Page 1 of 21
European Journal of Applied Sciences – Vol. 10, No. 2
Publication Date: April 25, 2022
DOI:10.14738/aivp.102.12244. Deussom Djomadji, E. M., Fouba, B. A. R., & Kenfack Wamba, J. G. (2022). Performance Evaluation of the Eicic Technique Applied
to a Heterogeneous 4g Mobile Network. European Journal of Applied Sciences, 10(2). 540-560.
Services for Science and Education – United Kingdom
Performance Evaluation of the eICIC Technique Applied to a
Heterogeneous 4G Mobile Network
DEUSSOM DJOMADJI Eric Michel
College of Technology, University of Buea, Buea, Cameroon
National Advanced School of Posts
Telecommunications and Information and Communication Technologies
University of Yaoundé I, Yaoundé, Cameroon
FOUBA Bienvenue Arsene Roger
National Advanced School of Posts
Telecommunications and Information and Communication Technologies
University of Yaoundé I, Yaoundé, Cameroon
KENFACK WAMBA Jean Gutenbert
National Advanced School of Posts
Telecommunications and Information and Communication Technologies
University of Yaoundé I, Yaoundé, Cameroon
ABSTRACT
The evolution of LTE-A (Long Term Evolution -Advanced) has enabled the
development of the Heterogeneous Network (HetNet) which uses multiple cell
types due to the explosion in the number of connected subscribers. Different cell
types of different sizes are deployed to share the load in the heterogeneous
network. Unplanned deployment of small cells with low power, planning of the
existing network, the power difference between cells; physical and meteorological
phenomena affect the signal quality of radio links; crowded areas of metropolises
with extremely high traffic hotspots force operators to increase the network
capacity in a more economical way by installing small cells Several techniques
have been developed in the literature. Among them, eICIC has emerged as the best
and is used to overcome these interference problems in the heterogeneous 4G
network. The performance evaluation of eICIC has been performed analytically
and numerically in the time domain due to its use with the ABS (Almost Blank
Subframe) of the OFDM system. The simulation areas are in center of the picocells,
in the cell range extension (CRE) areas and in the macrocells. Analytically, we used
the signal-to-interference-plus-noise ratio (SINR) to evaluate the bit rate, RSRP
and RSRQ. The numerical method consists in proposing an algorithm for
calculating the analytical parameters using Matlab software. The results obtained
from the simulations are convincing and are not far from those obtained in the
literature. The values obtained from the SINR correspond to the SINR-CQI table
found in the literature. The throughput calculated in this work in relation to radio
conditions (RSRP, RSRQ) reflects everyday realities.
KeyWords: LTE-A, HetNet, eICIC, ABS, CRE, OFDM, SINR, RSRP and RSRQ.
Page 2 of 21
541
Deussom Djomadji, E. M., Fouba, B. A. R., & Kenfack Wamba, J. G. (2022). Performance Evaluation of the Eicic Technique Applied to a
Heterogeneous 4g Mobile Network. European Journal of Applied Sciences, 10(2). 540-560.
URL: http://dx.doi.org/10.14738/aivp.102.12244
INTRODUCTION
The electronic communications ecosystem has been strengthened by the developments in
mobile technologies initiated by the 3GPP (Third Generation Partnership Project) and IEEE
802.x Wifi (Wireless Fidelity) standards. After 3G (third generation mobile network), 4G
(fourth generation mobile network) has made it possible to exchange data at very high speeds
in mobile networks. The number of users of these networks continues to grow, creating new
needs such as new real-time, financial, banking and other applications and services. This
makes it necessary to use new frequency bands to ensure and guarantee an acceptable quality
of service for users. Long Term Evolution (LTE) addresses these various requirements by
using an adaptive coding and modulation scheme with spectral flexibility using bandwidth
while ensuring transmission robustness to increase spectral efficiency.
To achieve the expected performance of this promising technology, the 4G mobile networks
still pose a large number of problems and challenges. Inter-cell, intra-cell and inter-channel
interference are detrimental factors in degrading throughput and quality of service for users
at the edge or within cells. While this is a classic problem in cellular networks, it is already
becoming more serious in LTE networks given the high data rates and load on the cells. It is in
this context that the concept of mobile radio interference coordination, ICIC (Inter-Cell
Interference Coordination), appears in 3GPP Releases 8 and 9. However, ICIC becomes
ineffective when the network has to be extended to a heterogeneous network. Researchers
have improved the ICIC technology and introduced an enhanced version: eICIC (enhanced
Inter-Cell Interference Coordination) presented in 3GPP releases 10 and 11. The present work
is focused on proposing a new method for evaluating and managing eICIC in mobile 4G
networks.
General context
HetNet (heterogeneous networks) is an attractive way to increase the capacity of wireless
communication systems. It is typically composed of various radio access technologies,
architectures, transmission solutions, and base stations characterized by different
transmission powers [1]. The HetNet is composed of macro-cells, pico-cells, femtocells, and
relays. The latter three constitute the small cells and are mainly used to increase the capacity
of hotspots with high user demand and to fill in areas not covered by the macro network, both
externally and internally. They also improve network performance and quality of service by
offloading large macrocells.
Table 1: Types of Cell in Hetnet [1]
Cell types Powers transmitting Coverage radius
Macro-cell 40W or 46dBm ≥500 m
Pico-cell 23 à 30 dBm < 300 m
Femto-cell < 23dBm < 50 m
Relay 30 dBm 300 m
Page 3 of 21
542
European Journal of Applied Sciences (EJAS) Vol. 10, Issue 2, April-2022
Services for Science and Education – United Kingdom
Figure 1: Example of HetNet Architecture [2]
Although considered the most attractive way to improve network capacity, HetNets face many
challenges [3]. We can list the following:
• Frequency allocation [4],
• Self-organization [5]
• Backhauling [5]
• The Handover process [6, 7]
• The Interferences
In this part, we will describe the eICIC technology and analyze its performance. The concept of
eICIC will be presented and its characteristics will be identified.
Characteristics of the eICIC concept
The eICIC has two main characteristics according to the 3GPP and this has been presented by
the author in [8] [9][10] these are:
Ø ABS (Almost Blank Subframe) technology defined in 3GPP Release 10 to prevent
neighboring macro equipment from interfering with cell edge UEs in small cells. Indeed
the two cells always use the same radio resources, but in different time intervals (sub- frames);
Ø The CRE (Cell Range Expansion) technology defined in 3GPP Release 11 which extends
the coverage of a small cell so that a greater number of UEs close to the edge of the cell
can access the small cell.
Page 4 of 21
543
Deussom Djomadji, E. M., Fouba, B. A. R., & Kenfack Wamba, J. G. (2022). Performance Evaluation of the Eicic Technique Applied to a
Heterogeneous 4g Mobile Network. European Journal of Applied Sciences, 10(2). 540-560.
URL: http://dx.doi.org/10.14738/aivp.102.12244
Figure 2: eICIC technology [11]
ABS Technology
Definition of the concept of ABS
According to [11], to avoid intercellular interference, the basic idea of eICIC is to allow a
macro-cell and its neighboring small cells to provide data, subframe by subframe, by using
different time ranges. When communicating with users at the cell edge, the small cells use the
subframes not used by their neighboring macro-cell, to avoid interfering with the users of the
macro cell.
Operation Principle
The eNodeB and user equipments (UE) continuously exchange data in radio frames. A radio
frame consists of 10 subframes. The decision on how many ABS subframes to include in a
radio frame is made based on traffic load and network operators’ policy.
Figure 3 illustrates the operation mode of the eICIC. A Macro eNodeB (M-eNB) determines
which subframes will carry data ("0") and which one will not ("1"), then saves them as ABS
model information ("0011000110"). The M-eNB prepares the charging information message
to send it to the pico eNodeB (P-eNB) through the X2 interface. Upon receipt of this message,
the P-eNB, from the template, learns which subframes are to be used by M-eNB and provides
data to cell edge UEs via ABS subframes uniquely.
Page 5 of 21
544
European Journal of Applied Sciences (EJAS) Vol. 10, Issue 2, April-2022
Services for Science and Education – United Kingdom
Figure 3: ABS principle [11]
CRE technology
Definition of the concept
The heterogeneous network consists of low power nodes such as pico-cells, femto-cells, and,
relay nodes are deployed within a macro-cell. These deployed small cells mainly allow to
increase the capacity of the hot spots where the users demand is high and also to cover the
hollow areas by the macro network, both outdoors and indoors. They help to unload large
cells in other to improve the system performance.
Principle of cell range extension
A major problem in the heterogeneous network is to ensure that the small cells really and
sufficiently serve the users. One solution is to increase the area served by the small cell, which
can be achieved through the use of an offset positive cell selection relative to the received
power of the small cell. Extending the coverage of a cell by connecting a UE to a cell weaker
than the detected cell and stronger is called cell range extension [12] [3].
The macro cell transmits at a very high power compared to the low power node. So, as shown
in the figure below, there are two bands and a cell border.
Figure 4: Illustration of the extension area [12] [3]
Page 6 of 21
545
Deussom Djomadji, E. M., Fouba, B. A. R., & Kenfack Wamba, J. G. (2022). Performance Evaluation of the Eicic Technique Applied to a
Heterogeneous 4g Mobile Network. European Journal of Applied Sciences, 10(2). 540-560.
URL: http://dx.doi.org/10.14738/aivp.102.12244
METHODOLOGY
The principle consists in deploying small cells within a macro cell where the traffic is very
dense; then, configuring the antennas by activating the eICIC technology and improving the
gain of user propagation channel in a given region.
We focus these studies on the downlink link. The users are randomly distributed in the
different analysis cell zones. Service cells are selected based on the highest received power.
The amount of radio resources allocated to each user is determined based on the size of the
data that the user is willing to transmit and receive; the traffic load in the cell and the MCS
(Modulation and Coding Scheme) is optimal. This load is determined according to the user
and the radio conditions likely to be disturbed, attenuated, and scrambled.
Analytical method
It consists in performing the calculations of the different SINRs, the user’s rate, and the
parameters of the radiofrequency conditions.
The calculation of the antenna gain will allow us to show the different SINRs (in the macrocell,
in the picocell, and in the CRE). From this, we will deduce the resulting user throughputs. The
analysis of the RSRP and RSRQ will be used for handover evaluation.
Propagation in the mobile radio environment
During propagation from the transmitter to the receiver, the electromagnetic phenomena
generate several components of the transmitted signal (multipath propagation). At the
receiving end, these components arrive at different times with given attenuations and phase
shifts.
In [13], the author modeled the propagation channel gain ��,� in the mobile environment
between the serving base station (k) and the user (i) by:
��,� = �. ����
��. ( ��
��,�
)� (1)
h represents Rayleigh multipath fading modeled by an exponential distribution. The
parameter �& is a standard Gaussian random variable, σ is the standard deviation of
shadowing in dB, ��,� is the distance between user i and Base Station k, d0 is a reference
distance equal to 500 meters and finally, α is the exponent of the path loss.
SINR calculation
The signal-to-noise-interference ratio (SINR) on a wireless link is an important basis for
considering the degradation, throughput, and spectral efficiency of a mobile network. It is,
therefore, necessary to know the SINR distribution within these networks, especially in
heterogeneous cellular networks, as these will dominate future network deployments. Until
recently, the distribution of SINRs in heterogeneous networks was studied almost exclusively
by simulation, for chosen scenarios representing predefined arrangements of users and
heterogeneous network elements such as macro cells, femtocells, etc [9].
In this document, � will refer to a user in a set I, � is the index of a macro cell, �'
is the pico cell
index, and �'' is the one of the CRE.
In our context, we define SINR for user i by :
Page 9 of 21
548
European Journal of Applied Sciences (EJAS) Vol. 10, Issue 2, April-2022
Services for Science and Education – United Kingdom
The data rate of the user (i) in CRE k’’ is :
(��_���)� ,� = �
��_��� ��� [� + O��������)�,����P ∗ (�) (20)
The equation (12) in (20), we are obtaining a variation of the ratio according to the distance:
(��_���)� ,� = �
��_��� ��� [� + �����M��,�-- . �. ����
��. B ��
��,�--C
�
K − ����� M ∆�. �� +
∑ ��,�. �. ����
��. B ��
��,�
C
� ��
�A� + ∑ ��,�- . �. ����
��. B ��
��,�--C
� ��2�
�-A� K ∗ (�)
(21)
ABS Ratio Validation
In order to maintain the balance and the load sharing between the macro cell and the pico cell
extension zone, it is necessary to share the throughput during ABS. That is, to balance
equations (17) and (21). We obtain the same maximum users speed, which will generate the
optimal ABS ratio, given by equation (22):
� = �
��
����*�
(22)
Simulation hypothesis
In the present scenario, the engineering data will be used by considering a heterogeneous LTE
network. These engineering parameters are given in table 1
Figure 5: HetNet topology
Page 10 of 21
549
Deussom Djomadji, E. M., Fouba, B. A. R., & Kenfack Wamba, J. G. (2022). Performance Evaluation of the Eicic Technique Applied to a
Heterogeneous 4g Mobile Network. European Journal of Applied Sciences, 10(2). 540-560.
URL: http://dx.doi.org/10.14738/aivp.102.12244
Table 2: simulation Parameters
PARAMETERS VALUES
system size Macro-cell= 500 m
Pico-cell = 150 m
Power transmission Macro-cell = 46 dBm
Pico-cell = 30 dBm
Gain antenna Macro-cell = 14 dB
Pico-cell = 6 dB
High of antenna Macro-cellule = 45 m
Pico-cellule = 15 m
UEs = 1,5m
Number of cell Number of macro-cell = 1
Number of pico-cell = 5
Bandwitch of system 20 MHz
System Noise -174 dBm/Hz
Frequency 1 800 MHz
Packet scheduling of eNB Round Robin
SINR maximal 20 dB
Débit maximal througput 50 Mbps
Space of sub-frame 15 KHz
Pathloss exposant 3 dB
RESULTS AND INTERPRETATIONS
The numerical simulation is on one hand based on the numerical resolution of the systems of
equations (23), (25), and (26) to obtain the SINR, RSRQ, and RSRP and on the other hand on
the use of the MATLAB functions to execute the steps of the proposed algorithm.
SINR Simulation
We are interested in three specific zones: the macro cell, the picocell, and the pico cell
extension zone (CRE).
Figure 6 shows us that the evaluated SINR (blue curve) in a pico cell decreases on a
logarithmic scale over the range of 500 meters. Indeed, the pico cell is defined on a radius of
150 meters and beyond this distance, the values become very bad. This is due to the
contribution of external powers which act on the pico users. As the range increases, the power
of the interfering signal increases, therefore the SINR decreases. The SINR values obtained by
simulation vary between -10dB to 40 dB.
As far as macro-cells are concerned, the quality of the SINR (curve in green) is very good in
our simulation conditions and the range of the SINR corresponds to that of the literature. We
have a variation between 5 dB and 45 dB obtained during the simulation. The following table
presents some variation of SINR obtained in the literature.
For users in the CRE of pico-cell the SINR (red curve) is acceptable up to 500 meters where it
is close to -5dB. This proves that there was dominance of the signal of the macro cell in the
users of the CRE, therefore there is an increase in the SINR of the users of the extension zone.
We can say that the users who are in the center of the pico cell that suffers from interference