International Conference on Information and Communication Technology and Its Applications
(ICTA 2016)
Federal University of Technology, Minna, Nigeria
November 28 – 30, 2016
Analysis on Energy Efficient Traffic Load Balancing in Downlink LTE-Advanced
Heterogeneous Network
A .K. Danburam1, A. D. Usman1, S. M. Sani1, and M. A. Gadam2
1
Department of Electrical and Computer Engineering, Ahmadu Bello University, Zaria Nigeria
2
Department of Electrical and Electronic Engineering, Federal Polytechnic Bauchi, Nigeria
danburamayuba@gmail.com, aliyuusman1@gmail.com, smsani@abu.edu.ng, agmohammed@fbtb.edu.ng,
Abstract—In this paper, a comprehensive analysis of energy (COMP) and Support for multi-tier deployment also known
efficiency for traffic load balancing using cell range expansion as Heterogeneous Network (HetNet) [5]. A network with a
(CRE) for Pico cells is presented. The study focused on composition of MeNB and low-power nodes (femto, pico,
evaluating the energy efficiency for traffic load balance of micro and relay nodes), mixed access modes, and backhaul is
Heterogeneous Network (HetNet) deployment scenario. Energy referred to as HetNet [6]. Intelligent HetNet deployment and
efficiency was modeled as ratio of total throughput to power planning strategies is one of ways to improve the energy
consumption, thus power consumption is evaluated using base efficiency in a mobile network [7]. Using high density
station power consumption parameters. Throughput is deployment of low power small base stations compared to
modeled based on the Signal Interference and Noise Ratio
low density deployment of high power macro base stations,
(SINR) link adaption, considering spatial distribution of User
Equipment (UE). Simulations were carried out using 3rd
has proven to decrease the power consumption. The fact
Generation partnership (3GPP) Long Term evolution (LTE) being that a base station hereafter referred to as eNodeB
system level simulator. The result obtained have shown that, (eNB), closer to mobile users lowers the required transmit
for some traffic situations, the energy efficiency improves with power due to advantageous path loss conditions [8]. Network
balanced traffic load which further provided more insight for deployment based on smaller cells such as Micro, Pico or
successful deployment of green heterogeneous cellular even Femto cells is a possible solution to reduce total power
network. consumption of a cellular network [9]. Heterogeneous
networks (HetNets) using Long Term Evolution (LTE)-
Keywords—Heterogeneous Networkt; Pico Cell Range Advanced system in 3GPP, achieve an overlay low power
Expansion; Energy Efficiency; Traffic Load Balance eNB onto high power macro eNB coverage using spectrum
reuse of one. HetNets are being increasingly deployed by
operators with macro-pico deployment as the most preferred
I. INTRODUCTION deployment strategy [10]. In a typical macro-pico
The number of mobile subscribers is greatly increasing deployment scenario, Pico eNB (PeNB) with smaller
over the years. Currently there are over 7 billion mobile transmission power and size compared to Macro eNB
cellular telephone subscribers and over 3 billion active (MeNB) are deployed within the coverage area of a MeNB
mobile broadband subscribers in the world [1]. The ITU-R to increase capacity. Another benefit of deploying PeNB is to
report anticipated that the mobile data traffic will increase reduce coverage holes, where radio signal strength from
tremendously in all countries and areas in the world. MeNB is low that mobile stations, referred to as User
Attractive mobile broadband services and improved device Equipment (UEs) are not served by MeNB [10].
capabilities drive the strong increase in unprecedented traffic However, HetNet deployment brings about new
volumes and consumer data rate [2]. From Sample cases the challenges; due to diverse transmit power levels of different
mobile data traffic revenues are not commensurate to the eNBs in HetNet [11]. Most UEs prefer to associate with the
actual traffic growth. For traffic growth of 350%, the total highest power eNB, when the conventional Reference Signal
data revenues increased only by 30% [3]. The mobile Received Power (RSRP)-based association scheme is
network operators spend about 25% of the total network employed [6]. This shifts the handover boundary between
operation cost for electric energy which is largely generated MeNB and PeNB closer to PeNB as depicted in Figure 1.
from fossil fuel [4]. Since Traffic grows faster than revenue, This result in uneven distribution of traffic load among
networks must become more efficient. different eNBs and in turn underutilization of the resource at
The LTE-Advanced system with advanced technologies low power PeNBs [11]. The 3GPP as part of it
was meant to cost effectively address the increasing demand standardization effort proposed the Biased Reference Signal
for quality of service (QoS), high data rates, and coverage Received Power (BRSRP) based user association, to
extension to mobile users. These advanced technologies proactively offload users to smaller cells using an association
include; carrier aggregation (CA), Advanced MIMO bias [12]. BRSRP-based association also known as Pico Cell
techniques, coordinated multipoint transmission/reception Range Expansion (PCRE) is a potential technique to solve
191
International Conference on Information and Communication Technology and Its Applications (ICTA 2016)
the problem of traffic imbalance [11]. In such a technique, an Micro cells and Pico cells per Macro BS will result in sub-
arbitrary fixed bias is added to the received signal power optimal of Area Energy Efficiency (AEE). In [17] a heuristic
from low-power small cell PeNBs that helps offloading more algorithm for eNB selection was proposed. The algorithm
users from MeNBs to PeNBs. The value of the bias can be maximizes energy efficiency by reducing the energy
configured individually per cell Thus, setting bias greater consumption in LTE HetNet without compromising the QoS
than 0 for the PeNB and bias of 0 for the macro MeNB will of UEs, defined as minimum data rate. In [18] a path loss-
results to PCRE [12]. This will therefore shift the handover based eNB selection procedure to realize CRE was proposed.
boundary to the MeNB as depicted in Figure 1. The algorithm associates UEs to eNBs with the lowest path
loss.
Other works focus on biased receive power based user
associations as PCRE technique. In [12] it was indicated that
MeNB RSRP PeNB RSRP
PCRE bias values have to be carefully set to achieve optimal
load-aware performance. The global optimal solutions for
dynamically selecting optimized bias was proposed in [19]
Handover Boundary and [12], and it was observed that there is a gap between an
Handover
Boundary + CRE
optimized but static PCRE and the globally optimal solution.
Bias Static PCRE has the advantage of offering much lower
complexity and overhead (both computational and
MeNB CRE Bias messaging) than optimizing the PCRE for each network
PeNB realization. The effects of PCRE on energy efficiency was
Figure 1. BRSRP based Association Scheme. investigated in [20]. This paper intends to investigate how
PCRE affect energy efficient and traffic load balancing in
PCRE bias value does not virtually enlarge the configuration 1 of the 3GPP HetNet deployment scenario.
transmission power from PeNBs, but makes User Equipment
(UE) do handover earlier to the PeNBs since they have a III. SYSTEM MODELS, SCENARIO DESCRIPTION AND
positive PCRE bias value [13]. The coverage area is not SIMULATION ASSUMPTIONS
affected by load imbalance in the uplink because the UE
possesses equal transmit power [6]. PCRE provides
A. System Models
significant improvement for UEs in the uplink as a result of
reduce path loss since the link distance are reduced [14]. The system performance evaluation of PCRE
However, in the downlink transmission, pico cell-edge UEs technique was carried out using a multi-cell system level
are exposed to severe interference from MeNB for two simulation according to LTE specifications as defined in [21]
reason: first the cell-edge UEs are furthest away from the and [22]. The investigated scenario is HetNet configuration
serving PeNB. Secondly, this UEs are much closer to the 1. Table I gives the summary and definitions of the RSRP
interfering MeNB which consequently reduce their rate. and PCRE association scheme and other variables which is
Hence PCRE for pico cells lead to uplink downlink traffic considered in this paper.
imbalance [14]. This reduce the overall throughput The conventional RSRP cell association was modeled as:
consequently reducing the downlink transmission energy
efficiency of the network. = max { , } (1)
In this paper, a comprehensive analysis of the impact
PCRE on transmission energy efficiency and traffic load Whereas the PCRE was modelled as
balance in LTE-Advanced HetNet is presented. The
objective of this paper is to evaluate the transmission energy = max { , + } (2)
efficiency, average UE throughput, and pico UE proportions
of different PCRE bias values. In order to demonstrate the For this work, single antenna receivers and transmitters
impact of PCRE association in LTE-Advanced HetNet. are assumed, and therefore, only large-scale parameters are
considered in the channel model according to [22].
II. RELATED LITERATURE 𝑃𝑅𝑋 − 𝑃𝑇𝑋 = 𝐿𝑃 + 𝐹𝑆 + 𝐺𝐴 + 𝐿𝑚𝑖𝑠𝑐 (3)
The work in [15] Investigates the impact of deploying
different numbers of small nodes on reducing area power The downlink Signal to Interference and Noise Ratio
consumption, or alternatively, on enhancing the throughput between any serving eNB and any UE is given in [7] as:
power consumption of access networks. In [16] a power
consumption model for LTE and LTE-Advanced macro cell SINR (uid,d) = 𝑃𝑇𝑋 + 𝐺𝐴 – N - I– 𝐹𝑆(d)– 𝐿𝑃(d)– PLN, (4)
and femto cell eNB was proposed and a suitable energy
efficiency measure was developed, to compare the design of Where: N and I are the noise and the inter-cell-interference
LTE to energy efficient LTE-Advanced Networks. The work (ICI) power from all the interfering eNBs at the UE location
in [7] presented a theoretical modeling of energy efficiency respectively. PLN is the wall penetration loss for signals
in Heterogeneous networks (HetNets). Simulation result received by indoor UE. Finally PL (d) and (d) are the path
shows that the pico cell strongly impacts the energy loss and shadow loss in dB respectively measured at
efficiency of the HetNet as compared to micro cell. More different UE positions. The Shanon approximation formula
specifically the work demonstrated that certain ratios of for the spectral efficiency was modeled according to [23] as
192
International Conference on Information and Communication Technology and Its Applications (ICTA 2016)
Assuming static power consumption irrespective of
traffic load situations, the base station power consumption is
defined as in [7] as:
TABLE I. SUMMARY OF VARIABLES
= Nsec*Nant (Ai* 𝑃𝑇𝑋 + Bi) (10)
Variables Definitions
CellID Cell where UE receive maximum RSRP Where Nsec and Nant denote the eNBs’s number of
, RSRP from MeNB and PeNB respectively sectors and the number of antennas per sector, respectively.
CRE bias value for PeNB Pci is the average total power of base station(s) in a cell and
SINR efficiency 𝑃𝑇𝑋 is the power fed to the antenna as defined in equation
SINR value corresponding to the 26 MCSs level. (3). The coefficient Ai accounts for the part of the power
consumption that is proportional to the transmitted power,
𝑃𝑅𝑋 UE received power
which include Radio Frequency (RF) amplifier power and
𝑃𝑇𝑋 eNB transmit power
feeder losses. While Bi denotes the power that is consumed
𝐿𝑃 Path loss
independent of the average transmit power which include
𝐹𝑆 Fading due to shadowing
signal processing and site cooling [7]. The value of the
𝐺𝐴 Directional antenna gain
parameters are specified in table II.
𝐿𝑚𝑖𝑠𝑐 Any miscellaneous loss such as feeder cable loss
The energy efficiency is defined as the ratio of the total
Bandwidth efficiency throughput (R) within a cell and the total power consumption
of the cell (PCi), which is expressed as [7]:
Link adaptation was used to map SINR to corresponding = = (11)
Transmission Block Size (TBS). Link adaptation requires the
selection of a proper Modulation and Coding Scheme (MCS)
according to the channel quality which is usually indicated Where: RCi is the overall throughput in bits/s within a
by the SINR reported by each UE. Following the LTE cell, and PCi is the total power consumption of the cell in
specification in [21], three modulation levels of Quadrature watts and is the transmission energy efficiency to all
Phase Shift Keying (QPSK), 16-QAM and 64-QAM are UEs in bits/joule within the cell.
supported. Together with turbo coding there are 26 MCSs
levels, this imply that there are 26 Channel Quality B. Scenario Description and Simulation Assumptions
Indicators (CQI). The SINR to the effective SINR ( ) Based on the 3GPP LTE system level simulations
mapping was modeled as: toolbox define in [25], a system of 7 wraparound sectored
MeNB (21 cells) with 4 PeNB per sector is considered in this
= max { , } work. The PeNBs are randomly drop within a MeNB area
(6) with minimum inter-site distance constrains. Each sector has
a directional antennas at 120 degrees apart one for each
Is the SINR as a result of the UE’s sector, while the PeNB has Omni-directional antenna. Users
instantaneous channel conditions as in equation (4). The are uniformly distributed throughout the coverage area
mapping of SINR to TBS of the 26 MCSs levels, assuming a following the HetNet configuration 1 topology. Mobility is
Block Error Rate (BLER) target of 10% according to [24] represented by users having different locations in each drop.
was modelled as: Other related system level simulation parameters are
specified in Table II.
TBS(uid,d) = TBS( ) (7) The performance evaluation was carried out in a 3GPP
LTE system level simulator. The following metrics was used
Throughput (R) for a UE i is given in [23] as: for performance evaluation:
PeNB UEs (PUE) proportion: Number of UEs
connected to PeNBs.
(8)
Average cell energy efficiency: energy efficiency
averaged over all simulated cells from all simulation
Where is The physical transmission block drops.
information capacity (in bits) for the each UE CQI state I, Cell average PUE and MeNB UE (MUE)
and is the average BLER, TTI is the transmission throughput: average UE throughput will indicate
time interval and is the number of resource block how well the traffic load is balanced between PeNBs
allocated to UE i. In this paper round robin resource and MeNB [25].
scheduler is considered which is modeled as:
IV. RESULTS AND DISCUSSION
NRB(uid,d) = (9) In this section, the overall simulation results for the
conventional RSRP cell association scheme and PCRE
association schemes with different bias considered in this
Where: NRB(uid,d) is the number of resource block work is presented. The simulation was carried out for
allocated to a user at distance d from an eNB. different number of UEs in the HetNet configuration 1. The
193
International Conference on Information and Communication Technology and Its Applications (ICTA 2016)
proportions of UEs connected to the PeNBs increased with CDF of spectral efficiency (SE) is presented in Fig. 2, Fig. 3
the increase in PCRE bias due to the offloading of more UEs and Fig. 4 respectively.
from MeNB to PeNBs as a result of the effect of PCRE bias.
100
The proportion of UEs connected to PeNB for PCRE with
bias of 3dB, 6dB, 9dB, 12dB and 16dB were found to be 90
about 7% 9% 15% 20% and 26% higher than the
80
conventional RSRP cell association scheme respectively.
Pico UE Proportion *100 [%]
70
TABLE II. SYSTEM LEVEL SIMULATION PARAMETERS 60
Parameter Setting/Description 50
Cell layout 7 Hexagonal MeNBs; 3
40
sectors; reuse 1
MeNBs radius 500m 30 Conventional RSRP
Bandwidth and carrier frequency 10MHz at 2000 MHz RSRP with CRE = 3dB
Hotspot radius 4 20 RSRP with CRE = 6dB
RSRP with CRE = 9dB
Hotspot radius 40m 10 RSRP with CRE = 12dB
Minimum distance MeNBs and 75m
RSRP with CRE = 16dB
between PeNBs 0
10 20 30 40 50 60 70 80 90 100
Among PeNBs 40m Number of UEs per Cell
MeNBs and UE 35m
PeNBs and UEs 10m Figure 2. PeNB UE Proportion
Transmission power MeNBs 46 dBm
100
PeNBs 30 dBm Conventional RSRP
Path-loss MeNBs 128.1 +37.6log10 (r [km]) 90 RSRP with CRE = 3dB
[21] RSRP with CRE = 6dB
PeNBs 140.7 +36.7log10 (r [Km]) 80 RSRP with CRE = 9dB
RSRP with CRE = 12dB
[21] 70 RSRP with CRE = 16dB
Number of UEs per 10,20,...,100
CDF (percentage)
sector 60
UE distribution Uniform distribution [21]
50
Packet scheduler Round Robin
, 0.75 , 1.25 [22] 40
Power consumption Macro:Ai = 21.45; Bi = 30
parameters 354.44, Pico: Ai = 5.5;
Bi=38[6] 20
10
For the individual bias values, the proportion of PeNB
0
UEs increase up to 30UEs in the system, but allowing up to -40 -30 -20 -10 0 10 20 30
40 UEs into the system, however, caused a decrease in the SINR [dB]
connection ratio. It subsequently stabilized when more UEs Figure 3. The CDF of SINR with Different PCRE Bias
were allowed into the system beyond 40 UEs. Therefore, it
can be deduced that the best offloading gain for all the bias
100
values is achieved when 30UEs are allowed in the system.
Nevertheless, the connection ratio does not show significant 90
difference with the rest of number of UEs for all the bias 80
values. This is consistent with what is reported in [26].
The cumulative distribution functions (CDF) of the SINR 70
CDF (percentage)
of PCRE cell association schemes with 4 PeNBs and 100 60
UEs per sector lie above the SINR CDF of conventional
50
RSRP as the reference cell association scheme. The worst
affected UE by interference in all the cell association 40
schemes are the cell edge (worst 5%) UEs [26]. 30
Conventional RSRP
Essentially, any offloading due to increase in PeNB cell RSRP with CRE = 3dB
RSRP with CRE = 6dB
range will result in SINR performance degradation of the 20
RSRP with CRE = 9dB
offloaded UEs [27]. This is due to the interference effect 10 RSRP with CRE = 12dB
suffered by pico cell-edge UEs from the high transmission RSRP with CRE = 16dB
0
power of MeNBs. Consequently, the SINR CDF for the cell 0 0.5 1 1.5 2 2.5 3 3.5 4
edge UEs of the PCRE with 16dB, was found to be the worse Spectral efficiency [bps/Hz]
followed by 12dB, 9dB, 6dB than the SINR CDF of the Figure 4. The CDF of SE with Different PCRE Bias
conventional RSRP respectively. PCRE with 3dB did not
show significant difference with the conventional RSRP. The spectral efficiency (SE) is the measure of utilization
This shows that without effective interference mitigation the of bandwidth measured in bps/Hz, the corresponding
cell edge UEs will be in an outage, with large PCRE bias performance for the conventional RSRP and PCRE is
values. The pico UE connection ratio, CDF of the SINR and depicted in Fig. 4. The average (50% CDF) SE was not
194
International Conference on Information and Communication Technology and Its Applications (ICTA 2016)
found to differ between the conventional RSRP and the with 6dB has the lowest difference in the average UE
RSRP with 3dB, 6dB and 9dB. But that of PCRE with 12dB throughput between the MeNB UEs and PeNB UEs.
and 16dB lie slightly above the conventional RSRP for less PCRE with bias of 9dB exhibited a more balanced
than 70% CDF after which it was not found to differ. The SE average UE throughput performance for high traffic load
of the cell edge (worst 5%) UEs for the conventional RSRP (10UE per cell). For high traffic load, the difference between
and all the PCRE bias were poor due to the poor load the average throughput performance of the PeNB UEs and
balancing in the case of the conventional RSRP and poor MeNB UEs is 1.494Mbps, 1.001Mbps, 0.42Mbps,
SINR in the PCRE scheme. 0.13Mbps, 0.83Mbps and 1.81Mbps for conventional RSRP,
RSRP with bias of 3dB exhibited a more balanced PCRE with 3dB, 6dB, 9dB, 12dB and 16dB respectively.
average UE throughput performance for low traffic load Hence, PCRE with 9dB has the lowest difference in the
(10UE per cell). For low traffic load, the difference between average UE throughput between the MeNB UEs and PeNB
the average throughput performance of the PeNB UEs and UEs as depicted in Fig. 7.
MeNB UEs is 10.8Mbps, 1.1Mbps, 1.6Mbps, 8.4Mbps, For all the traffic load considered the average PUE
10.3Mbps and 15.2Mbps for conventional RSRP, PCRE with throughput decrease with increase in bias. This can be
3dB, 6dB, 9dB, 12dB and 16dB respectively. Hence, PCRE attributed to the fact that PCRE essentially offloads UE from
with 3dB has the lowest difference in the average UE MeNB to PeNB, the higher the PCRE bias the more the
throughput between the MeNB UEs and PeNB UEs. offloading gain. Therefore, the higher PCRE bias resulted to
PCRE with bias of 6dB exhibited a more balanced overcrowding the PeNB thereby lowering the average
average UE throughput performance for medium traffic load throughput of the PUEs due the round robin scheduler
(50UE per cell). Fig. 5 and Fig. 6 shows the average UE employed. The round robin resource allocation makes UEs to
throughput for low and medium traffic load respectively. share the limited resource blocks in the pico cell equally.
Also as the PCRE bias increase pico cell-egde UEs increase,
35
Conventional RSRP
such UEs are greatly impacted by interference from MeNB
RSRP with CRE = 3dB which consequently reduce their rate. Conversely, the
30 RSRP with CRE = 6dB average MUEs throughput increase with increase in bias.
25.6
RSRP with CRE = 9dB
This can be attributed to the fact that, as UEs are offloaded to
Average UE Throughput [Mbps]
RSRP with CRE = 12dB
25
RSRP with CRE = 16dB PeNBs from MeNB, fewer UEs are left in the MeNB to
22
20.8
20.3
share the available resources and such UEs are not affected
20
by interference. Therefore, such UEs achieve high data rate
16.3
15.8
which is similar with what is reported in [27] and [28].
14.7
14.7
13.1
12.8
12.8
15
12.4
12.3
12.3
11.7
11.5
10.4
35
9.5
Conventional RSRP
10
RSRP with CRE = 3dB
30 RSRP with CRE = 6dB
5 RSRP with CRE = 9dB
Average UE Throughput [Mbps]
RSRP with CRE = 12dB
25
RSRP with CRE = 16dB
0
PUE Throughput MUE Throughput All UE Throughput 20
Figure 5. Average UE Throughput for Low Traffic Load
15
35
Conventional RSRP 10
RSRP with CRE = 3dB
30 RSRP with CRE = 6dB
2.86
5
2.14
1.82
1.52
0.819
1.45
1.46
0.646
1.41
1.36
1.32
1.33
1.32
1.32
1.17
1.05
RSRP with CRE = 9dB
1.1
2
Average UE Throughput [Mbps]
RSRP with CRE = 12dB
25 0
RSRP with CRE = 16dB PUE Throughput MUE Throughput All UE Throughput
Figure 7. Average UE Throughput for High Traffic Load
20
The PCRE with 16dB bias achieve the worst average
15
UE throughput (All UE throughput) and traffic load balance
9.79
for all the traffic load considered. This can be attributed to
10
poor SINR performance with 16dB and redundancy
4.83
introduced to the MeNB due to limited UEs allowed in the
4.14
3.36
3.7
2.92
2.98
2.82
2.77
2.72
2.74
2.77
2.59
5
2.55
MeNB. It can also be observed that the conventional RSRP
2.35
2.09
1.92
1.41
achieved the best total UE throughput. This is because the
0 conventional RSRP has the best SINR performance.
PUE Throughput MUE Throughput All UE Throughput
However, the conventional RSRP achieve a poor traffic load
Figure 6. Average UE Throughput for Medium Traffic Load
balance. This is due to low offloading of UEs from PeNB to
For medium traffic load, the difference between the MeNB.
average throughput performance of the PeNB UEs and Despite the poor performance of the conventional RSRP
MeNB UEs is 2.73Mbps, 1.44Mbps, 0.33Mbps, 1.15Mbps, in terms of traffic load balance, it was found to perform
2.4Mbps and 7.7Mbps for conventional RSRP, PCRE with better in terms of energy efficiency. The conventional RSRP
3dB, 6dB, 9dB, 12dB and 16dB respectively. Hence, PCRE achieved the best energy efficiency for all traffic load
195
International Conference on Information and Communication Technology and Its Applications (ICTA 2016)
simulated as depicted in Fig. 8. PCRE with 16dB achieved [5] 3GPP, “Evolved Universal Terrestrial Radio Access ( EUTRA );
the worst energy efficiency due to poor SINR performance Radio Frequency ( RF ) requirements for LTE Pico Node B, TR
36.931,” 2011
which lowers the total achievable throughput.
[6] M. A. Gadam, M. A. Ahmed, C. K. Ng, N. K. Nordin, A. Sali, and F.
Hashim, "Review of Adaptive Cell Selection Techniques in LTE-
120 Advanced Heterogeneous Networks," Journal of Computer Networks
and Communications, vol. 2016, 2016.
100 [7] A. A. Abdulkafi, S. K. Tiong, D. Chieng, A. Ting, A. M. Ghaleb, and
J. Koh, “Modeling of Energy Efficiency in Heterogeneous Network,”
Eng. Technol., vol. 6, no. 17, pp. 3193–3201, 2013.
Average EE [Kb/joules]
80 [8] C. Desset, B. Debaillie, V. Giannini, A. Fehske, G. Auer, H.
Holtkamp, W. Wajda, D. Sabella, F. Richter, M. J. Gonzalez, H.
Klessig, I. Gódor, M. Olsson, M. A. Imran, A. Ambrosy, and O.
60 Blume, “Flexible power modeling of LTE base stations,” IEEE Wirel.
Commun. Netw. Conf. WCNC, pp. 2858–2862, 2012.
[9] X. Ge, "Energy Efficiency Modelling and Analyzing Based on Multi-
40 Conventional RSRP cell and Multi-antenna Cellular Networks," KSII Transactions on
RSRP with CRE = 3dB Internet and Information Systems, 2010.
RSRP with CRE = 6dB
20 [10] S. Konishi, “Comprehensive analysis of heterogeneous networks with
RSRP with CRE = 9dB
RSRP with CRE = 12dB
pico cells in LTE-advanced systems,” IEICE Trans. Commun., vol.
RSRP with CRE = 16dB
E96-B, no. 6, pp. 1243–1255, 2013.
0 [11] T. Zhou, Y. Huang, W. Huang, S. Li, Y. Sun, and L. Yang, "QoS-
10 20 30 40 50 60 70 80 90 100
Number of UEs per Cell aware user association for load balancing in heterogeneous cellular
networks," in 2014 IEEE 80th Vehicular Technology Conference
Figure 7. Average Energy Efficiency (VTC2014-Fall), 2014, pp. 1-5.
[12] J. G. Andrews, S. Singh, and X. Lin, “And Beyond Cellular Networks
An Overview of Load Balancing in HetNets : Old Myths and Open
V. CONCLUSION Problems. IEEE Wireless Communications April, pp. 18–25, 2014.
[13] K. Kitagawa, T. Yamamoto, and S. Konishi, “E ff ect of Cell Range
HetNet deployment have the potential to improve Expansion to Handover Performance for Heterogeneous Networks in
capacity as well as energy efficiency. However poor cell LTE-Advanced Systems,” no. 6, pp. 1367–1376, 2013.
association and poor HetNet deployment limit the potential [14] K. Davaslioglu,. Energy Efficiency and Load Balancing in Next-
of HetNet in improving energy efficiency and traffic load Generation Wireless Cellular Networks, PhD Dissertation.
balance. Therefore, in this work, a comprehensive analysis Department of Electrical and Computer Engineering, Faculty of
Engineering, University of California, Irvine, 2015.
on the effect of RSRP and Pico Cell Range Expansion (CRE)
association scheme on energy efficiency and traffic load [15] A. B. Saleh, Ö. Bulakci, S. Redana, B. Raaf, and J. Hämäläinen,
“Evaluating the energy efficiency of LTE-Advanced relay and
balance is presented. The modelling of the energy efficiency Picocell deployments,” IEEE Wireless Communication Network
was based on base station power consumption and data rate. Conference. WCNC, pp. 2335–2340, 2012.
Thus the power consumption is evaluated using power [16] M. Deruyck, W. Joseph, B. Lannoo, D. Colle and L. Martens
consumption parameters and the data rate was modelled “Designing Energy-Efficient Wireless Access Networks :,” IEEE
based on link adaptation considering spatial distribution of Computing Society January 2013.
UEs. From simulation result it was found that, while RSRP [17] A. Pourmoghadas and P. G. Poonacha, “A Base Station Association
achieves the best performance, PCRE reduce energy Algorithm for Energy Reduction in LTE Heterogeneous Networks,”
Proc. of Int. Conf. on Advances in Communication, Network, and
efficiency and overall average UE throughput. However for Computing, CNC 2014.
low medium and high traffic load, PCRE with bias 3dB, 6dB [18] J. Wu, S. Jin, L. Jiang, and G. Wang, “Dynamic switching off
and 9dB achieves the best traffic load balance respectively. algorithms for pico base stations in heterogeneous cellular networks,”
RSRP with bias of 16dB and conventional RSRP achieved EURASIP Journal onWireless Communications and Networking
poor traffic load balance. Therefore, PCRE with bias of 3dB 2015.
to 9dB can achieve a tradeoff between traffic load balance [19] Q. Ye, B. Rong, Y. Chen, M. Al-Shalash, C. Caramanis, and J. G.
and energy efficiency. Further work should look at achieving Andrews, "User Association for Load Balancing in Heterogeneous
Cellular Networks," IEEE Transactions on Wireless
a tradeoff between traffic load balance and energy efficiency Communications, vol. 12, pp. 2706-2716, 2013.
by jointly optimizing the two metrics.
[20] Y. F. Lou, W. Guo, and W. H. Xiong, "Energy Efficiency of Cell
Range Extension of Picocell," Applied Mechanics and Materials, vol.
340, pp. 507-511, 2013.
REFERENCES [21] 3GPP, “Evolved Universal Terrestrial Radio Access ( E-UTRA );
[1] C. V. N. Index. Forecast and methodology, 2014-2019 white Radio Frequency ( RF ) requirements for LTE Pico Node B, TR
paper.Retrieved 23rd September.. 2015 36.931,” 2011.
[2] M. A. Joud,. Pico Cell Range Expansion toward LTE-Advanced, [22] A. Abubakar, T. Mantoro, S. Moedjiono, H. Chiroma, A. Waqas, “A
M.Sc. Thesis. Department of information and communication Support Vector Machine Classification of Computational Capabilities
technologies, Universitat Politècnica de Catalunya (UPC) 2013. of 3D Map on Mobile Device for Navigation Aid”, International
[3] Nokia Siemens Networks, “Mobile broadband with HSPA and LTE – Journal Of Interactive Mobile Technologies, 2016.
capacity and cost aspects,” p. 12, 2010. [23] P. Mogensen, W. Na, I. Z. Kováes, F. Frederiksen, A. Pokhariyal, K.
[4] S. K. Khadka, J. Shrestha, S. R. Shakya, and L. Lal, "Energy demand I. Pedersen, T. Kolding, K. Hugl, and M. Kuusela, “LTE capacity
analysis of telecom towers of Nepal with strategic scenario compared to the shannon bound,” IEEE Veh. Technol. Conf., no. 1,
development and potential energy cum cost saving with renewable pp. 1234–1238, 2007.
energy technology options," International Journal of Research in [24] C. Khirallah, D. Rastovac, D. Vukobratovic, and J. Thompson,
Engineering and Science (IJRES), vol. 3, pp. 01-08, 2015. "Energy Efficient Multimedia Delivery Services over LTE/LTE-A,"
196
International Conference on Information and Communication Technology and Its Applications (ICTA 2016)
IEICE Transactions on Communications, vol. 97, pp. 1504-1513, [27] M. A. Gadam, A. A Maryam, N. K. Nordin, A. B. Sali, and F.
2014. Hisyam, “ Decentralized Time Domain Muting for Interference
[25] M. A Gadam, N. g Chee Kyun, N. K. Nordin, A. Sali, and F. Hashim, Mitigation In LTE-Advanced Heterogeneous Network,” in 2015
“Hybrid Channel Gain and Access Cell Association for Load IEEE Conference on Sustainable Utilization and Development In
balancing in Downlink LTE-Advanced HetNets,” in 6th IEEE Engineering and Technology, (2015 IEEE CSUDET), 2015, pp. 17-
International Conference on Computer and Communication 22.
Engineering (Submitted to IEEE ICCCE 2016), 2016. [28] Y. Wang, H. Ji, and H. Zhang, “Spectrum-efficiency enhancement
[26] M. A. Gadam, C. N. Ng, N. K. Nordin, A. Sali, & F. Hashim,. Hybrid in small cell networks with biasing cell association and eICIC: An
channel gain prioritized access‐aware cell association with analytical framework,” Int. J. Commun. Syst., vol. 29, no. 2, pp. 362
interference mitigation in LTE‐Advanced HetNets. International –377, Jan. 2016.
Journal of Communication Systems. 2016.
197