ADAPTATION OF THE OKUMURA-HATA

MODEL TO THE ENVIRONMENT OF

ACCRA
Joseph Kweku Arthur, PhD Ashitei Theophilus Amartey,MPhil William Brown-Acquaye, PhD
Faculty of Engineering /Department of Faculty of Engineering/ Department Of Faculty of computing/Department of
Telecommunication Engineering Electrical Engineering Information Systems
Ghana Technology University College Ghana Technology University College Ghana Technology University College
Accra, Ghana Accra,Ghana Accra, Ghana
[email protected] [email protected] [email protected]

Abstract— Interference is a major drawback for cellular During the 90’s, cellular systems became digital. Japan
systems. Eliminating interference is near impossible hence the migrated to PDC (Personal Digital Communication), the
need to employ ways or algorithms to reduce or mitigate its USA developed the first Coded Division Multiple Access
effect. Radio Frequency interference during the propagation of (CDMA) standard IS-95 and Europe standardized GSM
a signal can degrade the quality of a cellular network (pathloss
between the eNB and user equipment (UE)). With the adoption
(Global System for Mobile Communication), the most
of the 4G networks in Ghana, it is imperative to explore ways by successful technology from the 2nd generation deployed all
which a network can be deployed or expanded by predicting the over the world and still carrying important amounts of voice
propagation path loss. This is done by modifying the Okumura- traffic nowadays. GSM is responsible for spreading the
Hata (OH) pathloss model by using the linear least squares massive use of cellular phones mainly for voice and text
method(LLSM) with the aim of better predicting the path loss messages (SMS) but the data services were slow and limited;
and thereby managing interference in the environment for first under GPRS (General Packet Radio Service), then using
which it may be implemented. The modified model has proven EDGE (Enhanced Data rates for GSM Evolution) with rates
to be an effective way of predicting the path loss for 4G in the order of 150 Kbit/s per cell [2]
networks. The terrain environment mismatch between the
predicted pathloss and the proposed pathloss is taken care of by The data rate capacity kept increasing up to 42 Mbit/s
the LLSM. A standard deviation of 10.2 of the empirical model with the introduction of HSPA and HSPA+ (High Speed
as opposed to 10.5 of the proposed model, which shows the Packet Access) but the market’s response being
dispersion in the data set. This makes it possible to deploy a 4G overwhelmingly positive led the 3GPP consortium into
network with less pathloss in an urban city like Accra, which
meets the requirements of an urban large city but with few high- defining new technical requirements for the 4th generation of
rise structures. cellular networks, called the Long Term Evolution (LTE).
Drastic changes on the core network and the introduction of
Keywords— Okumura-Hata propagation model; 4G network; Orthogonal Frequency Division Multiple Access (OFDMA)
pathloss; Linear Least Squares Method and Multiple-antenna (MIMO) radio technologies allow data
I. INTRODUCTION rates up to 150 Mbit/s in the downlink and 50 Mbit/s in the
uplink. At the beginning of 2013, LTE deployments had
The genesis of mobile communications dates to the 1940s captured more than 50 million clients in some 20 countries
where only in the United States thousands of analogue radios and strong investments make promise for worldwide use of
were used mainly for public services. In the early 60’s the this technology in the short-term future [3].
number of mobile users in that country grew to more than a
million with the development of cordless telephones, Interference among communication systems is a common
however, they were not connected to the public switched problem. It is widely held that stronger interfering signals
telephone network and the coverage areas were restricted [1]. cause more harm. In fact, “harmful interference” is often
described as an excess of interfering signal power at the
Important advances in related areas such as signal receiver [4]. This degradation does not necessarily result
processing and electronics enhanced the capacity of the radio from theoretical limitations but from the techniques
devices to the point that in the early 80’s the cellular concept, employed [5]. Most Fourth-Generation (4G) systems,
previously developed in Bells Labs gave birth to the first including WiMAX 802.16m [6] are targeting single
generation of cellular networks. There were variety of frequency capacity, it also severely degrades the performance
technologies such as AMPS (Advanced Mobile Phone experienced by cell edge users due to the increased
System) in the USA, TACS (Total Access Communication interference caused by out-of-cell transmission [7].
System) in Japan and the UK, NMT (Nordic Mobile Interference management schemes are critical to improve the
Telephone) in the Scandinavian countries, Radiocom2000 in performance of the network. This can also be done when
France and C-NETZ in Germany. These systems were based predicting the environment for the network to be deployed by
on analogue modulation and coding, frequency division considering the various pathloss models.
multiple access (FDMA), and were very limited in terms of
coverage. Additionally, the user terminals had important The implementation of 4G standards by service providers
dimensions, which limited their portability will soon be in the ascendancy. MTN-Ghana (voice and data
service provider) being the first to implement. Interference
impacts negatively on the performance of the network.
Interference which is because of various factors can be

Authorized licensed use limited to: Universiti Teknikal Malaysia Melaka-UTEM. Downloaded on June 24,2024 at 11:08:41 UTC from IEEE Xplore. Restrictions apply.
captioned in the pathloss models which considers all the
various aspects of interference in an urban city such as Accra.
This research aims at modifying the Okumura-Hata
model, which was developed using environmental conditions
of Japan. The proposed model will take into account peculiar
environmental conditions of Accra,Ghana.
This will benefit the network operators tremendously.
Deployment of a new technology or the expansion of an
existing network can be capital intensive, by predicting the
environment and optimizing the pathloss by way of managing
interference using a simulation software can be very useful to
the network operators. Various components and parameters
of the simulated network can be varied or configured to either
improve the network or best suit the network for the operator.
Interference affects the propagation of a signal tremendously
and can reduce the QoS (Quality of Service) and QoE
(Quality of Experience) of a network and its capacity.

Figure 1 LENA model overview (Source: Nicola Baldo -

II. RESEARCH METHODOLOGY
The LENA ns-3 module)
The research looks at predicting the pathloss for a 4G
A. PATHLOSS MODEL
network in a large city like Accra. This is done by
considering Radio Frequency interference during the The Okumura-Hata model for path loss is considered in
propagation of the signal by predicting the area for which it simulating the path loss between the eNB and the UE. The
may be implemented. This section also covers simulations of path loss equation is given as follows:
the modified modelled problem and suggests solutions using
the Okumura-Hata in NS3 (network simulator 3). 𝑃𝐿(𝑑𝐵) = 69.55 + 26.16 log10 𝑓𝑐 − 13.82 log10 ℎ𝑏 −
𝑎(ℎ𝑢 ) + 𝐵 log10 (𝑑) eqn (1)
NS3 is a discrete-event network simulator, targeted
primarily for research and educational purposes. NS3 is a free Where:
software licensed under the GNU GPLv2 license [8], and 𝐵 = 44.9 − 6.55 log10 ℎ𝑏 eqn (2)
publicly available for research, development and use.
𝑓𝑐 is the carrier frequency in MHz.
The LTE model selected for this research is developed by
the LENA project as shown in fig. 1. LENA is an open ℎ𝑏 is the height of the eNB antenna.
source product-oriented LTE/EPC network simulator that ℎ𝑢 is the height of the user equipment
allows LTE small/macro cell vendors to design and test Self
Organized Network (SON) algorithms and solutions. Target d is the distance of separation in kilometres (km)
applications for LENA include the design and performance
This model is used to model an open area pathloss for
evaluation of the downlink (DL) and uplink (UL) schedulers,
long distances (i.e. > 1 Km) to include all the possible
Radio Resource Management Algorithms, Inter-cell
frequencies usable by LTE. In fact, the original Okumura
Interference Coordination solutions, Load Balancing and
Hata model was designed for frequencies ranging from 150
Mobility Management, Heterogeneous Network (HetNets)
MHz to 1500 MHz. The COST231 extends it for the
solutions, End-to-End QoE provisioning, Multi-RAT network
frequency range from 1500 MHz to 2000 MHz. Another
solutions and Cognitive LTE systems. LENA is based on the
important aspect is the scenarios considered by the models.
popular ns-3 network simulator for internet systems.
All the models were originally designed for urban scenario
except the standard OH and the COST231 OH, which are
extended to suburban. Therefore, the model cannot cover all
scenarios at all frequencies. The following are inputs for the
proposed model
The pathloss expression of the COST231 OH is:
𝐿 = 46.3 + 33.9 log 𝑓 − 18.82 log ℎ𝑏 + (44.9 −
6.55 log ℎ𝑏 ) log 𝑑 − 𝐹(ℎ𝑀 ) + 𝐶 eqn (3)
Where
𝐹(ℎ𝑀 ) =
(1.1 log(𝑓)) − 0.7 × ℎ𝑀 − (1.56 × log(𝑓) − 0.8)
{ + eqn (4).
3.2 × (log(11.75 × ℎ𝑀 ))2
For medium and large cities

Authorized licensed use limited to: Universiti Teknikal Malaysia Melaka-UTEM. Downloaded on June 24,2024 at 11:08:41 UTC from IEEE Xplore. Restrictions apply.
 0𝑑𝐵 𝑓𝑜𝑟 𝑚𝑒𝑑𝑖𝑢𝑚 − 𝑠𝑖𝑧𝑒 𝑐𝑖𝑡𝑖𝑒𝑠 𝑎𝑛𝑑 𝑠𝑢𝑏𝑢𝑟𝑏𝑎𝑛 𝑎𝑟𝑒𝑎𝑠 The creation and configuration of the LTE entities was
𝐶={ made possible by the module “lte-helper.h”. The general
3𝑑𝐵 𝑓𝑜𝑟 𝑙𝑎𝑟𝑔𝑒 𝑐𝑖𝑡𝑖𝑒𝑠
responsibility of the lte-helper is to create various LTE
And objects and arrange them together to make the whole LTE
𝑓 = frequency (MHz) network, that is:
ℎ𝑏 = eNB height above the ground (m)  downlink spectrum channel (pathloss model/fading
model)
ℎ𝑀 = UE height above the ground (m)  uplink spectrum channel (pathloss model/fading
𝑑 = distance (km) model)
 eNodeB node(s) (mobility model, eNodeB device(s),
𝑙𝑜𝑔 = is a logarithm in base 10 EPC related models)
This model is only for urban scenarios. The pathloss  UE node(s) (mobility model, UE device(s))
expression of the standard OH in urban areas is: Spectrum channels were created automatically: one for
𝐿 = 69.55 + 26.16 log 𝑓 − 13.82 log ℎ𝑏 + (44.9 − DL and one for UL. The antenna model used was defined as
6.55 log ℎ𝑏 ) log 𝑑 − 𝐶𝐻 eqn (5) an isotropic antenna model with a maximum gain of 18dB
[9]. The uplink and downlink bandwidths are set to 25 MHz.
Where for small or medium sized city the E-UTRA band used is the band 3, which has a downlink
𝐶𝐻 = 0.8 + (1.1 log 𝑓 − 0.7)ℎ𝑀 − 1.56 log 𝑓 eqn (6) range of 1200-1947 and an uplink range of 19200-19949. The
pathloss environmental attribute was set to urban
and for large cities environment and the city size set to large city. This is because
for the deployment of 4G or expansion of an existing network
8.29(log(1.54ℎ𝑀 ))2 − 1.1 𝑖𝑓 150 ≤ 𝑓 ≤ 200
𝐶𝐻 { to 4G, network operators usually begin from areas, which
3.29(log(11.75ℎ𝑀 ))2 − 4.97 𝑖𝑓 200 ≤ 𝑓 ≤ 1500 will fetch them the needed revenue. First, the various nodes
eqn (6) are created, that is one eNodeB and 10 UE nodes as shown in
fig 2.
The extension for the standard OH in suburban is
𝑓 2
𝐿𝑆𝑈 = 𝐿𝑈 − 2 (log ) − 5.4 eqn (7)
28

Where
𝐿𝑈 = pathloss in urban areas
The extension for the standard OH in open area is
𝐿𝑂 = 𝐿𝑈 − 4.70(log 𝑓)2 + 18.33 log 𝑓 − 40.94 eqn (8)

B. MODELLING OF THE PROPOSED SYSTEM

USING THE OKUMURA-HATA MODEL IN NS3
The system was modelled by importing the various
modules or header files required by the LENA NS3 topology.
The core modules include
 core-module.h
 network-module.h Figure 2 Sample simulation view on netanim
 lte-module.h The eNodeB is set to a constant position with a height of
 mobility-module.h 50m. The UE positions are also set to constant positions with
 config-store.h varying distances from the eNodeB, which are set from 1km
 lte-helper.h to 10km. The EPS(evolved packet systems) bearer which is a
 epc-helper.h virtual connection between UE and PGW(Packet Data
 gnuplot.h Network Gateway), identifies a data sent and received
 point-to-point-module.h between two end points with specific QoS attributes [10] is
 propagation-module.h activated on all the UEs, which also activates the data radio
 lte-global-pathloss-database.h bearer. The pathloss is then monitored in two centralized
objects, one for downlink and the other for up-link. The
The command line arguments used to run the program are distances between the UEs and the eNodeB is kept in another
the enbDist, which is the distance between two eNodeB. The object.
radius of the disc or eNodeB where UEs are placed around
the eNodeB was set at 10,000 m. In addition, “simTime”
which is the total duration of the simulation in seconds was
also set at 20s.

Authorized licensed use limited to: Universiti Teknikal Malaysia Melaka-UTEM. Downloaded on June 24,2024 at 11:08:41 UTC from IEEE Xplore. Restrictions apply.
 C. THE MODIFIED PATHLOSS MODEL I – 𝐿𝑜𝑠𝑠𝑢𝑟𝑏𝑎𝑛 (𝑑𝐵) = 69.55 + 26.16𝑙𝑜𝑔10 (𝑓) −
13.82𝑙𝑜𝑔10 (ℎ𝑏 ) − 𝑎(ℎ𝑚 ) + (44.9 −
Linear Least Squares Method (LLSM) was used to
6.55𝑙𝑜𝑔10 (ℎ𝑏 ))𝑙𝑜𝑔10 (𝑑)
achieve a minimum mean error and acceptable standard
deviation between measured (these values have been J – Is f< 200MHz
assumed) and predict pathloss [11].
K – 𝑎(ℎ𝑚 ) = 3.2(𝑙𝑜𝑔10 (11.75ℎ𝑚 ))2 − 4.97
L – Is f< 200MHz?
Start M – 𝑎(ℎ𝑚 ) = 823(𝑙𝑜𝑔10 (1.54ℎ𝑚 ))2 − 1.1
(a ∗b ) (−a12 ∗b2 ) (−a21 ∗b1 )
N – A1 = ( 22 1 ) + ( ) ; A2 = ( )+
det det det
A ((a11 ∗ b2 )/det)
O – 𝑎11 = 𝑙𝑜𝑔10 (𝑑[𝑖]); 𝑎12 = (𝑙𝑜𝑔10 (𝑑[𝑖])2 ; 𝑎21 +=
1; 𝑎22 = 𝑙𝑜𝑔10 (𝑑[𝑖]); 𝑏1 = ∆𝐿[𝑖] ∗ 𝑙𝑜𝑔10 𝑑[𝑖]; 𝑏2 =
B E
∆𝐿[𝑖]; 𝑑𝑒𝑡 = (𝑎11 ∗ 𝑎12 ) − (𝑎12 ∗ 𝑎21 );
NO
STOP – 𝐿𝑜𝑠(𝑑𝐵) = (46.3 + 𝐴1 ) + (33.9 ∗
NO YES
C D F
(𝑙𝑜𝑔10 (𝑓)) − 𝑎(ℎ𝑚 ) + ((((44.9 + 𝐴2 ) − (6.55 ∗
YES
𝑙𝑜𝑔10 (ℎ𝑏 ))) ∗ 𝑙𝑜𝑔10 (𝑑𝑖𝑠𝑡)) − 𝑎(ℎ𝑚 ) + 𝐶
NO
G H

YES
D. Mathematical Analysis
NO
J K
Lp – theoretical pathloss
YES
I Lm – measured pathloss
NO
L ∆L – difference between predicted and measured pathloss
YES A1 – additional factor for compensation of constant
attenuation
M i=0
A2 – additional factor for compensation of attenuation
about distance(d)
NO i<10? N
STOP
Lmod – the modified mathematical model, considering the
YES terrain environment of Accra.
O

Pathloss for Cost231-Okumura-Hata model:

STOP Lp(dB) = K1 + 33.9log10(f) – 13.82log10 – a(hm) + K2log10(d)
+ (-6.55log10(hb)log10(d)) + Cm eqn (9)
Figure 3 Flow chart of algorithm used in the modification of where
the empirical model
f: is frequency in MHz
Where
hb: is the effective eNodeB transmitter antenna height in
A – Get the f,hm,hb,d,city size and the environment meters
B – Declare 𝑎11 , 𝑎12 , 𝑎21 , 𝑎22 , 𝑏1 , 𝑏2 , 𝑑𝑒𝑡, ∆𝐿 hm: is the effective UE receiver antenna height in meters
C – F<=1500MHz? d: is the distance between eNodeB and UE in Km
D – citySize=LargeCity? a(hm): is the correction factor for effective UE antenna
E– 𝑎(ℎ𝑚 ) = 1.1 ∗ 𝑙𝑜𝑔10 (𝑓) − 0.7 ∗ ℎ𝑚 − (1.56 ∗ height
𝑙𝑜𝑔10 (𝑓) − 0.8) Cm: is 0 for median sized cities and suburban areas and
F – (ℎ𝑚 ) = 3.2 ∗ ((𝑙𝑜𝑔10 (11.75 ∗ ℎ𝑚 ))2 , C=3; 3dB for metropolitan areas
a(hm) for Urban area is:
G – citySize=LargeCity?
a(hm) = 823 (log10(1.54hm))2 – 1.1 for f<300MHz eqn (10)
H – 𝑎(ℎ𝑚 ) = 0.8 + (1.1 ∗ 𝑙𝑜𝑔10 (𝑓) − 0.7) ∗ ℎ𝑚 −
2
1.56 ∗ 𝑙𝑜𝑔10 (𝑓) a(hm) = 3.2(log10(11.75hm)) – 4.97 for f>300 MHz eqn (11)

Authorized licensed use limited to: Universiti Teknikal Malaysia Melaka-UTEM. Downloaded on June 24,2024 at 11:08:41 UTC from IEEE Xplore. Restrictions apply.
by adding A1 and A2 factors for tuning purpose, the Lmod 5 168.45 162.354
equation becomes:
Lmod(dB) = (K1 + A1) + 33.9log10(f) – 13.82log10(hb) – a(hm) + 6 171.124 165.101
(K2 + A2)log10(d) + (-6.55log10(hb)log10(d)) + Cm eqn (12)
7 173.385 167.424
For each test of certain station f, hb , hm are all fixed
values but the distance is varied [11].
8 175.343 169.436
∆L = (Lp – Lm)
9 177.071 171.211
The error equation can be written as:
E (A1 , A2) = ∑ni=1(∆Li – (A1 + A2log10di))2 eqn(13) 10 178.616 172.799
where n is the number of measurement points.
By using LLSM to find the values of A1, A2 to obtain a Table 1 is a comparison of the downlink pathloss for both
lower value of E (A1, A2). the empirical model and that of the modified or proposed
∂E ∂E model. Another table is drawn for that of the uplink pathloss
≈ 0 and ≈0 eqn (14)
∂A1 ∂A2 values and is captured in table 2.
By solving equation (14) the result obtained is : Table 2 Results from the simulation comparing the
(∑𝑛𝑖=1 𝑙𝑜𝑔10 𝑑𝑖 )𝐴1 + (∑𝑛𝑖=1(𝑙𝑜𝑔10 𝑑𝑖 )2 )𝐴2 = ∑𝑛𝑖=1 ∆ 𝐿𝑖 𝑙𝑜𝑔10 𝑑𝑖 pathloss values for the UL
eqn (15) Distance Predicted Modified model
(∑𝑛𝑖=1(1))𝐴1 + (∑𝑛𝑖=1(𝑙𝑜𝑔10 𝑑𝑖 ))𝐴2 = ∑𝑛𝑖=1 ∆ 𝐿𝑖 eqn (16) (km) pathloss (dB) (dB) UL
UL
From equation (15) and equation (16) result a matrix
equation (17): 1 144.082 137.338

𝐴 2 154.236 147.77
( 1) =
𝐴2 3 160.18 153.878
−1
∑𝑛 log 𝑑 ∑𝑛𝑖=1(log10 𝑑𝑖 )2 ∑𝑛 ∆𝐿 log10 𝑑𝑖
( 𝑖=1𝑛 10 𝑖 ) ( 𝑖=1 𝑛 𝑖 ) 4 164.399 158.212
∑𝑖=1(1) ∑𝑛𝑖−1 log10 𝑑𝑖 ∑𝑖=1 ∆𝐿𝑖
5 167.671 161.575
eqn (17)
6 170.345 164.322
This is then modelled alongside the empirical model.
7 172.606 166.645

III. RESULTS AND ANALYSIS 8 174.564 168.657

This section covers the data analysis, discussion and the 9 176.291 170.432
presentation of results obtained from the empirical Okumura- 10 177.837 172.019
Hata model and the modified Okumura-Hata model
proposed. Mainly by considering both the DL and UL of a
radio propagated transmission. The model was modified by A. DISCUSSION
using the linear least squares method. The empirical model
was simulated by setting the height of the eNB to 50m and The Okumura-Hata model is a radio propagation model,
the height of the UE to 1m, the distances between the eNB which is used in predicting the pathloss of a cellular
and the UEs was also varied from 1km to 10 km. transmission in exterior environments. It was built by using
data collected from the city of Tokyo, Japan. This model is
Table 1 Results from the simulation comparing the best used in cities with many urban structures but not tall
pathloss values for the DL. blocking structures. Accra has a similar environment in that
regard with Tokyo, hence the selection of this model. Despite
Distance Predicted Pathloss Modified
these similarities, they are both very different when it comes
(dB) (DL) Pathloss (dB)
(Km) to the nature and weather of each city. Therefore, by
(DL)
modifying the model that is, considering these differences
better predicts the pathloss of the cellular transmission.
1 144.862 138.118 Figures 3 and 4 shows that, the predicted the pathloss values
are significantly higher than those of the modified pathloss
2 155.015 148.55 equation which considers the environment for which it may
be implemented and therefore improving the network
3 160.959 154.657 significantly by reducing the pathlosses for both the UL and
DL. For each distance,(i.e from UE to the eNB) the graph
4 165.178 158.992 shows how significantly the pathloss in the empirical model
have been improved by the modified or proposed model. This

Authorized licensed use limited to: Universiti Teknikal Malaysia Melaka-UTEM. Downloaded on June 24,2024 at 11:08:41 UTC from IEEE Xplore. Restrictions apply.
is because A1 and A2 from equation (16) takes care of the Hata, which was the empirical propagation model. A new
terrain environment mismatch between the predicted pathloss model was then proposed using the Okumura-Hata model and
and the supposed pathloss. This makes it possible to deploy adapted to the city of Accra, Ghana using environmental
an LTE network with less pathloss in an Urban city like conditions peculiar to Accra, Ghana. Both models reviewed
Accra, which meets the requirements of an Urban Large city were examined and simulations were carried out at a carrier
but with few high-rise structures, as compared to cities in frequency of 2.5GHz, antenna gain of 18dB, height of the
Europe, Asia and America. eNB at 50m and height of the UE at 1m and by varying
distance from 1km to 10km. By using least squares method, a
proposed optimization method of a pathloss model was
proposed. This proposed model has proven to be a better way
of predicting the pathloss for the deployment of 4G network
in Accra. By using the proposed model a 4G network can be
deployed or expanded with less propagation losses. The
empirical model has a standard deviation of 10.195 as
compared to 10.475 of the modified/proposed model. This
can be interpreted as the modified model having more
grouped data set.
1 REFERENCES

[1] A. Goldsmith, Wireless communications, Cambridge University Press,

2005.

[2] H. Friis, “A Note on a Simple Transmission Formula,” p. 254–256,

1946.

[3] Y. Bouguen, E. Hardouin, A. Maloberti and F. Wolff, LTE et les

réseaux 4G. EYROLLES, 2012.
Figure 3 Graphical representation of the simulated results
(DL) [4] International telecommunications union, “Radio regulation,” , Geneva,
1971.
From fig 3, the graph shows that by comparing the DL of
both models the modified model has a more improved [5] A. B. Carleial, “Correspondence: a case where interference does not
propagation pathloss than that of the empirical model. This is reduce capacity,” IEEE transactions on information theory , 1975.
also true for the UL propagation pathloss as seen in figure 4
below. [6] IEEE 802.16m, “System Requirements Document,” 2010.

[7] N. Himayat, S. T. (. corporation), A. Rav and R. S. Alcatel-lucen,

“Interference management for 4g cellular standards;,” IEEE
communications magazine, August 2010..

[8] “GPL free software,” 29 June 2007. [Online]. Available:

http://www.gnu.org/copyleft/gpl.html, Version 3, . [Accessed 15 July
2017].

[9] B. Station and E. Spectrum, “LTE (FDD) Transmitter characterstics,”

pp. 3-7.

[10] A. Satpathy, “LTE in wireless,” 02 June 2011. [Online]. Available:

http://lteinwireless.blogspot.com/2012/12/bearers-in-lte.html.
[Accessed 12 August 2017].

[11] S. A. Mawjoud, “Comparison of Propagation Model Accuracy for LTE

Cellular Network,” International Journal for Computer Applications,
vol. 79, 2013.

[12] S. Sesia, I. Toufik and M. Baker, LTE The UMTS Long Term
Evolution, FROM THEORY TO PRACTICE., : John Wiley & Sons
Ltd, 2011.
Figure II-1 Graphical representation of the simulated results
(UL)

IV. CONCLUSION
In this paper, research was carried out on managing
interference in 4G networks by considering the Okumura-

Authorized licensed use limited to: Universiti Teknikal Malaysia Melaka-UTEM. Downloaded on June 24,2024 at 11:08:41 UTC from IEEE Xplore. Restrictions apply.