MULTI-CRITERIA HANDOVER DECISION FOR HETEROGENEOUS NETWORKS: CARRIER AGGREGATION DEPLOYMENT SCENARIO

International Journal of Computer Networks & Communications (IJCNC) Vol.12, No.4, July 2020
DOI: 10.5121/ijcnc.2020.12403 41
MULTI-CRITERIA HANDOVER DECISION FOR
HETEROGENEOUS NETWORKS: CARRIER
AGGREGATION DEPLOYMENT SCENARIO
Maryam Abdulazeez- Ahmed, Nor Kamariah Nordin, Aduwati Bint Sali and
Fazirulhisyam Hashim
Department of Computer and Communication Systems Engineering, Universiti Putra
Malaysia, 43400, UPM Serdang, Selangor Malaysia
ABSTRACT
Carrier aggregation and integration of heterogeneous networks allow the mobile user equipment to benefit
from wider bandwidth and radio coverage of different access technologies. However, these technologies
have increased handoff scenario probability through user equipment’s mobility, leading to high outage
probability and low throughput. Handover is an important aspect of mobility management which allows
users to migrate from one cell to another without losing connections. But no lone access technology meets
the requirements of providing seamless communication without loss and delay. Therefore, high-quality
service continuity and reliable user equipment access to network anywhere and at any time require the
design of an appropriate handover decision algorithm. In this paper, a multi-criteria based handover
decision-making algorithm is proposed to evade loss of communication and provide better performance to
the system. It adaptively makes handover decisions based on different decision criteria (load, availability
of resources, and the handover scenario type) in addition to signal interference to noise ratio. The multi-
criteria handover decision making algorithm-based method chooses among the cells that satisfy the
requirements for the handover. As compared to some existing handover decision algorithms, simulation
result shows that this algorithm improves system performance in terms of handover failure by 93%, 72%,
and 58%; radio link failure reduction by 77%, 43%, and 22%; and handover ping–pong by 81%, 59%, and
36% over the conventional received signal strength, received wireless transmission line, and multi-
influence factor hand over decisions respectively.
KEYWORDS
Carrier aggregation, handover decision, user equipment, heterogeneous network, Long Time Evolution-
Advanced, multi-criteria, signal-to-interference-plus-noise-ratio, interference-to-interference-plus-noise-
ratio
1. INTRODUCTION
Several works have proposed various handover decision algorithms in wireless communication
systems. In conventional handover, Received Signal Strength (RSS) and/or power level received
from candidate base stations is compared in [1]. Also, other parameters such as: (a) RSS with a
threshold (b) RSS with hysteresis (c) RSS with hysteresis and threshold [2][3] (d) RSS with
hysteresis and distance [4][5] (e) Signal-to-Interference-plus-Noise-Ratio (SINR) [6][7][8][9]and
(f) Interference-to-Interference-plus-Noise-Ratio (IINR) [10]. However, their handover decision
algorithms are based on a single parameter. In order to enhance handover decision through user’s
movement, other factors like bandwidth and availability of resources, velocity of mobile terminal,
distance between mobile stations, and Evolved Node Bs (eNodeBs) need to be considered so as
to maximize system performance and User Equipment (UE) satisfaction. This is because

42
traditionally, mobile devices may be moved for better service quality. Thus, employing only
single criteria causes service interruption and unbalanced network load which in turn degrades
user’s throughput and increases its outage probability.
In [11][12][13][14], handover decisions are taken based on multi-criteria factors but none
considers carrier aggregation techniques. Carrier aggregation in heterogeneous networks leads to
increased access in technologies, bandwidth, transmission rate, energy requirement, coverage,
and enhanced network protocols. For the purpose of efficiency, one infrastructure capable of
interconnecting multiple access networks is required to support data concurrence at all cell
locations. Thus, efficient handover mechanism is required for ensuring seamless connectivity and
uninterrupted service delivery for future generation wireless networks.
Carrier Aggregation (CA) is one of the main background features for Long Time Evolution-
Advanced (LTE-A). With CA, users would gain access to a total bandwidth of up to 100MHz.
The system bandwidth may be contiguous, non-contiguous, and inter-band noncontiguous [15].
The bandwidth of these Component Carriers (CCs) can vary widely, ranging from 1.4MHz to
20MHz for Long Time Evolution (LTE) carriers. Also, propagation characteristics differ in
different CCs. For example, the propagation characteristic of CCs in the 800MHz band differs
from a component carrier in the 2.4GHz band [16].
Also, the introduction of CA technology, brings with it, a new handover scenario, which can be
performed between the CCs serving under the same sector and the same eNodeB with a
probability of redefining the Primary Component Carriers (PCCs). This results in increased
handover probability that consequently increases throughput degradation. The handover scenario
here is being translated into an efficient Handover Decision (HoD) provided the serving PCC
gives enough RSS to the UEs on acceptable levels.
Though [17] has studied handover decisions in the CA scenario using multi-influence criteria, but
only macro celle NodeB is considered. This cannot give the maximum benefit of CA in terms of
capacity and coverage increment. On the other hand, a limited number of UEs can be served by
outdoor macro eNodeB in heavily populated areas with the available spectrum. Therefore, the CA
technique and heterogeneous networks are possible solutions [18][19][20] to improve cell
capacity and utilization of the available spectrum. On the other hand, the seamless mobility of
UEs in the CA scenario using the same parameters of a macro eNodeB faces challenges due to
the aggregation of multiple CCs [21].
Fundamentally, handover is considered during the period when a voice call is in progress and
moving toward the current eNodeB (i.e source to a target eNodeB). In this case, signal
measurements obtained at a UE from the neighboring eNodeB, are reported by the UE to the
serving eNodeB and the handover decision is based on the measurement report from the UE at
the source eNodeB. Whereas in CA-based system, due to multiple CCs, the source and the target
eNodeBs can be CCs of the same or different eNodeBs. Employing such a signal Measurement
Report (MR) for decision making is not speedily concluded. Thus, it leads to a high probability of
handover failure, increase throughput degradation, and user outage probability especially to the
high-velocity UEs [22][23].
The main idea of handover is to maintain the connection while the subcarrier moves out of the
coverage area of the source eNodeB and enters the vicinity of the target eNodeB. Availability of
services to satisfy the customer is one of the major goals in cellular networks communication
[24]. However, efficient handover decision algorithm leads to an improvement in continuous
connectivity and enhancement in the network performances are inadequate. Therefore, it is novel

43
to look into HoD algorithm that can satisfactorily handle the major tasks in cellular network by
enhancing the level of the services, and manage the resources of the network simultaneously, and
serve as motivation to carrying out this work.
In this work, a proposed algorithm called Multi-criteria Handover Decision (MCHoD) is
presented to take tactful decision on handover when heterogeneous Network (HetNet) and CA is
implemented in LTE-A system. MCHoD dynamically changes handover decision algorithm
based on the signal strength, availability of resources, UE distance from the serving eNodeB and
the handover scenario type. MCHoD algorithm seeks to improve the system performances by
providing higher efficiency in SINR, cell edge spectral efficiency, and low outage probability in
the event of users’ mobility. The major contribution of this work is the development of MCHoD
algorithm on CA, integrating macro-eNodeBs (MeNodeBs) and femto-eNodeBs (FeNodeBs) to
provide UEs uninterrupted services in users’ mobility network.
The remaining of this paper is organized as follows. Related Work is presented in Section 2
followed by handover with CA techniques in section 3. The proposed algorithm and the system
model are described in Section 4. In section 5, results and discussions are presented while
conclusion is given in Section 6.
2. RELATED WORK
The handover decision is taken by the serving eNodeB based on the measurement report that is
received from the served UE. MR contains a list of signal levels of neighboring cells, and it can
provide other information based on the HoD algorithm being implemented. There are several
methods of a handover decision algorithm. Generally, it may be classified into the following
categories:
1) The traditional RSS-based criteria
2) Fuzzy logic
3) Multi-influence Factor (MIF) based
RSS–based approach is the traditional handover decision algorithm that uses RSS as the main
criteria to trigger handover [2][25]. In this case, the signal strengths of neighboring eNodeBs are
compared and consequently, the handover is enabled. The RSS-based technique has been
enhanced by adopting the RSS threshold (hysteresis) [5] and by combining the RSS threshold
with the user’s distance [4], SINR [6], and IINR [10]. To suppress the Ping-Pong effects, a
threshold is defined and applied in relative signal strength. The handover is initiated if existing
eNodeB RSS is lower than the threshold value and the new eNodeB RSS is stronger than the
threshold. Though, RSS based method is simple in principle and easy to use in measuring the
characteristics and the quality of services. However, the RSS signal fluctuates due to path loss in
the wireless medium. This effect results in unnecessary handovers, ping pong, and throughput
degradation. Also, RSS with threshold has drawbacks of handover delay and high call drop.
However, recent studies have demonstrated that by implementing CA and HetNets technologies,
handover decision becomes more challenging. The RSS-based criterion is not complete enough
for a decision process in HetNet with CA technique.
There are studies which employed fuzzy logic in decision making process to improve the
intelligence of HoD [26][27] for network selection and performances. Fuzzy logic method has
advantage of inherent potential to solve the precision problems but with the cost of design
complexity.

44
Meanwhile, an MIF decision making criteria was studied in a coordinated carrier aggregation
deployment scenario [17]. MIF applied different parameters that influence the decision making
process. Though, the simulation result shows that ping pong has been reduced but the result is not
optimal when HetNet is considered in CA.
The algorithm based on multi-criteria reduces the handoff delay and service disruption time. This
in turn decreases the packet lost, and increases throughput during handover [11][13][28].
Because of this, in MCHoD the chosen criteria ensure that the accuracy of the decision making is
held high. The main criteria are either user-related or network related, such as RSS, mobility,
application, and bandwidth. Despite all the earlier works are based on different criteria in
HetNets and CA, the question of which of the parameters yield an optimal handover process is an
open question. So, to find out the best parameters for successful handover process, a new
algorithm called multi-criteria handover decision algorithm is proposed for efficient handover
decision making when CA and HetNet is considered in LTE-A system. Table 1 compares some
of the existing work on different algorithms.
Table 1. Comparison of related works
Algorithm Application Achievement Drawback
RSS HetNet Simple Inaccurate
High rate of unnecessary
handover
Ping-pong effect
Fuzzy Logic HetNet Accurate precision High design complexity
MIF Coordinated CA Ping pong effect and
unnecessary
handover reduced
Only single eNodeB was used
Therefore, the proposed MCHoD algorithm combines RSS, bandwidth, speed, handover scenario,
and SINR in decision process in other to achieve seamless mobility.
3. CARRIER AGGREGATION BASED HANDOVER TECHNIQUES IN HETNETS
The introduction of CA technique in LTE-A system raises the number of aggregated CCs that can
be deployed at one eNodeB and simultaneously assign it to a UE. These CCs are grouped into
two main types. The first one is known as a PCC, whereas the second type of CCs is called
Secondary Component Carrier (SCC) [29][30].
The CCs concept of CA techniques considered in this study are as follows. We assumed that the
PCC is always at the Micro eNodeB (MeNodeB), whereas, the SCC is at the Femto eNodeB
(FeNodeB). With this configuration, the UEs always have a stable anchor PCC. At the
commencement of radio resource connection (RRC), UE is associated to the PCC which provides
the best signal quality over all the active CCs as the serving cell that is always active throughout
the active mode of the UE [31][32][33]. Data transmission between the UE and the eNodeB; and
the control signaling information exchange are usually carried out with PCC. Apart from being
used as data transmission channel, PCC is employed for random access procedure and allocation
of SCC. In the process, radio link failure event is recorded only if there is radio link connection
failure over the PCC. On the other hand, SCC can be configured and activated by eNodeB as an
additional CC used for additional resources for the served UE where higher data rate is required.

45
However, SCC cannot be used for exchanging control signaling messages between a UE and an
eNodeB [34].
Addition more to the developments, Radio Link Failure (RLF) is an indicator recorded when a
connection over the PCC fails. If this happens, Radio Resource Control (RRC) re-establishment
procedure comes into action over the named PCC. In the same vein, the Non-access Stratum
(NaS) recovery procedure is triggered if the RCC re-establishment procedure over the PCC is
failed within T310. T310 is the maximum allowed time for recovering connection through the
RRC re-establishment procedure [30][35].
Unlike the LTE system, the CA in the LTE-A system has increased handover scenarios (i.e.
changing of PCC in the same sector) influenced by aggregation of multiple carriers in addition to
channel conditions of two or more adjacent cells that may completely differ for the specific UE.
This poses challenges for the target eNodeB to guarantee the reservation of sufficient system
resources for the incoming UE’s service requirement. Therefore, employing single criteria in the
handover decision-making process may result in inappropriate handover for the target eNodeB
thus, leading to high handover probability which in turn increases outage probability and
throughput degradation.
In this paper, in order to reduce the outage probability and increase the spectral efficiency, we
propose a multi-criteria handover decision method with additional parameters such as SINR,
bandwidth, and data rate. This algorithm provides a set of parameters for better handover.
4. PROPOSED MULTI-CRITERIA HANDOVER DECISION ALGORITHM
In response to handover decision challenges of CA concept in LTE-A and future generation
systems, different handover decision algorithms have been proposed but none of the algorithms
has given the optimal result expected, particularly for HetNet’s Carrier Aggregation Deployment
Scenario (CADS). Therefore, in this study, we propose an efficient multi-criteria handover
decision (MCHoD) algorithm that utilizes both the network parameter (network topology and
radio resources) and mobile parameter (load, and UE distance) as input for decision making.
Due to the deployment structure of Het-Net’s CADS, UEs may not receive a signal from only the
MeNodeB, but also from FeNodeB. UEs are mostly connected to MeNodeB for high signal
reception and quality of service but are handed over to FeNodeB and vice-versa when the signal
strength drops below acceptable levels. This consequently leads to load imbalance. Furthermore,
differences in the transmitted power between MeNodeB and FeNodeB often cause frequent
handover, especially, ping-pong handover which negatively affects the system throughput.
Therefore, in this study, we incorporate Cell Range Expansion (CRE) power control technique
standardized in the Third Generation Partnership Project (3GPP) [36], as a virtual bias, added to
actual UE received power. This assists the UE with the association decision required. CRE is a
practical power control technique that boosts the received signal power strength of FeNodeB
which forces the UEs to offload to it. The biasing technique limits the MeNodeB load and
stabilizes the UE when it transits around the cell edges. At the same time, it reduces ping-pong
probability and consequently increases the network throughput.
Meanwhile, CRE will also alter the UE’s handover position due to changes in the coverage area
by the FeNodeB. During the UE association process, the handover margin is used to adapt the
received signal strength (RSS) in MCHoD. The PCC handover decision is initiated by UE
performance parameters such as SINR Handover management (HoM) and SINR threshold; and

46
handover scenario types. With this, handover decisions can be adaptively executed according to
the following conditions as depicted in the flowchart in Figure 1.
1) If the target cell SINR is larger than the sum of the handover margin and the serving cell
SINR then handover to target cell for all UEs on PCC, subject to condition 2.
2) That the target cell load is less than the sum of the serving cell load and the load margin
on inter-frequency components.
On the other hand, if the handover type is changing the serving sector or serving eNodeB (macro
to Femto), the decision will be based on resource availability, UE distance to the eNodeB, and
SINR. The handover decision can then be executed if the following conditions are satisfied:
1) That the target SINR is larger than that of the serving SINR plus handover margin
2) That the serving cell load is larger than the target cell load by a predefined load margin
3) That the UE distance to the target cell is less than the UE distance to the serving cell.
The combination of both the MCHoD and CRE techniques does not only increase the system
throughput but also reduces the handover failure rate.
Figure 1: Flow diagram of the proposed multi-criteria decision algorithm

47
4.1. System Model and Simulation Setup
The proposed algorithm is implemented as a multicell system using MATLAB toolbox system-
level simulator to generate LTE-A heterogeneous network topologies. MATLAB have a wide
variety of inbuilt libraries and features that can convert C, C++ code into MATLAB codes. It also
allows the necessary manipulation of data, improves the chances of new technology, and
increases the security surrounding a new technology [33]. We employed MATLAB as used by
prior studies [17][11][1].
The network scenario configuration was implemented by overlaying FeNodeBs on the
MeNodeBs coverage area with a uniform user’s distribution using the 3GPP LTE-A standard
topology [34][37]. The considered simulation environment consists of 61 hexagonal cell layout of
macro eNodeB located at the center with 500m inter-site distance. The MeNodeBs is subdivided
into three sectors and Omni-directional coverage femtocells are deployed in each sector to form
the CCs in inter-band noncontiguous deployment. The operating frequencies for CC1 and CC2
are 2GHz and 3.5GHz respectively premised on the band scenarios for Release 13 and 14 [34].
The UEs are uniformly distributed across the femtocell coverage area. The simulated cell
accommodates up to 50 mobile UEs randomly moving within the coverage area with a uniform
speed of 3, 20, 30, 60, 90, and 120 km/h. The assumption is that the frequency re-use factor is
taken to be 1. The inter-cell interference to other sectors is generated only from the first tier of the
neighboring eNodeBs. This is because the second tier neighboring eNodeBs are far away from
the serving eNodeB. Moreover, their signals are assumed to be weak and therefore, have no
significant impact on the served UEs. Radio resource control (RRC), radio link failure (RLF), and
NaS recovery procedures are all considered in this simulation. Adaptive Modulation and Coding
(AMC) schemes as specified in [38] is considered. The handover process follows the procedures
introduced in [39] with the assumption of 600ms as a time to trigger and 6dB as handover
margin. In order to improve performance accuracy; Radio Link Failure (RLF), radio resource
connection re-establishment, and NaS procedures are considered throughout the simulation. The
summary of the parameters that are used is listed in Table 2.
Table 2. Simulation parameters and values
Parameters Values
Propagation model Macrocell/Femtocell urban model
OFDM
Number of Component Carriers 2CCs each 20MHz
Bandwidth (MHz) 40 (2X20) MHz
Number of Resource Block 100
Resource Block size 180kHz
Number of Subcarrier per RB 12 Subcarrier per RB
Subcarrier Spacing 15kHz
Number of OFDM Symbol per frame 7
MeNodeB Parameters
Transmit power (dBm) 46dBm
Antenna gain + feeder loss 15dBi
Antenna height 15m
Noise figure 5
FeNodeB Parameters
Transmit power (dBm) 10dB
Antenna gain + feeder loss 2dBi

48
Antenna height 1m
Noise figure 4dB
Biasing value 10
User Parameters:
Antenna Gain 0 dBi
Antenna Height 1.5m
Noise Figure 9dB
Thermal Noise Power (NP) NP = Nt +10log (BW x 106) dB
Time-To-Trigger (TTT) 600ms
Handover Margin (HoM) 6dB
T311 10s
Measurement Interval for PCC and SCC 50ms
Process Delay Time 10ms
Simulation Time 500min
5. RESULT AND DISCUSSION
The proposed MCHoD algorithm is simulated and validated using MATLAB tools for the
simulation following the evaluation methodology of the LTE-A system.
Generally, high-speed UEs present several challenges to handover algorithms. To evaluate the
performance of MCHoD, it is compared with other handover decision algorithms such as
Conventional Received Signal Strength (CRSS), Received Wireless Transmission Line (RWTL),
and Multi-influence Factor (MIF) earlier discussed in section II. To demonstrate the effectiveness
of our algorithm, a comprehensive investigation of MCHoD at various speeds was examined
based on the probabilities of the Radio Link Failure (RLF), Handover Failure Rate (HFR), ping-
pong and throughput as shown in Figures 2, 3, and 4 respectively.
In terms of Handover Failure (HoF) reduction, the performance gains of MCHoD over CRSS,
MIF-HoD, and RWTL is presented in Figure 2. Generally, it is observed that at low speeds, the
rate of HoF is lower than at high speeds as in Figure 2a. This can be attributed to the fact that the
number of cells that a UE crosses over a certain period of time is lower at those speeds (i.e. 3, 20,
and 30 km/h). However, an increase in UE’s speed accelerates the frequency of handover because
as the UE moves away from the serving cell, signal reception becomes weaker, and so also is the
data rate. This results in a high HoF rate and consequently reduces the performance of the
system.
A plot of the average HoF probability shows that CRSS has the highest HoF rate as shown in
Figure 2b. The reason is that CRSS employs only a single criterion for handover decision
making. Besides, it does not consider the SINR and the availability of the resources. Therefore, it
suffices to say that, since handover consumes more resources, it is bound to experience more HoF
rates if the target cell is not having the required resources. The other two baselines, RWTL and
MIF have a much lower HoF compared to CRSS because the former utilizes RSS and
transmission line losses whereas, the latter employs more than one criterion in the handover
decision making process implementing CA scenario. The use of CA facilitates wider bandwidth
and effective utilization of available resources to make decisions. Quantitative assessment
produced shows that MCHoD achieves higher reduction gain in HoF by approximately 93%,
72%, and 58% over CRSS, RWTL-HoD, and MIF-HoD respectively.

49
Figure 2: Probability of handover failure (HoF). (a) At different speeds (b) average handover failure
probability for different decision algorithms.
UE speed does not only affect handover, it has much to do with RLF probability. At high UE
speeds, radio link connection becomes weakened due to fading, leading to higher RLF failure.
Figure 3a illustrates the RLF rate at different speeds. Overall, it is observed that the probabilities
of RLF get higher as the UE’s speed increases but the effects vary among the four algorithms
examined. Similar to the evaluation of HoF, CRSS is much affected by RLF than the other
algorithms. Generally, RSS is known to be unstable [26]. So, its implementation as an only
deciding factor in the heterogeneous network which has different characteristic cannot give the
optimal result. Figure 3a shows the performances of RFL in relation to mobile speed where
MCHoD and MIF-HoD are both low. On the average (Figure 3b), MCHoD shows a good
performance (low RLF ratio) compared with the other algorithms, CRSS, RWTL, and MIF-HoD
by 77%, 43%, and 22%, respectively.

50
Figure 3: Probability of Radio Link Failure (RLF). (a) Failure at different UE’s speeds and (b) average
RLF for different decision algorithms
The plot of the Handover Ping Pong (HoPP) ratio against the speed is presented in Figure 4a. As
expected, due to its reliance on only the downlink signal quality, the CRSS method shows the
highest HoPP rate. Except for MCHoD which has insignificant ping-pong effects across the
different speeds examined, other decision criteria show the extensive influence of speed over
HoPP. Obviously, this phenomenon increases with increasing UE speed. In Figure 4a, as the RSS
in the downlink tends to fluctuate due to shadow fading, cell selection becomes unstable, leading
to a high HoPP ratio. The higher the HoPP, the more resources are wasted. In our algorithm,
HoPP is minimized much better by using MCHoD where the handover is triggered if the
requirements are met. Thus, the UE can be handed over to a suitable cell that satisfies the entire
requirements for a successful handover to the target cell. The MCHoD has the best performance.

51
It reduces the HoPP by approximately 81%, 59%, and 36% over the CRSS, RWTL, and MIF-
HoD, respectively.
The resultant effect of HoF, RLF, and HoPP reflects on the Cumulative Distribution Function
(CDF) probability UEs throughput. The higher the throughput value on the CDF probability plot,
the better the performance of the system (Figure 4b). It can be observed, therefore, that the
proposed MCHoD decision algorithm performs better than all the other algorithms evaluated in
this work. In short, MCHoD is designed to select the best target cell based on multi-criteria
decision method. This reduces the HoPP and increases the throughput gain.
Figure 4: (a) Average Ping-Pong Handover probability over different speeds and (b) CDF of UEs
throughputs
6. CONCLUSION
The proposed MCHoD performs better than previous handover decision algorithms as it
demonstrate a significant reduction in HoF, RLF, and HoPP rates. This is an achievement and can

52
be explained thus. First, there is a choice in the mapping of CCs, and PCC configured on
MeNodeB for wider coverage, and the SCC on FeNodeB to extend the signal to the cell edge
users. Second, the implementation of AMC schemes which allows UEs to be assigned the most
suitable channel irrespective of variations in the channel conditions linearly increases the SINR
which eventually leads to improvement in the entire system throughput. Third, the
implementation of CRE techniques to resolve the load imbalance problem among the configured
networks thereby offloading traffics from the congested MeNodeBs. One of the outstanding
performances of the MCHoD is that it minimizes the effects of the parameters examined (RLF,
HoF, and HoPP) across the observed UE speeds. It also provides efficient resource utilization as a
result of the increased number of aggregated CCs that can be deployed at an eNodeB and
simultaneously assigned to a UE. This development is useful whenever there are a number of
users that are required to add or remove one or more CCs. Finally, MCHoD contributes to the
enhancement of the system performances by increasing user SINR, spectral efficiency, and
reducing the user’s outage probability. Since the future generation wireless communication will
be built on these qualities, MCHoD stands a promising handover decision algorithm for carrier
aggregation, Het-Nets deployment in LTE-A and future generation of wireless communication
As with every research, there are limitations and this study is no exception. Since carrier
aggregation technique and heterogeneous networks deployment appears to be the prospect of
wireless generation networks for higher capacity and data rate, further study needs to be
conducted on effective deployment of Femtocell (FeNodeB) for high-speed outdoor application.
Parameters like security, password protection, interoperability, etc. can also be included.
Technically, a femtocell is deployed for use indoors, but in this study, it is considered for outdoor
use and tested over different UE speeds. Certainly, its deployment outside is practicable but may
result in a high rate of ping-pong especially with user speed above 120 km/h.
CONFLICTS OF INTEREST
The authors declare no conflict of interest.
REFERENCES
[1] S. Maaloul, M. Afif, and S. Tabbane, “Handover Decision in Heterogeneous Networks,” in IEEE
30th International Conference on Advanced Information Networking and Applications, 2016, pp.
588–595.
[2] G. P. Pollini, “Trends in Handover Design,” IEEE Commun. Mag., vol. 34, no. 3, pp. 82–90, 1996.
[3] S. Dhar ROY, “Performance Evaluation of Signal Strength Based Handover Algorithms,” Int’l J.
Commun. Netw. Syst. Sci., vol. 02, no. 07, pp. 657–663, 2009.
[4] I. Ken-Ichi, W. Soichi, S. Jen-Shew, and S. Takuro, “Performance of Handoff Algorithm Based on
Distance and RSSI Measurements VOL. 51, NO. 6, NOVEMBER 2002,” IEEE Trans. Veh. Technol.,
vol. 51, no. 6, 2002.
[5] S. Lal and D. K. Panwar, “Coverage Analysis of Handoff Algorithm with Adaptive Hystersis
Margin,” in International conference on Information Technology, 2007, pp. 133–138.
[6] K. Yang, I. Gondal, B. Qiu, and L. S. Dooley, “Combined SINR Based Vertical Handoff Algorithm
for Next Generation Heterogeneous Wireless Networks,” in IEEE Globecom proceedings,
2007, pp. 4483–4487.
[7] A. K., N. K.;, and a. D. P.Ayyappan, “SINR Based Vertical Handoff Scheme for QoS in
Heterogeneous Wireless Networks,” in International Conference on Future Computer and
Communication (ICFCC 2009)., 2009, pp. 117–121.
[8] A. A. Bathich, M. D. Baba, R. A. Rahman, and U. T. Mara, “SINR Based Media Independent
Handover in WiMAX and WLAN Networks,” in international conference on Computier Application
and iIdustrial Electronics (ICCAIE), 2011, pp. 331–334.

53
[9] P. Baruah, N. Sarma, and S. S. Satapathy, “SINR based vertical handoff algorithm between GPRS
and WiMAX networks,” J. Theor. Appl. Inf. Technol., vol. 14, no. 2, pp. 107–110, 2010.
[10] H. Choi, “An Optimal Handover Decision for Throughput Enhancement,” IEEE Commun. Lett., vol.
14, no. 9, pp. 851–853, 2010.
[11] G. Mahardhika, M. Ismail, and R. Nordin, “Multi-criteria vertical handover decision algorithm in
heterogeneous wireless network,” J. Theor. Appl. Inf. Technol., vol. 54, no. 2, pp. 339–345, 2015.
[12] M. A. Radhwan and A. Z. Zuriati, “Enhanced Handover Decision Algorithm in Heterogeneous
Wireless network,” Sensors, vol. 1626, no. 17, pp. 1–14, 2017.
[13] S. Pahal and P. Sehrawat, “Multi-Criteria Handoff Decision Algorithms In Wireless Networks,” J.
Mob. Comput. Appl., vol. 2, no. 2, pp. 2394–42, 2015.
[14] M. M. Alkhawlani, “Multi-Criteria Vertical Handover for Heterogonous Networks,” Int. J. Wirel.
Mob. Networks, vol. 3, no. 2, pp. 149–163, 2011.
[15] X. Lin, J. Andrews, and A. Ghosh, “Modeling, analysis and design for carrier aggregation in
heterogeneous cellular networks,” IEEE Transactions on Communications, pp. 1–12, 2012.
[16] X. Lin, J. G. Andrews, R. Ratasuk, B. Mondal, and A. Ghosh, “Carrier Aggregation in Heterogeneous
Cellular Networks,” IEEE Wireless Communication Symposium, pp. 5199–5203, 2013.
[17] I. Shayea, M. Ismail, R. Nordin, and H. Mohamad, “Adaptive Handover Decision Algorithm Based
on Multi-Influence Factors through Carrier Aggregation Implementation in LTE- Advanced
System,” J. Comput. Networks Commun., vol. 2014, p. 8, 2014.
[18] 3GPP, “TR 136 912 - V13.0.0 - LTE; Feasibility study for Further Advancements for E-UTRA (LTE-
Advanced) (3GPP TR 36.912 version 13.0.0 Release 13),” 2016.
[19] 4G Americas, “LTE-Advanved: ‘Carrier Aggregation Aggregation Technology Development and
Deployment worldwide’ http://www.4gamericas.org/files/8414/1471/2230/,” 2014.
[20] T. Anti, “LTE‐Advanced Standardization,” in LTE-Advanced: 3GPP Solution for IMT-Advanced, 1st
ed., H. Holma and A. Toskala, Eds. John Wiley & Sons, 2012, pp. 8–13.
[21] L. Liu, M. Li, J. Zhou, X. She, L. Chen, Y. Sagae, and M. Iwamura, “Component Carrier
Management for Carrier Aggregation in LTE-Advanced System,” in IEEE 73RD Vehicular
Technology Conference (VTC Spring), 2011, pp. 1–6.
[22] 3GPP, “3GPP TR 36.839 V11.0.0 (2012-09), Mobility enhancements in heterogeneous networks
(Release 11),” 2012.
[23] K. Vasudeva, M. Simsek, D. Lopez-Perez and I. Guvenc and S. Meryem, “Analysis of Handover
Failures in Heterogeneous Networks with Fading,” IEEE Transactions on Vehicular Technology, vol.
66, no. 7, pp. 6060–6074, 2017.
[24] R. Kaur and H. Singh, “An Efficient Approach for Handover Decision Making In Wireless
Networks,” Int. Res. J. Eng. Technol., vol. 05, no. 02, pp. 454–458, 2018.
[25] Sonia and G. S. Saini, “Optimized RSS based Algorithm for Heterogeneous Networks,” Indian J. Sci.
Technol., vol. 9, no. December, p. 6, 2016.
[26] P. Tillapart, T. Thumthawatworn, P. Viriyaphol, and P. Santiprabhob, “Intelligent Handover Decision
Based on Fuzzy Logic for Heterogeneouse Wireless Networks,” in Electrica/Electronics Engineering,
Computer , Telecommunication and Information Technology (ECTI-CON), 2015, pp. 1–6.
[27] V. A. Narayanan, A. Rajeswari, and V. Sureshkumar, “AN Intelligent Vertical Handover Decision
Algorithm for Heterogeneous Networks,” Am. J. Appl. Sci., vol. 11, no. 5, pp. 732–739, 2014.
[28] M. Khan, “Multi-criteria based vertical handover decision in heterogeneous wireless network,” in
Proceedings of the 30th Annual ACM Symposium on Applied Computing - SAC ’15, 2015, vol.
2015, pp. 720–721.
[29] Y. D. Lee, S. D. Chun, S. J. Yi, S. J. Park, and S. H. Jung, “Carrier aggregation management method,
system and devices., [Google Patents].,” US 2011/0028148 A1, 2011.
[30] P. J. Song and J. Shin, “Method for handover in multi-carrier system, [Google Patents], US8688120
B2.,” US8688120 B2., 2014.
[31] Z. Shen, A. Papasakellariou, G. Dirk, and F. Xu, “Overview of 3GPP LTE-Advanced Carrier
Aggregation for 4G Wireless Communications,” IEEE Communications Magazine, no. February, pp.
122–130, 2012.
[32] L. Liu, M. Li, J. Zhou, X. She, L. Chen, Y. Sagae, and M. Iwamura, “Component Carrier
Management for Carrier Aggregation in LTE-Advanced System,” IEEE 69th Veh. Technol. Conf.
spring (VTC2009-Spring), Barcelonal, Spain, pp. 1–6, 2011.
[33] “DOCOMO R4-102114: Simulation Assumption for mobility performance in Carrier Aggregation;

54
DoCoMo, Montreal, Canada,” 2010.
[34] 3GPP, “Radio Frequency (RF) system scenarios, (3GPP TR 36.942 version 14.0.0 Release 14),
FRANCE,” 2016.
[35] I. Mikio, K. Etemad, M. Fong, R. Nory, and R. Love, “Carrier Aggregation Framework in,” IEEE
Commun. Mag., no. August, pp. 60–67, 2010.
[36] Qualcomm, “R1-100701 Importance of Serving Cell Association in Heterogeneous Networks,” 2010,
pp. 1–6.
[37] 3GPP, “TR 136 932 - V14.1.0 - LTE; Scenarios and requirements for small cell enhancements for E-
UTRA and E-UTRAN,” 2017.
[38] 3GPP, “TS 136 211 - V14.2.0, Physical channels and modulation, (3GPP TS 36.211 version 14.2.0
Release 14),” 2017.
[39] 3GPP, “TS 36.300 , Overall description ; 3GPP TS 36.300 version 13.9.0 (Release 13).,” 2017.
AUTHORS
Maryam Abdulazeez- Ahmed received her first degree in Electronics and
Telecommunication Engineering from Kwara Polytechnic Ilorin Nigeria in 1989,
Master of Engineering Management in 2011 from Universiti Putra Malaysia
(UPM) and Ph.D. (in view), Wireless Communication Engineering from UPM.
Her research interests include mobile cellular, wireless, and green communication.
She has published several papers in journals and conferences.
Nor Kamariah Noordin received her B.Sc. in Electrical Engineering from the
University of Alabama, USA, in 1987. She became a tutor at the Department of
Computer and Communication System Engineering, UPM, and obtained a Master
Degree from Universiti Teknologi Malaysia (UTM) and Ph.D. from UPM. She is
currently the Director of Corporate Planning Division at the Office of the Vice-
Chancellor. She has received National, and research awards, being a researcher at
the Wireless and Photonic Network Research Center of Excellence (WiPNET),
UPM. She has over 100 publications in journals and conferences.
Aduwati Bint Sali is a Professor at Department of Computer and Communication
Systems, Faculty of Engineering, UPM. She obtained her Ph.D. in Mobile and
Satellite Communications from the University of Surrey in the United Kingdom, in
July 2009, her MSc in Communications and Network Engineering from UPM, in
April 2002, and her B. Eng in Electrical Electronics Engineering
(Communications) from University of Edinburgh, UK, in 1999. Her research
interests are radio resource management, MAC layer protocols, satellite
communications, wireless sensor networks, satellite-assisted emergency
communications, and 3D video transmission over wireless networks. She is a researcher at the WiPNET,
UPM.
Fazirulhisyam Hashim holds an MSc degree from Universiti Sains Malaysia and a
Ph.D. in Telecommunications Engineering from the University of Sydney,
Australia. His research interests include network security and quality of service of
next-generation mobile networks, green communication systems, cognitive
networks, and wireless sensor networks. He is a member of IEEE and ACM, and
currently a senior lecturer and researcher at the Wireless and Photonic Network
Research Center of Excellence (WiPNET), Universiti Putra Malaysia.

MULTI-CRITERIA HANDOVER DECISION FOR HETEROGENEOUS NETWORKS: CARRIER AGGREGATION DEPLOYMENT SCENARIO

More Related Content

What's hot (18)

Similar to MULTI-CRITERIA HANDOVER DECISION FOR HETEROGENEOUS NETWORKS: CARRIER AGGREGATION DEPLOYMENT SCENARIO (20)

More from IJCNCJournal (20)

Recently uploaded (20)

MULTI-CRITERIA HANDOVER DECISION FOR HETEROGENEOUS NETWORKS: CARRIER AGGREGATION DEPLOYMENT SCENARIO