Propagation Models & Scenarios:

Urban


























2012 by AWE Communications GmbH

www.awe-com.com

Contents
2012 by AWE Communications GmbH 2








Overview: Propagation Scenarios
- Rural and Suburban: Pixel Databases (Topography and Clutter)
- Urban: Vector databases (Buildings) and pixel databases (Topography)
- Indoor: Vector databases (Walls, Buildings)


Wave Propagation Model Principles
- Multipath propagation
- Reflection
- Diffraction
- Scattering
- Antenna pattern


Topography and Vector Data (buildings and/or vegetation)
- Map data
- Propagation models
- Evaluation with measurements
2012 by AWE Communications GmbH 3
Propagation Scenarios










Propagation Scenarios (1/2)



Different types of cells in a cellular network
Macrocells

Cell radius > 2 km

Coverage


Microcells
Cell radius < 2 km

Capacity (hot spots)


Picocells
Cell radius < 500 m

Capacity (hot spots)
2012 by AWE Communications GmbH 4
Propagation Scenarios






Propagation Scenarios (2/2)







Macrocell


Microcell


Picocell


Database type


Raster data

Vector data
Raster data


Vector data


Database

Topography
Clutter

2.5D building (vector)
Topography (pixel)

3D building
3D indoor objects


Path Loss
Prediction Models
Hata-Okumura
Two Ray
Knife Edge Diffraction
Dominant Path
Knife Edge Diffraction
COST 231 WI
Ray Tracing
Dominant Path
Motley Keenan
COST 231 MW
Ray Tracing
Dominant Path


Radius

r < 30 km
r > 2 km

r < 2000 m
r > 200 m


r < 200 m
2012 by AWE Communications GmbH 5
Wave Propagation Models






Propagation Models

Different types of environments require different propagation models
Different databases for each propagation model
Projects based on clutter/topographical data or vector/topographical data
Empirical and deterministic propagation models available
CNP used to combine different propagation environments



Types of databases
Pixel databases (raster data)

Topography, DEM (Digital Elevation Model)

Clutter (land usage)
Vector databases
Urban Building databases (2.5D databases polygonal cylinders)

Urban 3D databases (arbitrary roofs)

Indoor 3D databases
2012 by AWE Communications GmbH 6








Topography and Vector Data



Databases: Vector Building Databases




















3D vector oriented database

Buildings as vertical cylinders
with polygonal ground-planes

Uniform height above street-level

Limitation to vertical walls and flat roofs

Individual material properties of building surfaces

Topography can be considered optionally











Example: New York
2012 by AWE Communications GmbH 7










Topography and Vector Data



Consideration of Topography for Vector Scenarios

Topographical databases:
Topography in pixel databases

Resolutions of 20-30 m




Consideration in Prediction:

Shift transmitter and receiver

Shift buildings due to the topo

Approximation of topo with triangles




Effects on results:
Additional shadowing by hills

Changing LOS-area of the transmitter

No additional rays (scattering at topo)
2012 by AWE Communications GmbH 8












Topography and Vector Data



Databases: Vector Building Databases

Special features














Courtyards and Towers
















Multiple Courtyards and Towers
Vegetation areas

Vegetation areas are polygonal cylinders.
Rays get an additional attenuation (dB/m)
when passing the cylinder and receiver
pixels inside cylinder get an additional loss
2012 by AWE Communications GmbH 9








Topography and Vector Data



Databases: Material Properties

Global catalogue for different construction materials (at various frequency bands)



(In WallMan via menu Edit Materials Import)

User can add or modify materials
2012 by AWE Communications GmbH 10






Topography and Vector Data



Databases: Material Properties

Local material database (in building database)

only relevant for objects in this database

independent of global material catalogue
(modification of global catalogue does not affect material properties of objects in database)

can be updated with materials from global material catalogue



Settings of local material database

individual material properties for different frequency bands
(always the properties of the frequency band closest to TX frequency is used)

Material (incl. all properties) is assigned to objects (walls/buildings)

Always all material properties must be defined even if they are not
required for the selected propagation model

Individual colors can be assigned to the materials for better visualization
2012 by AWE Communications GmbH 11








Topography and Vector Data



Databases: Material Properties

Properties of a material
Properties affecting all propagation models

Transmission Loss (in dB)

Properties affecting Ray Tracing & Dominant Path Model

Reflection Loss (in dB)

Properties affecting Ray Tracing

GTD/UTD related properties
Relative Dielectricity

Relative Permeability

Conductance (in S/m)

Empirical reflection/diffraction model
Reflection Loss (in dB)

Diffraction Loss Incident Min (in dB)

Diffraction Loss Incident Max (in dB)

Diffraction Loss Diffracted (in dB)
2012 by AWE Communications GmbH 12












Topography and Vector Data



Propagation Models
COST 231 Walfisch-Ikegami

Homogenous parameters (street width, building
height,) for whole area

Individual determination of parameters according
to buildings in vertical plane between Tx and Rx



Ray Tracing

3D Ray Tracing IRT (with preprocessing)

2x2D Ray Tracing IRT (horiz. and vertical plane)

3D Ray Tracing SRT (standard, no preprocessing)



Dominant Path Model

3D path searching
2012 by AWE Communications GmbH 13










Topography and Vector Data



Propagation Models: COST 231 Walfisch-Ikegami


Model accepted by ITU-R

Evaluating building profile between transmitter and
receiver (vertical plane)

Consideration of additional losses due to building data

Reasonable results for Tx above rooftops

For Tx below rooftops limited accuracy (no wave guiding)

No multipath propagation considered
















Transmitter Considered propagation path Receiver
Buildings considered for determination of parameters
2012 by AWE Communications GmbH 14








t
h
Topography and Vector Data



Propagation Models: COST 231 Walfisch-Ikegami
WinProp: Vertical plane is analyzed for each predicted pixel individually!

Parameters of the model obtained from the buildings in the vertical plane






h
Roof
w
h
r







Height of transmitter h
TX


Height of receiver h
RX

b
d
Mean value of building heights h
roof


Mean value of widths of roads w

Mean value of building separation b






Vertical profile with topography
2012 by AWE Communications GmbH 15






Topography and Vector Data



Propagation Models: COST 231 Walfisch-Ikegami

Parameters of the model gained from the buildings in the vertical plane


LOS: l
b


= 42,6 dB
d

+ 26 lg

km

+ 20 lg
f
MHz
l
0
+ l
rts

+ l
msd

l
rts

+ l
msd
> 0

NLOS: l
b
=


l l + l < 0

0 rts msd

f r

Free space loss l
0
: l
0

= 32,44 dB + 20 lg

w

+ 20 lg

MHz km
f




h
Roof
h
r

Rooftop loss l
rts
: l
rts

= 16,9 dB 10 lg + 10 lg
m
d

+ 20 lg

MHz m
f b

Over rooftop loss l
msd
: l
msd

= l
bsh

+ k
a

+ k
d

lg + k

km
f

lg 9 lg

MHz m
|

18 lg1
h
t
h
Roof
|

| >
with l
bsh
=

\
+
m
.

0
h
t
h
Roof


h
t
< h
Roof


Factors k
a
and k
d

Empir. Correction of antenna heights
Faktor k
f

Adaption to different building
densities

Valid for: f MHz ................... 800 - 2000
h
t
m ................................. 4 - 50
h
r
m ................................. 1 - 3
d m ........................... 20 - 5000
2012 by AWE Communications GmbH 16










Topography and Vector Data



Propagation Models: Ray Tracing
Multipath propagation

Dominant effects:
diffraction and reflection

Up to 6 reflections and 2 diffractions are
determined as well as combinations

Computation of the path loss with
Fresnel coefficients (for reflection) and
GTD/UTD model (for diffraction).
Alternative: Scalable empirical
reflection/diffraction model for
prediction of path loss along the ray

Uncorrelated superposition of
contributions (rays)

Either full 3D or 2x2D (horizontal and
vertical plane)

Post-processing with Knife Edge
Diffraction model possible
2012 by AWE Communications GmbH 17








Topography and Vector Data



Propagation Models: Ray Tracing


Types of rays to be determined

Different types of rays: direct,
reflected, diffracted, scattered

Definition of max. number for
each interaction type

Definition of total interaction
number

Selection of Fresnel & GTD/UTD
or empirical interaction model

Additional thresholds for
computation of paths

2012 by AWE Communications GmbH 18














Topography and Vector Data



Propagation Models: Ray Tracing




Direct Single

Reflection














Double

Reflection
Single

Diffraction
2012 by AWE Communications GmbH 19














Topography and Vector Data



Propagation Models: Ray Tracing



Triple

Reflection













Double

Diffraction
Single

Reflection +
Single
Diffraction







Double
Reflection +
Single
Diffraction
2012 by AWE Communications GmbH 20








Topography and Vector Data



Propagation Models: Intelligent Ray Tracing (IRT)




Considerations to accelerate the time consuming process of path finding:

Deterministic modelling generates
a large number of rays, but only few
of them deliver most of the power

Visibility relations between walls and
edges are independent of transmitter
location

Adjacent receiver pixels are reached
by rays with only slightly different paths

Single pre-processing of the building database with determination of the
visibility relations between buildings reduces computation time
2012 by AWE Communications GmbH 21








Topography and Vector Data



Propagation Models: Intelligent Ray Tracing (IRT)

Pre-processing of the Building Database

Subdivision of the walls into tiles

Subdivision of the vertical and
horizontal edges into segments

Subdivision of the prediction area
into receiving points (grid)



stored information for each visibility relation:

angle between the elements

distance between centres

example: visibility between a tile and a
receiver pixel



























Tile
















o
max















o
min






|

min



|

max














Prediction Pixel
projection of connecting straight lines
into xy-plane and perpendicular plane

4 angles for each visibility relation
Segment Center of Tile
Center of horiz. Segm.
Center of vert. Segm.
2012 by AWE Communications GmbH 22








Topography and Vector Data



Propagation Models: Intelligent Ray Tracing (IRT)

Prediction with Pre-processed Data
Determination of all tiles, segments and receiving points, which are visible
from the transmitter

Computation of the angles
of incidence belonging to
these visibility relations

Recursively processing of
all visible elements incl.
consideration of the
angular conditions

Tree structure is very
fast and efficient


Direct ray


1.interaction





2.interaction






3.interaction
PREDICTION





PREPRO-
CESSING




transmitter receiving point tile / segment
2012 by AWE Communications GmbH 23






Topography and Vector Data



Problem of Database Accuracy in Ray Tracing models







Ray Tracing















T


















T



Building error
2012 by AWE Communications GmbH 24










Topography and Vector Data



Propagation Models: Urban Dominant Path (UDP)


Dominant Path (single path)

Determination of path with full 3D
approach

Unlimited number of interactions
(changes of orientation)

Parameters of path determined (e.g
length, number of interactions,
angles,.) and used to compute path loss
with semi-deterministic equations

Optional consideration of wave guiding
possible (wave guiding factor, based on
reflection loss of walls)

Short prediction time

High accuracy
Typical Channel Impulse Response




One path
dominates















Full 3D approach
2012 by AWE Communications GmbH 25








6 1
Topography and Vector Data



Propagation Models: Dominant Path Model
Determination of Paths

Analysis of types of wedges in scenario

Generation of tree with convex wedges

Searching best path

Computation of path loss

T

Layer 1

2 4 5


5 T 2
3
4
R
Layer 2

Layer 3

Layer 4
4 5


R 5 4


R
2 R 5


5 2
2 4


4 R 2


R


concave wedges convex wedges
1 3 6 2 4 5
2012 by AWE Communications GmbH 26






Topography and Vector Data



Propagation Models: Dominant Path Model
Computation of Path Loss




Path length l

Path loss exponents before and after breakpoint p

individual interaction losses f(,i) for each interaction i of all n
interactions

Gain due to waveguiding

Gain g
t
of base station antenna





4p

L 20 log


10 p log (l )
n
f ( , i) g
l

t
2012 by AWE Communications GmbH 27


=


+ + j + W +
i=0
2012 by AWE Communications GmbH 27






Topography and Vector Data



Propagation Models: Dominant Path Model
Parameters for prediction (1/2)

2012 by AWE Communications GmbH 28








Topography and Vector Data



Propagation Models: Dominant Path Model
Parameters for prediction (2/2)



Acceleration for large areas
Adaptive Resolution Management
Path loss exponents before and after
breakpoint can be defined individually
Breakpoint distance/computation can be
adapted to the users needs
Definition of different path loss exponents
for LOS (Line of Sight) and OLOS
(Obstructed Line of Sight)
Interaction losses (at points where the
path changes its orientation) can be
defined
Individual reflection loss assigned to
buildings influences wave guiding effect









TX












Wave guiding factor
2012 by AWE Communications GmbH 29








Topography and Vector Data



Propagation Models: Preprocessing with WallMan
Single pre-processing of building database required only for IRT model





Project File
Pre-processing
(*.pre)

Pre-processing

(Computation)
Pre-processed
Database Files
(oib, ocb opb)








Original Binary
Database file
(*.odb)



Materials (electrical properties) can
still be modified after pre-processing.

Re-assignment of materials to objects
is not possible after pre-processing.
Database Extensions:

*.odb Outdoor Data Binary




*.ocb Outdoor COST Binary

*.oib Outdoor IRT Binary

*.opb Outdoor Dom. Path Binary
2012 by AWE Communications GmbH 30












Topography and Vector Data



Propagation Models: Comparison


COST 231 Walfisch-Ikegami Ray Tracing (3D IRT) Dominant Path (3D)


Computation time: < 1 min Computation time: 3 min Computation time: < 1 min

Preprocessing time: < 1 min Preprocessing time: 30 min Preprocessing time: < 1 min


Not very accurate High accuracy in region of Tx
Limited accuracy far away
High accuracy everywhere
2012 by AWE Communications GmbH 31









Topography and Vector Data



Propagation Models: Indoor Penetration






Constant Level Model Exponential Decrease Model Variable Decrease Model

Considers defined
transmission loss


Homogeneous indoor level


Subtracting defined
transmission loss from
average level at outer walls
Considers defined
transmission loss


Additional exponential
decrease towards the
interior with attenuation rate
depending on building
depth (~ 0.1 dB/m)
Considers defined
transmission loss


Additional exponential
decrease towards the interior
with definable attenuation
rate (default 0.6 dB/m)
2012 by AWE Communications GmbH 32






Topography and Vector Data



Propagation Models: Prediction of LOS States
LOS: Line of sight between Tx and Rx

OLOS: Obstructed line of sight between Tx and Rx (only indoor)

NLOS: No line of sight between Tx and Rx

LOS-V: Line of sight regarding the buildings, but shadowing due to vegetation

NLOS-V: NLOS due to buildings and additional shadowing by vegetation


2012 by AWE Communications GmbH 33










Topography and Vector Data



Sample Large Urban Scenario incl. Topography





































Prediction of Hong Kong (334 km, 1.5 megapixel, 22030 buildings, comp. time: 15 min)
(transmit power: 40 dBm, GSM 900, directional antenna at 40 m height)
2012 by AWE Communications GmbH 35












Topography and Vector Data



Sample Urban Scenario

































2D view




Prediction of Manhattan (9 km x 18 km, 15758 buildings, comp. time: 6 min)
2012 by AWE Communications GmbH 36










Urban Evaluation





Evaluation with Measurement Data




Wave Propagation Models considering

Topography and Clutter Data

Topography and Vector Data



2012 by AWE Communications GmbH 37
Urban Evaluation









Evaluation with Measurements



Investigated Scenarios:

I. Helsinki, Finland

II. Hong Kong, China

III. Monaco, Monte Carlo

IV. Munich, Germany

V. Ilmenau, Germany

VI. Amsterdam, Netherlands
2012 by AWE Communications GmbH 38
Urban Evaluation





Scenario Information
Number of buildings 1651
Topo. difference none (flat terrain)
Resolution 5 m


Transmitter
Site 1 4.0 m, 2.5 Watt, 900 MHz
Site 2 41.5 m, 10 Watt, 2.1 GHz

Prediction heights 1.6 m, 2.5 m




Scenario I: Helsinki, Finland



3D view of database
2012 by AWE Communications GmbH 39
Urban Evaluation













Scenario I: Helsinki, Finland




Predictions for transmitter location 2






Prediction with COST 231
Walfisch-Ikegami
Prediction with 3D Ray
Tracing
Prediction with Urban
Dominant Path
2012 by AWE Communications GmbH 40
Urban Evaluation














Scenario I: Helsinki, Finland




Differences for transmitter location 2




























Difference of prediction
with COST 231 Walfisch-
Ikegami and
measurements
Difference of prediction
with 3D Ray Tracing and
measurements
Difference of prediction
with Urban Dominant
Path and measurements
2012 by AWE Communications GmbH 41
Urban Evaluation







Scenario I: Helsinki, Finland




Statistical evaluations for all transmitters









Site

Statistical Results
Empirical Model
(e.g. COST 231 Walfisch-
Ikegami)

Deterministic Model
(e.g. 3D Ray Tracing or Urban Dominant Path)

Mean
Value
[dB]

Std.
Dev.
[dB]

Comp.
Time
[s]


Mean Value
[dB]


Std. Dev.
[dB]


Comp. Time
[s]

2

-9.38

9.40

2

-1.041.94

5.926.30

2032

3

-5.84

8.35

2

-3.604.31

5.537.81

18.. 32

Avg

-7.61

8.88

2

-0.83...1.64

5.73...7.06

19.. 32

A standard PC with an AMD Athlon64 2800+ processor and 1024 MB of RAM
was used to determine the computation times
2012 by AWE Communications GmbH 42
Urban Evaluation









Scenario Information
Number of buildings 3306
Topo. difference 482 m
Resolution 10 m


Transmitter
Site 1 33.0 m, 28.5 dBm, 948 MHz
Site 2 94.0 m, 24.9 dBm, 948 MHz

Prediction height 1.5 m




Scenario II: Hong Kong, China




































3D view of database with topography
2012 by AWE Communications GmbH 43
Urban Evaluation










Scenario II: Hong Kong, China


Predictions for transmitter location 1




Prediction with Urban Dominant Path






Prediction with COST 231
Walfisch-Ikegami


Prediction with 3D Ray
Tracing
2012 by AWE Communications GmbH 44
Urban Evaluation
















Scenario II: Hong Kong, China


Differences for transmitter location 1










Difference of prediction with COST
231 Walfisch-Ikegami and
measurements










Difference of prediction with Urban
Dominant Path and measurements



Difference of prediction with 3D
Ray Tracing and measurements
2012 by AWE Communications GmbH 45
Urban Evaluation







Scenario II: Hong Kong, China


Statistical evaluations for all transmitters










Site

Statistical Results

Empirical Model
(e.g. COST 231 Walfisch-
Ikegami)

Deterministic Model
(e.g. 3D Ray Tracing or Urban Dominant Path)

Mean
Value
[dB]

Std.
Dev.
[dB]

Comp.
Time
[s]


Mean Value
[dB]


Std. Dev.
[dB]

Comp.
Time
[s]

1

-12.81

20.13

5

0.724.91

6.08 7.56

10127

2

1.34

9.02

5

-2.305.63

7.74 7.79

1680

Avg

-5.74

14.58

5

-0.79...5.27

6.94 ...7.65

13...104

A standard PC with an AMD Athlon64 2800+ processor and 1024 MB of RAM
was used to determine the computation times
2012 by AWE Communications GmbH 46
Urban Evaluation









Scenario Information
Number of buildings 1511
Topo. difference 646 m
Resolution 10 m
Transmitter 17.0 m, 31.0 dBm, 2.2 GHz
Prediction height 1.5 m




Scenario III: Monaco, Monte Carlo

































3D view of database
2012 by AWE Communications GmbH 47
Urban Evaluation














Scenario III: Monaco, Monte Carlo


Predictions for transmitter location 1





Prediction with COST 231
Walfisch-Ikegami
Prediction with 3D Ray
Tracing
Prediction with Urban
Dominant Path
2012 by AWE Communications GmbH 48
Urban Evaluation














Scenario III: Monaco, Monte Carlo


Differences for measurement route 50



























Difference of prediction
with COST 231 Walfisch-
Ikegami and
measurements
Difference of prediction
with 3D Ray Tracing and
measurements
Difference of prediction
with Urban Dominant
Path and measurements
2012 by AWE Communications GmbH 49
Urban Evaluation







Scenario III: Monaco, Monte Carlo



Statistical evaluations for all measurements routes









Route

Statistical Results
Empirical Model
(e.g. COST 231 Walfisch-Ikegami)
Deterministic Model
(e.g. 3D Ray Tracing or Urban Dominant Path)

Mean Value
[dB]

Std. Dev.
[dB]

Comp. Time
[s]

Mean Value
[dB]

Std. Dev.
[dB]

Comp. Time
[s]

50

-18.71

5.74




3

-4.73-2.94

3.924.36




15141
52

-20.12

8.09

-1.940.08

4.976.17

58

-25.28

9.04
-0.60-0.23

4.094.87

Avg

-21.37

7.62

3

-2.30...-1.15

4.73

15...141


A standard PC with an AMD Athlon64 2800+ processor and 1024 MB of RAM
was used to determine the computation times
2012 by AWE Communications GmbH 50
Urban Evaluation







Scenario Information
Number of buildings 2032
Topo. difference 14 m
Resolution 10 m
Transmitter 13.0 m, 10.0 Watt, 947 MHz
Prediction height 1.5 m




Scenario IV: Munich, Germany

3D view of database with topography
2012 by AWE Communications GmbH 51
Urban Evaluation














Scenario IV: Munich, Germany


Predictions for transmitter location 1




Prediction with COST 231
Walfisch-Ikegami
Prediction with 3D Ray
Tracing
Prediction with Urban
Dominant Path
2012 by AWE Communications GmbH 52
Urban Evaluation














Scenario IV: Munich, Germany


Differences for measurement route 0



Difference of prediction
with COST 231 Walfisch-
Ikegami and
measurements
Difference of prediction
with 3D Ray Tracing and
measurements
Difference of prediction
with Urban Dominant
Path and measurements
2012 by AWE Communications GmbH 53
Urban Evaluation







Scenario IV: Munich, Germany


Statistical evaluation for all measurement routes










Route

Statistical Results

Empirical Model
(e.g. COST 231 Walfisch-Ikegami)
Deterministic Model
(e.g. 3D Ray Tracing or Urban Dominant
Path)

Mean Value
[dB]


Std. Dev. [dB]

Comp. Time
[s]

Mean Value
[dB]


Std. Dev. [dB]

Comp. Time
[s]
0 -10.98 6.38




5
-5.262.80 7.137.17




14...20

1

-13.80

7.07

-2.011.34

6.206.73

2

-14.70

7.43

-3.150.31

7.948.04

Avg

-13.16

6.96

5

-3.47...1.48

7.09...7.31

14...20

A standard PC with an AMD Athlon64 2800+ processor and 1024 MB of RAM
was used to determine the computation times
2012 by AWE Communications GmbH 54
Urban Evaluation












Scenario V: Ilmenau, Germany
Trajectory in Urban Marco Cell (COST reference scenario)
Tx height: 26.5 m

Tx frequency: 2.53 GHz
Tx power: 46 dBm
Receiver: high resolution 3D channel sounder (RUSK, Medav GmbH)
Receiver moving with constant speed along trajectory (~ 54/123 m)
Rx height: 1.9 m

2012 by AWE Communications GmbH 55
Urban Evaluation

















[dBm] Mean Std. Dev.
Measured -62.38 2.24
Simulated -62.47 2.06
Difference 0.09 0.70

[ns] Mean Std. Dev.
Measured 195.33 17.11
Simulated 208.79 37.46
Difference 13.46 33.32

[bit/s/Hz] Mean Std. Dev.
Measured 6.31 0.13
Simulated 6.48 0.21
Difference 0.17 0.20




Rx Power:
(Route 41a-42)












Delay Spread:
(Route 41a-42)












MIMO Capacity (2x2):
(Route 41a-42)
2012 by AWE Communications GmbH 56
Urban Evaluation

















[dBm] Mean Std. Dev.
Measured -50.83 6.18
Simulated -50.85 5.33
Difference 0.02 1.65

[ns] Mean Std. Dev.
Measured 173.36 75.54
Simulated 172.43 70.61
Difference 0.92 27.21

[bit/s/Hz] Mean Std. Dev.
Measured 6.14 0.19
Simulated 6.26 0.26
Difference 0.12 0.24




Rx Power:
(Route 10b-9b)












Delay Spread:
(Route 10b-9b)












MIMO Capacity (2x2):
(Route 10b-9b)
Urban Evaluation














Scenario VI: Amsterdam, Netherlands
Trajectory in Urban Marco Cell
Tx height: 29 m
Tx frequency: 2.25 GHz
Tx power: 43 dBm
Receiver: high resolution 3D-Channel Sounder (TU Eindhoven)
Receiver moving with constant speed along trajectory (~ 420 m)
Rx height: 3.5 m




2012
Bridge / Tunnel
(not considered in simulation)
by AWE Communications GmbH 57

Urban Evaluation
2012 by AWE Communications GmbH 58


















[] Mean Std. Dev.
Measured 52.05 21.15
Simulated 49.79 32.96
Difference -2.25 24.99




Rx Power:

[dBm] Mean Std. Dev.
Measured -53.91 8.04
Simulated -53.90 7.10
Difference 0.01 4.03



Delay Spread:
[ns] Mean Std. Dev.
Measured 222.36 106.91
Simulated 216.07 130.23
Difference -6.29 109.63



Angular Spread (Rx):








Bridge / Tunnel
(not considered in simulation)
2012 by AWE Communications GmbH 59
Summary






Features of WinProp Urban Module

Highly accurate propagation models
Empirical: COST 231 Walfisch-Ikegami

Deterministic (ray optical): 3D Dominant Path, 3D Ray Tracing, 2x2D Ray Tracing

Optionally calibration of 3D Dominant Path Model with measurements possible but not
required as the model is pre-calibrated

Building data
Models are based on 2.5D vector data of buildings

Consideration of material properties (also vegetation objects can be defined)
Consideration of topography (pixel databases)
Antenna patterns
Either 2x2D patterns or 3D patterns

Outputs
Signal level (path loss, power, field strength)
Delays (delay window, delay spread,)
Channel impulse response
Angular profile (direction of arrival)
2012 by AWE Communications GmbH 60






Further Information





Further information: www.awe-com.com