Validation study of path loss models on

International Journal of Next-Generation Networks (IJNGN) Vol.6, No.2, June 2014
DOI : 10.5121/ijngn.2014.6202 17
VALIDATION STUDY OF PATH LOSS MODELS ON
WIMAX AT 2.6 GHZ FREQUENCY BAND IN
SUBURBAN ENVIRONMENT FOR CELL SIZE
PLANNING
Pratibha Maina1
,Gopal Chandra Manna2
, Namrata Sahayam3
1,3
Dept, of Electronics and Communication Engg., Government Engineering College,
Jabalpur, M.P. India
2
Inspection Circle, BSNL, Jabalpur, M.P.India
ABSTRACT
The radio wave propagation in form of path loss model plays very significant role in planning of any
wireless communication network. Measurement of signal strength of OFDM driven WiMAX technology at
2.6 GHz band is taken in Suburban Town of India. The results are analyzed and compared with Empirical
path loss models such as Hata-Okumura, Modified Hata and COST-231Hata. COST-231 model shows
highest path loss for suburban environment. These analyzed results establish that COST-231 model is
suitable for suburban environment also. Threshold RSSI estimates cell coverage probability in the area.
KEYWORDS
OFDM, WiMAX, IEEE 802.16, path loss exponent, Rayleigh fading, RSSI, CINR
1. INTRODUCTION
Mobile WiMAX offers full mobility cellular networks at high speeds. It is based on OFDM
technology
Radio wave propagation depends on various factors like reflection, refraction, diffraction etc.
This results in multipath fading at the received location. The fading is vary with time slowly as
well as with distance from transmitter. If we consider an Omni-directional transmitter, we can
presume that the signal strength along the circumference of a fixed radius around the transmitter
shall be same but actually it is not. Scientists have calculated Cell edge probability and area
coverage probability based on parameters like initial path loss, path loss exponent (γ) and
Rayleigh fading factor (σ). Jakes graph is usually deployed for prediction of cell radius. Now,
commercially available planning tools are deployed for estimation of coverage which includes
clutter diagram and a suitable prediction model. Base transceiver station location is plotted on the
predicted coverage map plot. Present paper uses WiMAX network operating at a frequency of 2.6
GHz for the study of path loss and compare with existing propagation models. Two rural
locations of Rajasthan were carried out for studies with BTS height of 40m during the months of
November 2009 to February 2010. It was a hilly area and mostly covered by dense trees [1]. For
GSM, studies were carried out at dense city like Kolkata and medium to small towns like Raipur,
Jabalpur, Katni, Jagdalpur etc of India during 2006 to 2009[2]. To estimate the coverage of
WiMAX based on IEEE 802.16d at Kalyani, a suburban area of Kolkata City, a study was carried
out in 2007. Measurements were taken for rooftop antennas on line of sight (LoS) condition up to

18
4km and backhaul throughput capacity of 2 mbps due to practical limitations. The throughput was
observed full and signal quality was found adequate [3].
Kariapatti in Tamil Nadu, has mostly suburban terrain profile. Present study on WiMAX based on
IEEE 802.16e recommendation was conducted at suburban location of Tamil Nadu. The
observations were taken in the Evening time; however, sky was clear during the period of
observations. The area covers single storey buildings with medium height trees and few populated
area. The study measured all the above components for effective service and leads to estimate
values of various communication parameters essential for estimation of coverage.
In this paper, two optimized models are used namely Modified Hata model and COST-231 Hata
model to predict path loss for WiMAX in 2640MHz to 2650MHz .The COST-231 model showed
best agreement with the measured path loss in terms of path loss exponent as compared to Hata
model and Modified Hata model in estimating the path loss in the 2640MHz to 2650MHz.
Section 2. discusses various propagation models based on which WiMAX coverage is predicted
and a statistical approach for gross estimation of path loss exponent and fading component. The
various studies already conducted on several mobile communication technologies and results
obtained has been stated in Section 3. Base station setup, Subscriber Station location
arrangements and methodology of data collection has been elaborated in Section 4. Processing of
data and results obtained has been given in Section 5. followed by discussions and conclusion has
been drawn in Section 6.
2. PATH LOSS MODELS
Path loss is defined as the difference between transmitted power and received power,
Path loss = Transmitted power + gain – Received power (1)
Here, we have used the transmitting power of 40 dbi and transmitting antenna gain of 17 dbi.
Path loss is the reduction in signal strength when it propagates from Tx to Rx. Various parameters
reduces the signal strength such as height of antenna, distance between Tx and Rx, obstacles such
as trees, buildings,etc. transmitted power and antenna gain.
A general model which includes path loss exponent and shadow fading factor is given as
PL(dB) = PL(d0) + 10 γ log( d/d0) + σ (2)
Where PL (d0) the intercept, γ is the slope and σ is the standard deviation of received signals
[5].Losses are measured in dBm and d is in meters.
RSSI is the received signal strength on the receiver side. It is the indication of received power
level. On the basis of RSSI values one can obtain the communication range of base station. A plot
of Received Signal Strength Indicator (RSSI) with distance from transmitter may be considered as
radial coverage. According to Jake’s graph, if Px0(R) is the probability of radial coverage where
received signal threshold is x0 dBm at distance R, then
( ) = −
√
(3)
Where, is the mean of the received signal strength. From Jake’s graph, the area coverage
probability for a radius coverage probability can be obtained for the corresponding radius.
.

19
2.1. Hata Model
Okumura Hata model is widely used for voice coverage in below 1500 MHz range and the same
has been deployed with modification for prediction of WiMAX coverage which works in 2500
MHz and 3500 MHz band. There are three different environments namely urban, suburban and
rural environment. Hata model has different path loss model for each environment. This model
can be used for both point to point and broadcast transmissions[8]. For urban environment path
loss is given as:
PL( U) = 69.55 + 26.16 log (f ) − 13.82 log (ht) – ahm + 44.9 − 6.55 log (ht). log(d) (4)
Where,
PL = path loss in dB,
f = frequency in MHz ,
ht = height of the transmitting antenna in meters,
ahm = HATA correction factor,
d = distance in km .
For Sub-urban environment,
PL (SU) = PL (U) − 5.4 − 2[log10( )]2
(5)
For rural environment,
PL (R) = PL (U) + 18.33 log10 (f) − 4.78[log10 (f) ]2
– K (6)
Where, K ranges from 35.94 to 40.94 for countryside and desert respectively.
For small to medium cities and for large cities the correction factor ahm is given by
ahm = (1.1 log10 (f) − 0.7)hr − (1.56 log10 (f)− 0.8) (7)
ahm = 3.2(log10 11.75. hr )2
− 4.97 (8)
2.2. Modified HATA Model
Modified Hata model is the modification of Hata model. It is modified for making it suitable for
given environment [8].
For urban environment,
PL (U) = 69.55 + 26.16 log10 (f) − 13.82 log (ht) – ahm
+ 44.9 − 6.55 log (ht). log(d) + Cm(U) (9)
For sub-urban environment,
PL(SU) = PL(U) − 5.4 − 2[log10( ) ]2
(10)
For rural environment,
PL(R) = PL(U) − 4.78[log10(f) ]2
+ 18.33 log10 (f) − Cm(R) (11)
Where, Cm (U) and Cm(R) are the correction factors for urban and rural environments and taken
as 12 and 29 respectively.

20
2.3. COST-231 HATA Model
COST-231 has also been used for calculation of path loss with correction factor Cm=0 as
applicable for sub-urban centers with medium tree density [4]. COST-231 is an extended form of
Hata model and used in the frequency range of 1500-2000 MHz. Here it is used from 2300 MHz
to 2600 MHz for WiMAX system. The path loss for this model is calculated as:
PL(dB) = 46.33 + (44.9 – 6.55log10(ht)) log10 (d ) + 33.9log10(f) -13.82log10(ht) - ahm + Cm (12)
Where,
f = frequency in MHz,
d = distance between BTS and receiving antennae in km
ht = transmitting antenna height in meters
The correction factor, Cm is given as Cm(U) = 3dB and Cm(SU) = 0dB respectively
The correction factor ahm for urban and suburban environments are defined as
ahm = 3.2(log10(11.75hr))2
- 4.97 (13)
ahm = (1.1 log10 ( f) -0.7)hr – (1.56 log10 (f) - 0.8 ) (14)
Where, hr is the receiving antenna height in meters.
3. RELATED WORKS
A comparison were made among different environments of Bangladesh using different path loss
models at 2.5 GHz frequency band where Free Space Path loss model referred for LOS condition
and SUI model showed lowest path loss and ECC-33 model showed highest path loss for NLOS
condition [6]. During 2010, optimization of HATA propagation prediction model in suburban
area was conducted in Malaysia by R. Mardeni and K. F. Kwan [7]. Similar study was conducted
at the sub-urban area of Kalyani, Kolkata for LOS and NLOS conditions where terrain is mostly
flat and building height is averaged at 7m [3]. Modified Hata model have been used for path loss
calculations which has proven more appropriate for GSM cellular communication [8]. Path loss
modeling for vehicle-to-vehicle communication were carried out for four different environments
e.g. highway, rural, urban, suburban. Measurement was conducted in Lund, Sweden, during
January 2011[9]. Different propagation models that would be used for Long Term Evolution
(LTE), a comparison were made among them and SUI showed lowest path loss for all
terrains[10]. A pilot project is being conducted at the IRT (Institute fur Rundfunk technik) to test
the digital video broadcasting ability of both fixed and mobile WiMAX at Munich city during
2009[11]. SUIT (Scalable, Ultra-fast and Interoperable Interactive Television) conducted field
trial at Aveiro-Portugal for mobile WiMAX over 10 MHZ bandwidth and at speed up to 140
kmph and measured throughput in sub-urban city environment for both unicast and multicast
mode during 2009[12]. In comparison to ECC-33 model, Hata Okumura and COST-231 models
showed the highest probability for same parameters. In addition the received signal strength
(RSS) of base station with noise and without noise is calculated for the same area by Mukesh
Kumar, Vijay Kumar, Suchika Malik during during February 2012[13]. Radio Measurements in
the WiMAX Band of 2.3 GHz for different antenna heights, was conducted in the urban coastal
region of Mumbai during June 2012 by Chhaya Dalela [14]. VoIP and video streaming
performance was measured over mobile WiMAX testbed deployed at the campus of Institute
Telecom SudParis during the period May 2012[15].
Study of all above reference reveals that there is very little approach to evaluate γ for Indian
environment. Further there is no installation which covers OFDM at 2.3 GHz range. In the
present study, OFDM at 2.6 GHz band has been used and γ has been calculated. The path loss

21
model thus obtained may be scaled to estimate path loss at a lower frequency range of 2.3 GHz
band. The results for both the bands are nearly same.
4. STUDY SETUP AND OBSERVATION
4.1. Base Station
Base Station is installed at Kariapatti in Tamil Nadu province of India. Kariapatti is located
around 24 Km in Tamil Nadu. It is situated near Madurai-Thuthookudi National Highway 45B.
Kariapatti tower is just 2 Km from National Highway in east direction shown in fig 1. The tower
shares other Antenna assemblies also and locations of WiMAX antenna assembly (left) and Radio
Resource Unit (RRU) (right) are shown as inset. The antenna assembly is located at a height of
35m.The output power at antenna connector is set at 40 dBm and antenna used has dual element
system (2 Rx/1 Tx) with gain of 17 dbi. The antenna orientations are set at 600
-1800
-3000
respectively. The frequencies used are in of 5 MHz bandwidth each and from 2.640 GHz to 2.650
GHz. The system uses Time Division Duplexing, convolution turbo code and Adaptive Multiple
Antenna.
4.2. Measurement environment
4.2.1. Alpha Sector Direction
The way of direction which is marked as Y goes towards few populated area. At around 0.065km
the Rx is in a close proximity to BTS results in a fall of signal shown in fig 9 by point A. At point
B around 0.204 km the way is surrounded by few populated area comprising of maximum single
storey buildings LOS exist between Tx and Rx and allow the signal to propagate with little
suffering from diffraction, reflection, absorption and scattering. Around 0.204 km to 0.659 km on
the left side of the way having many buildings stops the signal to reach the receiver results in fall
of signal which is indicated by point B to C. D to E is inside the city even if LOS is available, the
received signal undergoes deterioration/ improvement due to additional reflections and scattering
by road and local moving traffic e.g. bus, taxi. At the distance around 1.205 km to 1.554 km again
having number of buildings on left side causes poor LOS as shown in fig from E to F. On the way
from F to G around 1.554km to 1.974 km is a regular terrain with a few medium height trees
which causes ups and down of signal. From 1.974km towards National Highway at point G on the
right side of the way are having a few populated areas the signal is reflected from the buildings
and reaches the Rx. At the end of the way up to 2.5 km towards Madurai the Rx is on the National
Highway45B shown by point H. Field strength of signal decreases with increase in distance
between BTS and Rx.
4.2.2. Gamma Sector Direction
The way of direction which is marked as X goes towards BTS away from Madurai-Thuthookudi
National Highway 45B. The Kariapatti BS surrounded by moderately spaced one to two storey
buildings with trees. The landscape is even and the road crosses area having a factory on the left
side around a distance from 3.564Km to 3.394km,the signal is reflected from the building and
reaches the receiver causes LOS which shown in the fig6 from point E to D. From a distance of
3.6 to 3.4 km is an open field area shown in fig from point D to C. Around 2.5 km towards BTS
there is a regular terrain up to a distance of 1.58 km marked from C to B shows continuous fall of
signal. The signal obtained at point B is quiet weak shows frequent NLOS. From point B to A
around 1.5 to 0.103 km signal strength increases slowly as the distance between BTS and Rx is
decreasing. Route trace of Kariapatti Base station is shown in fig 1.

22
4.2.3. Subscriber Station (SS) Setup
An outdoor SS was chosen for the measurements. The SS, laptop, dongle and GPS all were
mounted inside a vehicle. The Global Positioning System (GPS) is a space-based satellite
navigation system that provides location and time information in all weather conditions,
anywhere on or near the Earth where there is an unobstructed line of sight to four or more GPS
satellites. A dongle is a small piece of hardware that attaches to laptop and that enables additional
functions such as audio, video, data, or other services. On the roof of the vehicle an outdoor
antenna was mounted to get maximum signal strength. The total height of the antenna was at 3 m
which, on an average, was equal to height of average buildings in suburban areas and line of sight
signal was mostly obtained. The output at the antenna connector was 23 dBm and had a gain of 3
dBm. The laptop was loaded with driver software for measurement of signal parameters and
attached to a server. RF performances were also monitored at Access Service Network (ASN)
Gateway for uplink parameters.
Subscriber substation is shown in fig.(2).
5. RESULTS AND DISCUSSIONS
All CINR, path loss exponent and path loss with distance, RSSI obtained for Gamma Sector are
shown in figs 3, 4, 5 and 6 respectively and the corresponding values for Alpha Sector at figs 7, 8,
9 and 10. RSSI values were compared with COST-231 model, Hata model, Modified Hata model
and path loss plotted with parameters applicable for sub-urban. It is observed that the path loss
slope is much less compared to other three models with high initial LOS. Figs 4 and 8 give path
loss exponents for Alpha and Gamma Sector respectively.
For Alpha sector, RSSI was better than -70 dBm up to a distance of 0.403 km shown in fig 9
marked by B. After which it drops from -80 to -85 dBm up to 0.72 km at C. From C to E it is
observed to improve again to -70dbm up to 1.268 km. CINR varied between 15 to 25 up to 1.301
km. RSSI is again varied between -70 to -85 up to 1.941 km and better than -70 dam at 1.941 km
indicated from E to G. CINR varied between 10 to 25 up to 1.941 km and between 15 to 20 up to
2.443 km. From G to H RSSI varied between -75 to -80 up to 2.442 km, after which it drops to -
90 dBm at 2.564 km. RSSI is observed to mostly varied between -70 to -80 dBm whereas CINR
between 15 to 25.
For Gamma Sector, RSSI varied between -60 dBm to -95 dBm up to a distance between 0.103 to
1.58 Km marked from A to B which shows frequent drop of signal at point B and CINR is nearly
2.5, after which the RSSI was observed to improved slowly up to a distance of 2.5 Km and better
than -70 dBm at point C and CINR is better than 20. From point C to D it was observed to vary
between -65 to -80 dBm up to 3.5 km. CINR varied between 25 to 10.
Observation of Alpha Sector and Gamma Sector were regular for terrain. For Gamma Sector there
was a continuous open field area up to several kms where direct signal is received by the receiver
which affects RSSI and CINR. The drop of signal shown in fig (6) is indicated by B which shows
the rapid signal drop. For Alpha Sector when the path loss is plotted against distance it was
shown that path loss increased as the distance between the Tx and Rx is increased hence shows
good results for RSSI and CINR.
The rate of propagation path loss is plotted against distance shown by the path loss exponent. If
the path loss exponent value is 2, then the environment propagation characteristic is close to free
space propagation. Overall path loss exponent (γ) for Alpha Sector was 2.351 and fading (σ)
5.20415 fig (12); the corresponding values for Gamma Sector were 2.196 and 6.93145 fig (11)
respectively.

23
Fig 1: Kariapatti BS (red) with. X and Y route towards north
Fig 2: Subscribers substation

24
Fig 3: Gamma Sector CINR
Fig 4: Gamma Sector Path loss exponent
Fig 5: Gamma Sector Path loss with distance

25
Fig 6: Gamma Sector RSSI
Fig 7: Alpha Sector CINR
Fig 8: Alpha Sector Path loss exponent
25
25

26
Fig 9: Alpha Sector RSSI
Fig 10: Alpha Sector Path loss with distance
26
26

27
Fig 11: Signal strength distribution and Fig 12: Signal strength distribution and
Rayleigh fading factor estimation Rayleigh fading factor estimation
for Gamma Sector. for Alpha Sector
Fig:13 Jakes graph
6. CONCLUSION
The obtained value of RSSI is higher than the calculated value which lead to overall value of γ<2
whereas for free space where γ is equal to 2. Low γ value indicates better coverage than predicted
by all models used by planning tools. This observation is similar to GSM as obtained in earlier
observations. However this situation continued up to 1 Km after which nearly free space
propagation was observed. The σ/γ value was observed as 2.6850 which is an average obtained
from above path loss exponent (γ) for Alpha Sector are 2.351 and fading (σ) 5.20415 fig (12); the
corresponding values for Gamma Sector are 2.196 and 6.93145 fig (11) respectively. The Cell
edge probability and corresponding area coverage probability can be stated with better degree of
confidence level using Jakes Graph; however RSSI was observed to be nearly -90 dBm at point
B; this drop of signal is shown in fig (6) where Jakes Graph is nearly conical as in fig (13). If -85
dBm is considered as threshold with a mean value of RSSI as -78 dBm as in fig (11) and average
27
Fig:13 Jakes graph
6. CONCLUSION
27
Fig:13 Jakes graph
6. CONCLUSION

28
σ/γ = 2.6850, we get radius coverage probability as 0.87 using equation (2). From fig (13) with
0.87 as Px0 (R) profile, the corresponding area coverage probability is ~95% for a radius
~2500m as evident from fig (8) and fig (9).Jakes graph and the graphs drawn on the basis of this
experiment are sufficient for cell planning of an area.
REFERENCES
[1] Gopal Chandra Manna, Bhavana Jharia “Mobile WiMAX Coverage Evaluation for Rural Areas of
India” Advanced Communication Technology (ICACT),13th International Conference on 13-16 Feb.
2011 at Seoul (India).
[2] S.K. Sarkar and G.C. Manna, “Comparison Among GSM, CDMA AND WIMAX As Fast Track
Access Network For Fourth Generation Mobile Terminal”, National Convention and Seminar on
Mobile Spectrum: Issues and Challenges to Engineers- Institution of Engineers (India) at Jabalpur, on
24th and 25th October, 2009
[3] S.K. Sarkar and G.C. Manna, “Performance Evaluation of IEEE 802.16 based System in
Sub-urban Area” , Telecommunications, Vol-59, Issue 2, March-April 2009.
[4] Mardeni. R, T.Siva Priya,“Optimized COST-231 Hata Models for WiMAX Path Loss Prediction in
Suburban and Open Urban Environments” Vol. 4, No. 9; September 2010.
[5] Taimoor Abbas, Johan Karedal and Fredrik Tufvesson, “A Measurement Based Shadow Fading
Model for Vehicle-to-Vehicle Network Simulations” February, 2014 - IEEE Xplore digital library.
[6] Md. Didarul Alam1 and Md. Rezaul Huque Khan2, “ Comparative Study of Path Loss Models of
WiMAX at 2.5 GHz Frequency Band” in International Journal of Future Generation Communication
and Networking Vol. 6, No. 2, April, 2013.
[7] R. Mardeni and K. F. Kwan “Optimization of HATA Propagation Prediction Model In Suburban Area
in Malaysia” Progress In Electromagnetics Research C, Vol. 13, 91-106, 2010.
[8] P.Saveeda, E.Vinothini, Vardhi Swathi and K.Ayyappan, “Received Signal Strength (RSS)
Calculation for GSM Cellular System at BSNL Pondicherry using Modified HATA Model”in IJSETR
Volume 2, Issue 1, January 2013.
[9] Johan Karedal, Alexander Paier, Fredrik Tufvesson and Andreas F. Molisch, “Path Loss
Modeling for Vehicle-to-Vehicle Communications” IEEE transactions on vehicular technology, Vol.
60, no. 1, January 2011.
[10] Noman Shabbir1, Muhammad T. Sadiq2, Hasnain Kashif3 and Rizwan Ullah4,“Comparison of
Radio Propagation Models For Long Term Evolution (LTE) Network” International Journal of Next-
Generation Networks (IJNGN) Vol.3, No.3, September 2011.
[11] Hennann Lipfert, Alois Zistler, Aylin Vogl, Madeleine Keltsch, “Performance Testing in a
WiMAX-Pilot at the Institute fur Rundfunktechnik”, April 8, 2009 - IEEE Xplore digital library.
[12] N. Coelho, N. Cabral, A. Pereira, A. Rocha and A. Navarro, “Mobile WiMAX Field Trials in a Sub-
urban Area”,IEEE Xplore Digital Library.
[13] Mukesh Kumar , Vijay Kumar, Suchika Malik, “Performance and Analysis of Propagation
Models for Predicting RSS for Efficient Handoff” International Journal of Advanced Scientific and
Technical Research Issue2, volume 1, February 2012 .
[14] Chhaya Dalela , “Radio Measurements in the WiMAX Band of 2.3 GHz, in Coastal Zone for
Different Transmitting Antenna Heights” in IJEAT, Volume-1, Issue-5, June 2012.
[15] Hongguang Zhang1, Mohammed Boutabia2, Hang Nguyen2, Noël Crespi2, Ai-Chun Pang3, Liang
Zhou4, Jianming Wei1“Mobile WiMAX Field Trial Test through Multimedia Performance
Evaluation” Author manuscript, published in "EURASIP Journal on Wireless Communications and
Networking (2012)."
Authors
Pratibha Maina has completed her Bachelor of Engineering degree in Electronics and
Communication Engineering from Samrat Ashok Technological Institute ,Vidisha, India.
She is currently an Master of Engineering (Communication system) student at Jabalpur
Engineering College (M.P.), India. Her area of interest is Wireless Communication.

29
Dr. Gopal Chandra Manna had done Ph.D. from Electronics and Telecommunications
Engineering of Jadavpur University, Kolkata. He had both graduated and post graduated
in Radio Physics and Electronics Engineering from University of Calcutta and undergone
trainings at Beijing University of Post and Telecom China in 1990 and DARTEC,
Montreal, Canada in 1999. He is working as Senior General Manager, Inspection Circle,
BSNL, Govt. of India. GSM, CDMA, WCDMA and WiMAX radio access are his area of
research. An one week course on Quality of Service Monitoring at ICTA, Mauritius as International
Expert through CTO, London during 2010 has been developed and conducted by him. From 1997 to
2002, he had worked as Deputy General Manager in Telecommunication Training Centre of DoT and
installed live training node for Internet Service Provider(ISP). He had conducted training programs for
participants from Asia Pacific Telecommunications (APT) and Sri Lankan Telecom and seminars with
international experts of UNDP/ITU.
Namarata Sahayam received the Bachelors Degree in Electronics and Communication
Engineering from Government Engineering College, Rewa, India in 2003 and Mtech in
Digital Communication from RGPV, Bhopal in 2008. She is currently working as an
Assistant Professor in Department of Electronics and Communication Engineering of
Jabalpur Engineering College (M.P.), India. Her research of interest is in the field of
Wireless Communication and Signal Processing.

Validation study of path loss models on

More Related Content

What's hot (15)

Viewers also liked (13)

Similar to Validation study of path loss models on (20)

More from ijngnjournal (20)

Recently uploaded (20)

Validation study of path loss models on