LATEST TRENDS on COMMUNICATIONS

An Open-Source Radio Coverage Prediction Tool

ANDREJ HROVAT*, IGOR OZIMEK*, ANDREJ VILHAR*,
TINE CELCER*, IZTOK SAJE+ and TOMAŽ JAVORNIK*
*
Department of Communication Systems,
*
Jozef Stefan Institute, +Mobitel, d.d.
*
Jamova cesta 39, SI-1000 Ljubljana, +Vilharjeva 23, SI-1000 Ljubljana
*+
SLOVENIA
andrej.hrovat@ijs.si http://www-e6.ijs.si

Abstract: - The cellular concept applied in mobile communication systems enables significant increase of
overall system capacity, but requires careful radio network planning and dimensioning. Wireless and mobile
network operators typically rely on various commercial radio network planning and dimensioning tools, which
incorporate different radio signal propagation models. In this paper we present the use of open-source
Geographical Resources Analysis Support System (GRASS) for the calculation of radio signal coverage. We
developed GRASS modules for radio coverage prediction for a number of different radio channel models, with
antenna radiation patterns given in the standard MSI format. The results are stored in a data base (e.g. MySQL,
PostgreSQL) for further processing and in a simplified form as a bit-map file for displaying in GRASS. The
accuracy of prediction was confirmed by comparison with results obtained by a dedicated professional
prediction tool as well as with measurement results.

Key-Words: network planning tool, open-source, GRASS GIS, path loss, raster, clutter, radio signal coverage

1 Introduction for radio network optimization and are intended for

Emerging user applications call for increased specific tasks such as WLAN network planning,
bandwidth of communication systems. calculation of radio coverage inside buildings, design
Consequently, higher frequencies are used in of radio-relay links, etc.
wireless systems while the size of radio cells is Those mentioned tools do not allow users to add
becoming smaller. The cellular concept enables new propagation prediction modules or to adjust the
lower transmission power and frequency reuse in existing ones. From the scientific point of view their
cells which are far enough from each other. usage is therefore very limited. These limitations can
However, due to the increased complexity, a wireless be avoided by using an open-source platform which
system has to be planned carefully. Cellular system can be upgraded by an arbitrary propagation model.
planning involves determining the number and the As the terrain relief significantly influences radio
locations of base stations, their hardware and wave propagation, a logical choice is to use an open-
software, frequency and code planning. One of the source geographical information system (GIS).
aims is to efficiently use the allocated frequency These systems also include built-in functions for
band and to assure high radio coverage. displaying results on geographical maps, importing
For the calculation of radio coverage, various different raster and vector GIS formats, converting
mathematical radio propagation models are being geographical coordinates, etc. Geographical
used [1, 2, 3, 4]. They can be divided into three Resources Analysis Support System (GRASS) is one
groups: (i) statistical models, (ii) deterministic (or of the most wide-spread open source GIS systems,
theoretical) models and (iii) combinatorial models. which has been successfully used for many years and
Various commercial programming tools are has a wide spectrum of already implemented
available for radio coverage calculation. The first modules [11].
representative tools were designed for mobile In the paper, the following section presents the
operators and national regulators, e.g. Planet [5], GRASS system with its main structure,
decibel Planner [5], Vulcano [6] and CS telecom nG characteristics and its applicability in the field of
[7]. Accordingly, their price was high while their radio communications. Next, a description of the
accessibility and spread of usage were low. Later on, radio coverage prediction software developed in
some cheaper yet functionally limited tools have GRASS is presented. The essential building blocks
appeared on the market, e.g. WinProp [8], RPS [9] calculating path loss, sectorisation, radio coverage,
and TAP [10]. These tools do not comprise modules and converting and evaluating input/output data are

ISSN: 1792-4243 135 ISBN: 978-960-474-200-4

 LATEST TRENDS on COMMUNICATIONS

explained. In addition, a module for tying various where each mapset represents either a subregion or
processing modules into a complete radio coverage data of a specific user. Users may read and copy data
tool is also described. In section 4, the GRASS from any mapset, while modifications are allowed
software package is evaluated by comparing only within their own mapsets. Such organization
simulation results with field measurements and enables efficient collaboration between users in a
simulations performed with a professional tool. The working group. The described structure of maps and
paper concludes by a description of our experience files is maintained automatically by GRASS.
with the GRASS system and by plans for our future GRASS GEOMETRY AND
LOCATION MAPSET
work. DATABASE ATTRIBUTE DATA

RASTER MAPS
/cats

/cell

2 GRASS Open Source GIS Tool /europe /PERMANENT

/cellhd

...
GIS systems find their applications in several /home/user/grassdata /slovenia /PERMANENT

different fields, including space planning, business /user1

/vector /border

VECTOR MAPS
/ljubljana
management, navigation, environmental protection, /avilhar
/roads
/toulouse .
demographical data management etc. Several .
...
. .
professional GIS tools exist on the market; however, . /dbf ...
.
due to their limitations such as price, long response .
.
to required changes and limited possibility of tool .

modifications, the open-source approach to the

programming part of GIS technology has also been Fig. 1: GRASS data organization
developed. Similarly to other technologies, GIS
technology also benefits from the open-source Modules for data processing are classified
approach. Some of the advantages are: continuous according to their functionality. They are invoked by
improvement and control carried out by developers commands of a form x.name, where x stands for a
from all over the world, heterogeneous approach to class and name stands for a specific task within this
development, accessibility and adjustability. class. Some class examples are:
GRASS [11] is one of the eight starting projects  g. (general commands),
of the OSgeo foundation [12], established in 2006,  r. (raster data processing),
which supports and promotes open source geospatial  v. (vector data processing),
software. Nowadays, GRASS is known as one of the  d. (commands for graphical display).
most important open source GIS tools. It operates
over raster and vector data and includes methods for 2.2 Usage in the Field of Radio
image processing and display. It is published under Communications
the GPL license, its usage is supported under various In its original form, GRASS can be used to analyse a
operating systems including Mac OS X, Microsoft radio coverage that has been either measured or
Windows and Linux. calculated beforehand by using an arbitrary tool. Of
GRASS comprises over 350 already implemented course, the data has to be imported into the GRASS
modules for processing, analysis and visualization of environment in a proper form. Alternatively, radio
geographical data. The core modules and libraries coverage models can be implemented inside GRASS
are written in the C programming language. A well due to its open-source nature. By doing so, potential
documented API (Application Programming inconveniences and/or errors that may arise from
Interface) with a few hundred C functions is data format conversion are avoided. The whole
available for the developers of new modules. The process can remain modular, as different modules are
documentation is up-to-date and is available online used for the data analysis and the propagation
[13]. For large projects, processing may be prediction.
automated by using scripting languages.

2.1 Structure of Data and Commands 3 Radio Coverage Project in GRASS

The organization of geographical data in GRASS is We have developed a modular radio coverage tool
depicted in Fig. 1. The data is divided into different characterized by a high level of flexibility and
locations, where each location is defined by its own adaptability. It performs separate calculation of the
coordinate system, map projection and geographical radio signal path loss using an arbitrary channel
boundaries. Each location can have many mapsets, model, and the inclusion of antenna radiation

ISSN: 1792-4243 136 ISBN: 978-960-474-200-4

 LATEST TRENDS on COMMUNICATIONS

patterns and setup parameters such as antenna tilting, responsible for a correct sequence of modules
azimuth and antenna height. Furthermore, it execution. Therefore, each individual module
performs computation of the maximum signal level represents realization of the radio calculations only,
at the receiver and additional processing of the while the script takes care for the input and output
output data. data management. The achieved modularity has
The software packet block diagram is presented in several benefits:
Fig. 2. It is composed of two basic compositions of  simple upgrade or substitution of existing
modules. The first part is composed of GRASS mathematical modules with new models,
modules for radio coverage calculations, which are  module independency from a specific
linked together with a script written in the Python network,
programming language. Additional modules for data  quick and simple recalculation for an
comparison and for adapting input data to the individual segment or chosen geographical
GRASS data structure build the second group of region,
GRASS modules.  possibility of parallelized calculation.
Besides the GRASS modules represented in Fig. In each step, realization of different modules with
2 as white squares, the input and output data are also the same or a similar task is feasible. Proper module
depicted as different colorized parallelograms. selection is performed through the script and
Textual input and output files are indicated in depends on the usage purpose. This is significant
orange, GRASS raster files in blue, while databases especially for the first segment where different
are denoted in yellow. mathematical models for radio signal propagation
The core of the radio coverage software is radio can be taken into consideration. In the first segment,
coverage calculation in Fig. 2 encircled with a four modules are implemented, r.fspl, r.hata,
dashed line. Radio coverage calculation for the r.hataDEM and r.cost231. Module r.sector is
whole cellular network is divided into three steps: currently the only module in the second segment
 path loss calculation for isotropic source, while the third segment contains two modules -
 calculation of influence of the antenna db.GenerateTable (creates an empty table for the
diagram, antenna tilt and azimuth, results) and r.MaxPower (arranges the computed
 storage of N highest calculated received data and writes it to the output file).
signal strength values into a database.
In the GRASS programming environment every 3.1 Implemented Path Loss Models
segment is implemented as a separate module. A Currently, four basic path loss prediction models for
script written in the Python programming language is

Fig. 2: Radio coverage prediction software block diagram

ISSN: 1792-4243 137 ISBN: 978-960-474-200-4

 LATEST TRENDS on COMMUNICATIONS

the open rural and suburban environments are r.sector module calculates the actual path loss for the
implemented. analyzed cell and writes the data to the output raster
The FSPL (Free Space Path Loss) model for further processing.
implemented in the r.fspl module calculates radio
signal propagation attenuation in free space with no 3.3 Arranging Cells According to Received
nearby obstacles to cause reflections or diffractions Power and Writing in Database
[14]. At higher carrier frequencies and in After the path loss for each individual cell located
environments without many reflections, the r.fspl within the analyzed area has been calculated, the
module can serve as the first approximation of the radio signal coverage prediction must be performed.
radio signal propagation prediction for the The received signal strengths for cells included in
geographical points that are in the transmitter’s line- simulations are calculated with the r.MaxPower
of-sight. Therefore, the calculation must be done in module. Values from different antennas at each
two steps. In the first step, the visibility between the receive location are arranged in a table in a
transmitter and each receive point in the area must be decreasing sequence. An empty table is first
determined with the already integrated r.los module. generated with the db.GenerateTable module.
Afterwards, the path loss at the LOS points is Furthermore, the r.MaxPower module also generates
calculated using the r.fspl module. an output raster file with the maximal received signal
The r.hata module implements the Okumura-Hata strengths for all individual points, which can be
model [1]. The model is founded on empirically graphically presented in GRASS GUI (Fig. 3).
determined radio propagation characteristics and
includes three variants: for the urban, suburban and
open environments. The model does not consider
terrain configuration neither the environment where
the mobile terminal is located, which are its main
drawbacks. Radio signal attenuation depends only on
the distance, antennas heights and carrier
frequencies. To improve the model accuracy, an
additional knife edge diffraction module must be
implemented.
The COST231 model, realized in the r.cost231
module, is an extension of the Okumura-Hata model
for higher frequencies [15]. It is suitable for medium
and large cities where the base station antenna height Fig. 3: Radio coverage calculation for flat terrain at
is above the surrounding buildings. The terrain 2040 MHz with r.hataDEM module
configuration is only partly taken into consideration.
Therefore, the signal is predicted also behind larger 3.4 Python Script
geographical obstacles, which significantly In order to enable interaction of all processing
contributes to the model inaccuracy. modules, we wrote a module, r.radcov, in the Python
In the r.hataDEM module, a modification of the scripting language. Creation of the user interface
Okumura-Hata model is implemented [16]. In itself was considerably simplified by the fact that
addition to the carrier frequency, the distance GRASS has built-in support for this, offering input
between the transmitter and the receiver, and the parameter parsing and checking against allowed
receiver and transmitter antenna heights, the model values, and also automatically generating graphic
takes into consideration also the terrain profile, user interface at run time if a user wants to use it.
clutter data and the spherical earth impact. This is the First, r.radcov reads an input table in the CSV
most accurate and sophisticated model implemented (Comma Separated Value) format, which specifies
in GRASS so far. the configuration of the radio cells comprising the
radio network, such as positions and orientations of
3.2 Antenna Radiation Diagram Influence antennas, etc. It also takes a number of parameters
After the path loss calculation of the isotropic source (as command-line arguments or via the GRASS’s
for a specific region, the antenna’s radiation diagram auto-generated GUI) specifying global simulation
is considered. Based on the input raster containing data such as radio transmission frequency etc. The
the path loss data for the isotropic source, and the script performs extensive checking of these
antenna’s radiation diagram (beam direction, parameters as well as the contents of the input table
electrical and mechanical tilt, antenna gain) the against valid values and reports eventual errors.

ISSN: 1792-4243 138 ISBN: 978-960-474-200-4

 LATEST TRENDS on COMMUNICATIONS

Next, r.radcov performs a two-step coverage the professional radio signal coverage prediction
computation by first calling modules for the selected programme TEMS.
propagation models (e.g. r.hata, r.hataDEM,…) and The performance of the new software package
then applying antenna transmission beam forming was investigated for different types of networks
using the actual antennas’ radiation patterns. (GSM, UMTS) and terrains (hilly and almost flat
Finally, r.radcov calls the r.MaxPower module, rural, urban, and suburban). Due to the limited scope
which joins all partial coverage results for individual of the article, only the analysis for one region with
antennas into a complete coverage data, given as a several UMTS base stations is included. Radio signal
bit map picture and a database (e.g. MySQL, coverage prediction using the custom built GRASS
PostgreSQL) for further processing. software takes longer than an identical prediction
with the TEMS software package.
The accuracy of the GRASS prediction software
4 Radio Coverage Tool Performance can be verified from charts in Fig. 4. On the left side,
the charts comparing the measurements and
Analysis calculations with the GRASS radio coverage
The performance and accuracy of the developed
prediction software are depicted, while graphs
modules for radio signal coverage prediction was
showing the comparison between the measurements
investigated through the comparison of simulation
and calculations with the TEMS software package
results with field measurements. The reference
are on the right. It is evident from the received power
values were obtained by comparing field
charts showing the simulation results from both the
measurements and simulation results acquired from
software and the field measurements that the

-20 -20
Meas. Meas.
-40 Sim -40 Sim

-60 -60
Rscp [dBm]

Rscp [dBm]

-80 -80

-100 -100

-120 -120

-140 -140

-160 -160
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3
N 4 N 4
x 10 x 10

Received power

60 60

40 40

20 20
Diff [dBm]

Diff [dBm]

0 0

-20 -20

-40 -40

-60 -60

-80 -80
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3
N 4 N 4
x 10 x 10

Difference between calculated and measured received power

Comparison between measurements and Comparison between measurements and

calculations with GRASS radio coverage calculations with TEMS software package
prediction software

Fig. 4: Comparison simulations and measurement results at 2040MHz

ISSN: 1792-4243 139 ISBN: 978-960-474-200-4

 LATEST TRENDS on COMMUNICATIONS

simulation results match with the measurements data. The achievment made so far represents a strong
rather well. The deviation among the measurements base for future work and is interesting both from the
and simulations for both software applications is point of view of researchers as well as network
depicted in the second raw of diagrams in Fig. 4. It is developers.
evident that the difference between the diagrams on
the left- and on the right-hand side is minor. Thus, it can
be concluded that the results from the developed Acknowledgment
radio coverage tool are comparable with the results The authors gratefully acknowledge for suggestions,
from the TEMS software package. Some negligible advices and the results of measurements and radio
differences between the results originate from the cover predictions the Radio networks department at
fact that the implemented path loss model in the Mobitel d.d.
TEMS software used in the simulation is not entirely
available. Thus, it cannot be realized in the GRASS
software with all the minor details. References:
[1] M. Hata, Empirical Formula for Propagation
Loss in Land Mobile Radio Services, IEEE
5 Conclusion Transactions on Vehicular Technology, Vol. 29,
Efficient calculation of predicted radio coverage is No. 3, August 1980.
required for precise planning of cellular radio [2] S. R. Saunders, Antennas and Propagation for
telecommunication systems. Existing programming Wireless communication systems, John Wiley &
tools are either expensive or limited in functionality Sons, 1999.
and do not allow any modifications. Open-source [3] Y.Okumura, E. Ohmori, T. Kawano, K. Fukada,
systems enable adjustments to specific requirements Field Strength and its Variability in VHF and
of developers, and improvements of existing models, UHF Land-Mobile Radio Service, Review of the
based on measurements. Electrical Communication Laboratory, Vol. 16,
A representative open source system GRASS has No. 9-10, September-October 1968.
been presented in this paper. A description of the [4] G. L. Stuber, Principles of mobile
radio coverage prediction software developed in communications, Kluwer Academic Publishers,
GRASS has been given. The tool includes London 2001.
propagation and sectorisation modules, a module for [5] Planet and decibel Planner (Marconi),
radio signal coverage calculation, and additional http://www.ericsson.com/.
modules for preparing input data and analyzing [6] Vulcano, Siradel, http://www.siradel.com/.
simulation results. The r.radcov module, written in [7] CS telecom nG http://www.atdi.com/.
Python, which interconnects individual modules into [8] WinProp, AWE Communications,
a complet radio signal propagation software is also http://www.awe-communications.com/.
described. At the end, the developed software is [9] RPS - Radiowave Propagation Simulator,
evaluated by comparing to field measurements and http://www.radiowave-propagation-simulator.de/.
simulation result obtained from a professional [10] TAP - Terrain Analysis Tool,
software application. http://www.softwright.com/.
The radio signal coverage prediction software [11] M. Neteler, H. Mitasova, Open source GIS – a
implementation was quite straightforward, as API is GRASS GIS aproach, third edition, Springer,
well developed and documented. The set of built-in 2008.
C functions is adequate. The possibility to study [12] OSgeo, http://osgeo.org.
parts of the already implemented code is also very [13] GRASS 6 Programmer's Manual,
helpful. http://download.osgeo.org/grass/grass6_progman/
The developed software package with the [14] K. Bullington, Radio Propagation for Vehicular
implemented path loss modules gives satisfactory Communications, IEEE Transaction on Vehicular
accuracy compared to professional simulation tools. Technology, Vol. VT-26, No. 4, November 1977,
Additional model tuning based on field pp. 295-308.
measurements will be performed. In our future work, [15] D. J. Cichon, T. Kurner, Propagation prediction
we also plan to expand the functionalities of the models, COST 231 Final Rep., Available on:
developed software package and build additional http://www.lx.it.pt/cost231/.
path loss modules for urban and hilly rural [16] Ericsson Radio Systems AB, TEMS CellPlanner
environments that will include also the elements of Universal Common Features, Reference Manual,
ray tracing techniques and additional environment April 2006.

ISSN: 1792-4243 140 ISBN: 978-960-474-200-4