviation [dB]
Point location probability Area location probability
Loc.Prob. 6 7 8 6 7 8
10 % -7.69 -8.97 -10.25 -19.34 -20.18 -21.11
20 % -5.05 -5.89 -6.73 -14.14 -14.85 -15.59
30 % -3.15 -3.67 -4.2 -10.92 -11.45 -11.99
40 % -1.52 -1.77 -2.03 -8.42 -8.76 -9.09
50 % 0 0 0 -6.24 -6.39 -6.5
60 % 1.52 1.77 2.03 -4.18 -4.11 -4
70 % 3.15 3.67 4.2 -2.08 -1.76 -1.4
75 % 4.05 4.72 5.4 -0.95 -0.49 0.01
80 % 5.05 5.89 6.73 0.29 0.91 1.56
90 % 7.69 8.97 10.25 3.44 4.48 5.56
95 % 9.87 11.51 13.16 5.95 7.36 8.8
96 % 10.5 12.25 14.01 6.68 8.19 9.73
97 % 11.28 13.17 15.05 7.56 9.2 10.87
98 % 12.32 14.38 16.43 8.72 10.54 12.38
99 % 13.96 16.28 18.61 10.54 12.63 14.74

Multiple server location probability
 In previous coverage analysis only a single cell was assumed
 In a real network, multiple base stations provide coverage, and
probability for having sufficient signal level is higher
 Assuming that the signals from different cells do not correlate:
 A  B   A  B
 For example:
A  B  A  B 
50%  50%  50%  50%  75%

 Single cell location probability compared to multiple server
location probability; zero correlation between signals assumed
1,00
0,95
0,90
0,85
0,80
0,75
0,70
0,65
0,52
0,56
0,60
0,65
0,69
0,73
0,77
0,80
0,84
0,86
0,89
0,91
0,93
0,95
0,96
0,97
0,98
0,98
0,99
0,99
Single cell area location probability

Indoor location probability
 The differences between outdoor and indoor location probability
calculations are
 When calculating indoor location probability standard deviation is
replaced by standard deviation of signal measured indoors
 Average building penetration loss is added to the calculated slow fading
margin.
 Because in indoor locations the standard deviation of signal is
higher than standard deviation outdoors, for a given location
probability the slow fading margin is higher

Location probability and environment
 Conclusions/notes about location probability
 STD has an influence on the location probability (point & area, single &
multiple)
 STD varies as a function of frequency (900/1800 MHz) and
environment type
 STD is different for macro cell and micro cell planning as well as for
outdoor and indoor
 Propagation slope (environment type; urban/rural) has an effect on the
location probability
 Number of received signals in case of interference?
STD of one signal
0 dB 6 dB 8.3 dB 12 dB
n mean std mean std mean std mean std
1 0,00 0,00 0,00 6,00 0,00 8,30 0,00 12,00
Number of
2 3,00 0,00 4,58 4,58 5,61 6,49 7,45 9,58
received
3 4,50 0,00 6,90 3,93 8,45 5,62 11,20 8,40
signals 4 6,00 0,00 8,43 3,54 10,29 5,08 13,62 7,66
5 7,00 0,00 9,57 3,26 11,64 4,70 15,37 7,13
6 7,50 0,00 10,48 3,04 12,69 4,41 16,74 6,74

Power budget to planning system
DOWNLINK
BASE STATION UNIT VALUE

RF Power dBm 45 A
Combiner loss dB 3 B
Power amplifier dB 4 C
Diversity TX/RX Cable loss dB 3 D
Antenna Antenna TX antenna gain dBi 16 E
Peak EIRP dBm 59 F = A-B+C-D+E

Antenna
Height
Antenna
Cable MOBILE STATION UNIT VALUE
RX antenna gain dBi 0 G
Cable loss dB 0 H
MS sensitivity dBm -105 I
Base
Station
Inteference degradation margin dB 1 J
Minimum reception level dBm -104 K = -G+H+I+J
Isotropic path loss dB 163 L = F-K

Planning margins
 Building / in-vehicle penetration loss
Verified with measurements
 Body loss
Human body affects the antenna radiation pattern
 Antenna orientation loss / polarisation loss
 Interference degrade margin / additional fast fading (TU3)
 Slow fading margin

Building penetration loss (BPL)
 Building loss varies based on
 Different types of outside wall construction
 Changes in floor elevation
 Propagation environment near the building
 Different percentage of window areas in the outside walls
 Different building orientations with respect to the direction of the antenna
illumination
 Different types of window treatment used to reflect sunlight and heat
Floor
Building 1 Basement 1st 2nd 3rd 4th 5th 6th 7th 8th 9th 10th
BPL [dB] 40 31 42 39 37 27 24 15 --- --- ---
Building 2 Basement 1st 2nd 3rd 4th 5th 6th 7th 8th 9th 10th
BPL [dB] 22 --- 22 --- 20 --- 20 --- 19 --- 17

Planning threshold
 Example of downlink planning threshold at 900 MHz in an urban environment
 Building / in-vehicle penetration loss + 20  10 dB , 10 dB = -64 dBm
 Other losses (body loss, antenna orientation loss) + 5-10 dB , 5-10 dB = -84 dBm
 Interference degrade margin / additional fast fading + 1-3 dB = -94 dBm
 Slow fading margin + 5 dB (90%) = -97 dBm
 Mobile sensitivity level - 102 dBm
Indoor Outdoor (BPL = 0 dB)

Location Probability over Cell Area 90.0% 80.0% 70.0% 90.0% 80.0% 70.0%
Slow Fading Margin + BPL (dB): 22.8 17.9 14.4 4.5 0.9 -1.8
Coverage Threshold (dBm): -77.2 -82.1 -85.6 -95.5 -99.1 -101.8

Learnings
TOPIC LEARNINGS
1. Defines probability when signal is over a given
threshold over a cell area
2. There can be different definitions!

Slow fading
1. Due to obstacles like buildings, hills, etc
2. Also call log-normal fading

Slow fading maring
1. Used to increase location probability
2. Standard deviation affect the margin

Body loss
1. Loss due to body proximity
Building penetration loss
1. Varies due to materials, thichness of walls, etc.
Coverage thresholds
1. The targeted field strenght value in planning
system
Link budget balancing
1. Coverage planning must be done according to
the weaker direction
Downlink prediction 1. Almost all propagation models work in downlink
direction
Additional fast fading margin 1. Compensates sensitivity in challanging
environment

Propagation prediction
Fundamentals
In free space the attenuation of propagating signal corresponds to
LFree Space ~ 20 log 10 d 

When radio waves travel over land they interact with the earth’s surface. The
received signal is a complex sum of all its components. The following equation
presents the sum of the components.
 
E  E0 1  Rc e i  1  Rc F e i  ...
E is the received signal, E0 is theoretical field strength for propagation in free
space, Rc is the reflection coefficient, F() the complex surface wave attenuation,
 is the phase difference between the reflected and the direct wave and i is the
imaginary number -1. Equation can be simplified if it is used at high frequencies,
becauseF()<<1 is valid for the most of situations. This assumption can be made
if using typical antenna height in transmitting and receiving ends. The simplified
equation, so called 2-ray model, is as follows:

E  E0 1  Rc e i 
Fundamentals
The path loss of the presented signal is

L ~ 20 log 1  Rc e i 
In the most cases (d >> 5hthr, where ht is height of transmitting antenna and hr is
the height of receiving antenna), the phase difference  can be approximated
4ht hr

d
Also, the reflection coefficient can be approximated by assuming that Rc  -1.
This assumption is valid for the most terrain types. If studying propagation bit
further from the transmitter, it can be assumed that d >> 4 hthr. Then the total
path loss can be approximated
ht hr
L  20 log 10
d2

Knife-edge diffraction
Diffraction allows radio signal to propagate behind obstacles. Diffraction parameter
defines the attenuation,
2 1 1 
  hm   
  dt d r 
according to equation
6  9  1.27 2 0    2.4
Ad  
13  20 log 10    2.4
The total path loss is thus given as
Atotal = Afree space + Ad
Diffraction in knife-
edge obstacle.
hm
dt dr

Multiple diffractions
 To improve accuracy, multiple knife edge diffraction calculation is applied.
 For example, Deygout method can be used up to three edges. The procedure
goes as follows. [1]
 Find the obstacle giving highest loss
 Find the other obstacles (up to two) of each sub-path and calculate losses
 Calculate the total loss as a sum of all the calculated losses
Deygout method.

Propagation models
 Coverage prediction, and hence the radio network planning, is
based on adequate signal level predictions
 Classification of propagation models
 Empirical models
 Equations derived from massive field measurements
 Macrocell models
 Low computation, low accuracy
 Semi-empirical models
 Theoretical model derived assuming some physical ideality
 Microcell / small macrocell models
 Average computation, average accuracy
 Deterministic models
 Theoretical, based on numerical such as ray optical methods
 High computation, high accuracy

Empirical models
 Propagation models which are commonly used for macro cells –
Okumura-Hata and Juul-Nyholm
 These models are developed by combining propagation theory and
extensive measurement campaigns
 The models take into account several parameters like
 Effective antenna height
 Terrain type (morphology) and terrain height (topography)
 Frequency, EIRP..
 These two models are macrocell models which have their
limitations in terms of frequency, calculation ranges, and base
station antenna heights

Okumura-Hata model
L  A  B log10 f  13.82 log10 hb  a( hm )  ( 44.9  6.55 log10 hb ) log10 d  Cm
150 – 1000 MHz 1500 – 2000 MHz

A 69.55 46.3
B 26.16 33.9
a(hm )  (1.1log10 f  0.7)hm  (1.56 log10 f  0.8) for small/medi um city

a(hm )  3.2(log10 (11.75hm )) 2  4.97 for large city
L Pathloss [dB]
A Constant
B Constant
f Frequency [MHz]
hb Base station effective antenna height [m]
hm Mobile station antenna height [m]
d Distance between base and mobile station [km]
Cm Area type correction factor

Limitations of Okumura-Hata
 Frequency range 150-1000 MHz and 1500-2000 MHz
 Transmitter antenna height between 30 and 200 meters
Base station antenna has to be above average roof top level
 Calculation range 1-20 km

Effective BS antenna height
hBS
h > 0
h  0
heff= hBS+ h
heff= hBS
h < 0
hBS
Base station effective antenna height calculation using the method suggested by CCIR.

Effective BS antenna height
2A
heff 
d
B C
A= B + C
d
Alternative method for calculating the effective antenna height.

Juul-Nyholm model
F  10 log10 P  k 0  k1 log10 hb  (k 2  k3 log10 hb ) log10 d
F Field strenght [dBµV/m]

Range 1 - 20 km Range 20 - 100 km
P Effective radiated power (ERP) [W] k0 23 30
k0 …k4 Coefficient (see Table 5.2.) k1 13 27
k2 47 50
hb Base station antenna height [m] k3 -7 3.6
d Distance between mobile and base station [km]
 Frequency range 800-1000 MHz and 1700-2000 MHz,

 Transmitter antenna height between 30 and 150 meters
 Calculation range 1-100 km.

Semi-empirical models
 For microcells or small macrocells
 More precise description of the environment than empirical models
 Typically based on physical optics and some assumptions made for geometry of
buildings
 More accurate than empirical models but require also more computation time
 Walfisch-Ikegami
 Separated formulas for LOS and NLOS
 Particularly suited for urban cells
 Diffraction taken into account
 Empirical corrections included

Deterministic models
 Microcell and picocell/indoor
 Very detailed and accurate input parameters
 Generally based on ray optical (ray tracing/launching) and some diffraction methods
(UTD, GTD) or finite difference/volume time domain (FDTD/FVTD) method
 Propagation can include
 LOS
 Reflected
 Transmitted TX''
 Diffracted
 Scattered
 and combined rays
 The most time-consuming part is to find TX
relevant paths from transmitter to receiver RX

 Can be applied in all kinds of environments
TX'
 Very good accuracy
 Lots of computation time needed
 Emerged as a highly promising models

Propagation measurements
 Measurements are needed in order to tune the propagation model
GPS receiver
Dead Reckoning Laptop computer Receiver
Speed sensor Compass
A block diagram of a measurement system.

 Measure only interference free frequencies
 Measure only in the main lobe of the transmitting antenna
 Avoid or erase LOS measurement points
 Use differential GPS if possible or match the coordinates with the
map
 Check coordinate conversion parameters
 Measure all the cable losses (both in transmitting and receiving
end)
 Measure the output power of the transmitter
 Check transmitter antenna installation and ensure that there are no
obstacles nearby
 Document the measurements very carefully

 Example of offset in position
Measured
road
Measurement
points

Prediction model tuning
Site and cell data Digital map System information
Coordinates
Calculate measurement route
Map matching
Model tuning
Measurement data
Field strenght
Compare
Analysis
No
Satisfactory model
Yes
End
Prediction model tuning algorithm

 Prediction model tuning areas
 Propagation slope
 Effective antenna height
 Morphographic corrections
 Calculation distance
 Topographic corrections
 Street orientation

 Assessment of propagation slope
 Okumura-Hata, correction factor C:
L  A  B log 10 f  D log 10 hb  (C  6.55 log 10 hb ) log 10 d
 Effective antenna height

 Okumura-Hata method specifies the effective antenna height as the
difference between the transmitting antenna and the average terrain height between 3
and 15 km. This definition is valid when calculation distance is more than 15 km. If the
distance is shorter than the antenna height is defined as follows.
 0 – 3 km: the average terrain height is calculated from base station to mobile station.
The effective antenna height is the difference between the absolute antenna height and
the average terrain height.
 3 – 6 km: the average terrain height is calculated as a sliding average over 3 km from
the mobile station towards to the base station.
 6 – 15 km: the average terrain height is calculated from 3 km (from base station) to
the mobile station.

 Example of morphographic corrections
 The distance between the base station and the mobile station is 1.5 km. On the
digital map there are 30 pixels (50 m x 50 m) between the base station and the mobile.
Each pixel presents the terrain type within the 50 m x 50 m area. The
morphographic correction is calculated as an average of the pixels between the
mobile station and base station. In Figure, there are 6 urban pixel, 9 suburban pixels, 4
park pixels, 6 open pixels and 5 water pixels between the mobile and the base station.
The average of the correction factors in this example is –9.4 dB. The basic propagation
model is corrected by adding the calculated correction to the prediction result
(correction factor Cm in Okumura-Hata model).
30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
Terrain type U U U O O U U U O O O O S S S S P P P P W W W W W S S S S S
Correction factor [dB] 0 0 0 -15 -15 0 0 0 -15 -15 -15 -15 -5 -5 -5 -5 -8 -8 -8 -8 -23 -23 -23 -23 -23 -5 -5 -5 -5 -5
Pixel size: 50 m x 50 m
The following notations are used: U = Urban, S = Suburban, P = Park, O = Open and
W = Water.
Example: morpho corrections tuning
-40 90
Measured
Predicted 80
-50
70
-60 60
Signal level [dBm]
50
-70
40
-80 30
20
-90
10
0
-100
-15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
1 51 101 151 201 251 301 351 401 451 501
dB
Measurement points

 Calculation distance
In the previous example, there are few remarks that must be made. The
distance between the base station and mobile was 1.5 km. It is not very likely
that the area close to the base station has a great impact on the received power
of the mobile station. The areas close to the mobile are more important for the
prediction, thus there are ways to weight the areas close to the mobile station.
First, the calculation distance can be shorter than the distance between the
mobile station and the base station. This means that only the pixels close to the
mobile stations are considered when calculating the correction. If in the
previous example the calculation distance is changed from 1.5 km down to
500 meters the average of the correction factors is –14 dB. In this case the
difference between the corrections is 4.6 dB which means that the selection of
calculation distance is important.

 Calculation distance can also be weighted average
Calculation distance
1.0 1.0
2.0
1.0
10 9 8 7 6 5 4 3 2 1
Terrain type W W W W W S S S S S
Linear weights for terrain type
Correction factor [dB] -23 -23 -23 -23 -23 -5 -5 -5 -5 -5
correction factors (example). The
Weights 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9
average of the normalized
correction factors is –12.33dB. Normalized weights 0.67 0.73 0.80 0.87 0.93 1.00 1.07 1.13 1.20 1.27
Normalized correction factors -15 -17 -18 -20 -21 -5 -5.3 -5.7 -6 -6.3

 Topographic corrections
The Okumura method for calculating topographic correction include
undulation
slope
diffraction parameters
 The parameters for undulation corrections are presented below.

The correction is done with the aid of two curves defined by the
parameters.
Rolling Hilly Terrain Correction

Correction Fine Correction
Kh Undulation [m] Khf Undulation [m]
0 0 0 0
-6 50 9 50
-10 100 12 100
-10 300 12 300
Calculation distance 10 km

Correction
5
Kh (dB) -5
-10
-15
0 50 100 300
Undulation (m)
Fine correction
20
15
Khf (dB)
10
0
0 50 100 300
Undulation (m)
 Slope correction is calculated for the given calculation distance
using the specified step. For each point, the slope in milliradians is
calculated and the average slope is defined. The slope is calculated
by using the curves defined by the equations given below. The
value of the slope is defined for both negative and positive slopes.
Note that the slope is defined for distances less than 10 km and
more than 60 km. For the distances between 10 km and 60 km a
linear interpolation is used.
Positive slope Negative slope

d < 10 km K slope  k1  x ( mr ) K slope  k 2  x ( mr )
K slo p e  k 3  x ( mr ) K slope  k 4  x ( mr )
d > 60 km

20.0
d < 10 km
d > 60 km
10.0
Slope correction [dB]
0.0
-10.0
-20.0
-20 -15 -10 -5 0 5 10 15 20
Average angle of slope [mr]
Slope corrections for different distances

 Clearance angle
Negative angle
Slope correction is based

on the negative or positive
angle.
Positive angle
0
dB
Slope corrections -2
-4
-6
-8
-3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1
milliradians

 Street orientation (in Walfish-Ikegami)
 10  0.354 0    35

L  2.5  0.075(  35) 35    55
4.0  0.114(  55) 55    90


Transmitter
0
Correction (dB)
-2
-4
Slope corrections
-6
-8
-10
-12
0 10 20 30 40 50 60 70 80 90
Angle (deg.)

Example: quality of tuning
-40
-50
-60
Signal level [dBm]
-70
-80
-90
-100
100 1000 10000
Distance [m]

Example: tuning results
Morpho Class Value [dB]
Open -20
Water -25
Forest -11
Quasi-Open -5
Houses -12
Sub-Urban -10
Urban -2
Buildings 7
Industrial buildings -4
High rise buildings 18

Tuning process
Correction factors
Basic path loss

model Select correction factors
to be tuned
Modify model
Set upper and lower limits
Recalculate model
Tune the correction factors
Analyze results
Analyze the results
OK NO
OK? OK NO
OK?
Save the model Forget the model

Save the model Forget the model
End
End
Model tuning Correction factor tuning

Digital maps
 One of the key elements of successful coverage predictions
 Topography, morphography, building rasters, roads

Digital maps
 Problems of inaccurate maps
50 m
50 m 50 m
54 52 55 56
50 m
Urban
53 53 54 56
Forest
52 54 55 57
53 55 56 58 Water
65 m ?? m
60 m
56 m
55 m
55 m
50 m
Pixel 1 Pixel 2 Pixel 3

Learnings
TOPIC LEARNINGS
Slope, decay index In urban areas slope is between 30 – 40 dB/dec, in
rural areas closer to 30 dB/dec.
Okumura-Hata Commonly used propagation model.
Can be used for large cells, be careful in urban
areas.
Juul-Nyholm Very similar to Okumura-Hata
Effective antenna height The definition of antenna height depends on the
topographics of the terrain.
Effective antenna height presents antenna height
relative to the surroundings.
Model tuning The target is to find parameters for the basic
propagation model and terrain type correction
factors in a such way that the predicted field
streght differs as little as possible from the
measured field streght.
Model tuning measurements In model tuning measurements it is important that
all the data is correct.
Digital maps Digital maps can have several layer depending the
information presented. Note that accuracy is not
nessesary the same as resolution.

Learnings
Walfish-Ikegami Not a real microcell model. Good for small

macrocell.
Microcell models Currently the microcell models require a lot of
computation time if the predictions are targeted to
be accurate.
Antenna selection The target area affects the selection of antenna.
Both horizontal and vertical 3dB-beamwidths have
an effect on the selection.
Calculation area Calculation areas for cells should be selected very
carefully to achieve optimum results. Too large
areas increase the computation time but too small
areas give unaccurate results.

TLT-6407 Section 3

Uploaded by

Copyright:

Available Formats

TLT-6407 Section 3

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

TLT-6407 Section 3

Uploaded by

Copyright:

Available Formats

TAMPERE UNIVERSITY OF TECHNOLOGY

Institute of Communications Engineering

TLT-6406 RADIO NETWORK PLANNING

TLT-6406 Radio Network Planning 2

TLT-6406 Radio Network Planning 3

TLT-6406 Radio Network Planning 4

TLT-6406 Radio Network Planning 5

TLT-6406 Radio Network Planning 6

TLT-6406 Radio Network Planning 7

TLT-6406 Radio Network Planning 8

Location probability Ph = 50%

Location probability Ph = 74%

TLT-6406 Radio Network Planning 9

TLT-6406 Radio Network Planning 10

TLT-6406 Radio Network Planning 11

If making the following substitutions

TLT-6406 Radio Network Planning 12

Location probability over

TLT-6406 Radio Network Planning 13

TLT-6406 Radio Network Planning 14

TLT-6406 Radio Network Planning 15

TLT-6406 Radio Network Planning 16

TLT-6406 Radio Network Planning 17

TLT-6406 Radio Network Planning 18

BASE STATION UNIT VALUE

Peak EIRP dBm 59 F = A-B+C-D+E

Inteference degradation margin dB 1 J

Minimum reception level dBm -104 K = -G+H+I+J

Isotropic path loss dB 163 L = F-K

TLT-6406 Radio Network Planning 19

TLT-6406 Radio Network Planning 20

TLT-6406 Radio Network Planning 21

Indoor Outdoor (BPL = 0 dB)

TLT-6406 Radio Network Planning 22

2. There can be different definitions!

2. Also call log-normal fading

2. Standard deviation affect the margin

TLT-6406 Radio Network Planning 23

LFree Space ~ 20 log 10 d 

TLT-6406 Radio Network Planning 26

TLT-6406 Radio Network Planning 27

TLT-6406 Radio Network Planning 28

TLT-6406 Radio Network Planning 29

TLT-6406 Radio Network Planning 30

L  A  B log10 f  13.82 log10 hb  a( hm )  ( 44.9  6.55 log10 hb ) log10 d  Cm

150 – 1000 MHz 1500 – 2000 MHz

a(hm )  (1.1log10 f  0.7)hm  (1.56 log10 f  0.8) for small/medi um city

TLT-6406 Radio Network Planning 31

TLT-6406 Radio Network Planning 32

TLT-6406 Radio Network Planning 33

TLT-6406 Radio Network Planning 34

F  10 log10 P  k 0  k1 log10 hb  (k 2  k3 log10 hb ) log10 d

F Field strenght [dBµV/m]

d Distance between mobile and base station [km]

 Frequency range 800-1000 MHz and 1700-2000 MHz,

TLT-6406 Radio Network Planning 35

TLT-6406 Radio Network Planning 36

relevant paths from transmitter to receiver RX

TLT-6406 Radio Network Planning 37