New weight function for adapting handover margin level over contiguous carrier aggregation deployment scenarios in lte advanced system

Vol.:(0123456789)
Wireless Personal Communications
https://doi.org/10.1007/s11277-019-06463-2
1 3
New Weight Function for Adapting Handover Margin Level
over Contiguous Carrier Aggregation Deployment Scenarios
in LTE‑Advanced System
Ibraheem Shayea1,2
· Mahamod Ismail1
· Rosdiadee Nordin1
· Mustafa Ergen2
·
Norulhusna Ahmad3
· Nor Fadzilah Abdullah1
· Abdulraqeb Alhammadi4
·
Hafizal Mohamad5
© Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract
In this paper, an Adaptive Handover Margin algorithm based on Novel Weight Function
(AHOM-NWF) is proposed through Carrier Aggregation operation in Long Term Evolu-
tion—Advanced system. The AHOM-NWF algorithm automatically adjusts the Hando-
ver Margin level based on three functions, f(SINR), f(TL) and f(v)
, which are evaluated
as functions of Signal-to-Interference-plus-Noise-Ratio (SINR), Traffic Load (TL) , and
User’s velocity (v) respectively. The weight of each function is taken into account in order
to estimate an accurate margin level. Furthermore, a mathematical model for estimating
the weight of each function is formulated by a simple model. However, AHOM-NWF
algorithm will contribute for the perspective of SINR improvement, cell edge spectral effi-
ciency enhancement and outage probability reduction. Simulation results have shown that
the AHOM-NWF algorithm enhances system performance more than the other considered
algorithms from the literature by 24.4, 14.6 and 17.9%, as average gains over all the con-
sidered algorithms in terms of SINR, cell edge spectral efficiency and outage probability
reduction respectively.
Keywords Adaptive handover margin · Weight function · Carrier aggregation · LTE-
advanced · Cell edge throughput and outage probability
1 Introduction
Handover (HO) is an important process which supports service continuity and balances
traffic loads between cells in the entire network area throughout an active user call. How-
ever, the handover process must be imperceptible to the user to provide a seamless connec-
tion with good service quality. Thus, the Handover Decision (HOD) Algorithm (HODA)
must be well-designed and intact, and the Handover Execution (HOE) process must be fast
and reliable. This paper focuses on the handover decision during CA implementation in
* Ibraheem Shayea
ibr.shayea@gmail.com
Extended author information available on the last page of the article

I. Shayea et al.
1 3
LTE-Advanced System Release 10 to 13 (Rel.10 & Rel.13) based on two contiguous Com-
ponent Carriers (CCs).
In mobile communication systems, several HODAs have been proposed to provide
a seamless connection with good service quality. These algorithms have been proposed
based on different parameters, such as the distance [1, 2], Received Signal Strength (RSS)
[2–4], SINR [5–8] and Interference to other- Interferences-plus-Noise Ratio (IINR) [9].
The most common HODA among those are proposed is taken based on RSS quality, which
is divided into three categories [3, 10, 11]. The first category is taken based on RSS only,
which is decides to initiate the handover procedure once the target RSS become better than
the serving RSS. The second category is deisgned based on RSS with the threshold level. It
decides to initiate handover procedure once the serving RSS level fall below the threshold
level and target RSS become better than the serving RSS. The third category considered
RSS with margin level, which determines the initiation of the handover procedure once the
target RSS becomes better than the serving RSS at a marginal level. It is the most practi-
cal algorithm recently used for taking handover decision [3, 10, 12–15]. The margin level
between the serving and target RSS contributes for reducing the unnecessary handover pro-
cedure, which leads to prevent the ping-pong effect. While the ping-pong effect is a fre-
quent unnecessary handover scenario between two neighbour stations which is caused by
rapid fluctuations in the signal strength from both stations [16]. Thus, HOM is considered a
very sensitive control parameter for taking HOD, as well as a suitable HOM level contrib-
uting to an intact HOD, which in turn contributes to a reduction in throughput degradation
and outage probability.
Decreasing or increasing HOM levels lead to a noticeable effect on system perfor-
mance, both negatively and positively. In the first case, if HOM level is decreased the ping
pong effect will increase [17, 18], which leads to increase the waste of network resources
(throughput degradation) [19] and inefficient communication between the User Equip-
ment (UE) and serving network will be resulted. On the other hand, decreasing HOM leads
to reduction in the handover failure rate [20] and outage probability [17, 21], which are
required in mobile wireless systems. In the second case, if HOM level is increased the
ping pong effect will reduce [17, 18], leading to increased probability of connection stabil-
ity. Meanwhile, increasing HOM level leads to increase users’ outage probability [17, 21],
which is undesirable in mobile communication applications. Therefore, to balance the pro
and cons of varying the HOM level, AHOM Algorithm (AHOMA) has been proposed to
adapt HOM level between a minimum and maximum levels in order to select the suitable
margin level that can contribute for taking a proper HOD.
There have been several AHOM algorithms proposed based on single and multiple
parameters in a homogeneous network. AHOM algorithms based on a single parameter,
such as distance [10, 13], service type [22], velocity [23–25] and traffic load [26], adjust
the margin levels automatically based on variation of the corresponding parameter. The
estimated level can contribute for enhancing system performance compare to the fixed-
HOM level. However, the estimated level cannot be accurate, since it is estimated in per-
spective of single parameter only, while there are other influence parameters which have
not been considered, as explained in the next section. The the AHOM algorithms based on
multiple parameters such as Cost Function (AHOM-CF) [12] the margin level is automati-
cally adjusted based on multiple parameters. In this case, the estimated HOM level may
be more accurate than that is estimated based on a single parameter only. However, there
are uninfluential factors such as service type which should be ignored, and other influence
factors (i.e. distance, channel condition, noise, interferences) which should be considered.
Consequently, a new AHOM algorithm based on multiple influence parameters is needed.

New Weight Function for Adapting Handover Margin Level over…
1 3
Furthermore, the sensitivity of high outage probability through the users’ mobility needs
more optimal algorithm that can estimate more accurate HOM level.
In this paper, AHOM-NWF is proposed through CA operation in LTE-Advanced sys-
tem. This algorithm attempts to adjust the margin level automatically based on SINR, traf-
fic load, and UE’s velocity. A mathematical model of the proposed algorithm has been
formulated based on a multiple functions, which are evaluated as a function of SINR, traf-
fic load and UE’s velocity. Moreover, a mathematical formula for estimating the weight of
each function is modelled in this paper. However, this proposed algorithm is designed for
throughput enhancement and outage probability reduction through CA operation in LTE-
Advanced system only. It is investigated and compared to two different adaptive handover
algorithms in order to point out its achievable enhancement.
The remainder of this paper is organized as follows: Sect. 2 describes the Background
and Related Work, while Sect. 3 presents the Proposed Algorithms. System Model is
described in Sect. 4, followed by Results and Discussion in Sect. 5. Finally, Sect. 6 con-
cludes this paper.
2 Background and Related Work
In cellular mobile communication systems, handover is the main and essential Radio
Resource Management (RRM) process that is required to support reliable UE connectiv-
ity at different mobility conditions [27–30]. It always maintains the radio link connection
for the UE to the best serving cell in the coverage area. The term handover, also called as
Handoff, can be defined, in general, as the process of switching a radio link connection
from the source to the target Base Stations (BSs). Therefore, the mobile UE can maintain
its radio connection during its movement within the cells by performing a handover process
from the serving Evolved Node B (eNB) to another eNB that provides better signal qual-
ity. Furthermore, the efficient handover can support service continuity and enhancing UE’s
throughput, ideally, without any service interruption. In wireless systems, there are two
types of handover procedures which can be performed between cells, known as horizontal
and vertical handovers. In horizontal handovers, the procedure can be performed between
cells in a homogeneous network only, such as the handover procedure between two eNBs
in LTE network. In vertical handovers, the procedure can be performed between two cells
from different networks, such as the handover procedure from eNBs under LTE network
to a BS under WiMAX network. However, this paper focuses only on handover decisions
in terms of horizontal handover in a homogeneous network (LTE-Advanced System). In
horizontal handover, several studies have focused on handover decision with a fixed HOM
level [17–21] and AHOM level. Fixed handover margin level means that the margin level is
a constant through all the Transmission Time Intervals (TTIs), while AHOM level means
that the margin level is automatically adjusted periodically based on different factors as
illustrated in Fig. 1.
In [10, 13] AHOM algorithm based on Distance (AHOM-D) has been proposed, sim-
plified as follows:→ max
[
Mmax
(
d
R
)4
, Mmin
]
, where d is the distance between UE and
serving eNB, and R represents the cell radius in meter. Mmax and Mmin represent maxi-
mum and minimum handover margin levels, respectively. However, this algorithm
dynamically determines the HOM level as a function of the distance between the UE
and the serving eNB. Therefore, based on this algorithm, HOM level increases when the

I. Shayea et al.
1 3
UE oncoming toward the serving eNB, while it is decreased when the UE going away
from the serving eNB toward the target eNB. It is a useful algorithm when the network
resources and a good channel condition are always available with low UE’s movement
speed, but this not always be available. In such a case, AHOM-D cannot estimate the
suitable HOM level, since this algorithm considers only distance and other influencing
factors have not been considered.
In [23–25], AHOM algorithm based on user’s Velocity (AHOM-V) has been pro-
posed. The algorithm adjusts the margin level (ΔH) by utalizing the following model:
→ ΔH = r ⋅ Thdrop , where r is expressed by → r = log2(1 + v)
, in which v represents
UE’s velocity and Thdrop is the minimum RSS level that the quality of the radio link
below it become unacceptable. In this algorithm, higher margin level is estimated when
the UE speed is increased, while, a lower margin level is estimated when the UE speed is
decreased. It is a useful algorithm since it contributes for reducing the unnecessary hando-
ver procedure through the high UE’s movement speed by estimating high margin level. But
in other hand, it cannot estimate an accurate margin level since it considers UE’s velocity
only, while other influence factors have not been considered for adjusting the margin level.
In [26], AHOM algorithm has been proposed according to the Traffic Load (AHOM-
TL) of the serving and target eNBs in LTE system. The adaptive model is expressed as
a function of serving and target eNBs loads by → MH(e, k) = f(xe − xk) , where xe and xk
represent the loads of the serving and target eNBs, respectively. In this algorithm, the esti-
mated margin level is increased when the serving eNB’s load is decreased and target eNB’s
load is increased, while it is decreased when the serving eNB is overloaded and the tar-
get eNB is less-loaded. Although this algorithm contributes for balancing loads between
cells, but it estimates the margin level in perspective of traffic loads only. That leads to an
Max HO margin value
Average HO margin Value
Min HO margin Value
Qrslevmin
RSS over the Target CC
RSS over serving PCC
Factors
Received
Signal
Strength
(RSS)
Adaptive Margin Value
Fig. 1 Adaptive HO margin level in LTE-Advanced System

1 3
inaccurate estimate of margin level compared to the algorithm that consider multiple influ-
ence factors.
In [12], AHOM-CF has been proposed for adjusting the margin level in LTE Net-
work based on multiple parameters, which are the load difference between the serv-
ing and target cells, UE’s velocity, and the service type. In AHOM-CF, the HOM level
is adaptively estimated by this proposed expression: → M = Mdefault + ΔM , where,
Mdefault is the default margin level, while ΔM represent the margin level between the
serving and target eNB, which is expressed by ΔM = 𝛼.fl,v,s . 𝛼 is a factor expressed by
𝛼 = Mmax − Mdefault or 𝛼 = Mdefault − Mmin . While, fl,v,s represents the cost function,
which is simplified by the following formula:→ fl,v,s = 𝜔lNl + 𝜔vNv + 𝜔sNs . Where Nl , Nv
and Ns represent the normalized functions of the load difference between the serving and
target Cells, UE’s velocity, and the service type respectively. While, 𝜔l, 𝜔v, 𝜔s represent
the weight for the respective normalized function, where the sum of the weights must be
one (𝜔l + 𝜔v + 𝜔s = 1)
. However, these three normalized functions are the main factors
which contribute to adaptation of margin level. Although AHOM-CF considers multiple
parameters for estimating the margin level, it is not an optimal algorithm, as there are other
influencing factors which have not been considered and non-influtential factors have been
considered.
In Munoz et al. [31] proposed Fuzzy Logic Controller (FLC) algorithm to adaptively
modify the handover margin level only, while the Time-To-Trigger (TTT) interval is set
to 100 ms. The FLC adjusts the HOM level based on the average Call Drop Rate (CDR)
and Handover Ratio (HOR) per cell. Based on these ratios, the HOM level is optimized for
each cell individually, and it is restricted between 0 and 12 dB. The optimization opera-
tion is performed systematically in every Transmission Time Interval (TTI). However, the
authors have shown that, adjusting HOM levels based on FLC given a better reduction
gains in terms of call drop rate as compared to the conventional handover parameter opti-
mization algorithm.
In [32], a new handover self-optimization algorithm in LTE system based on a fuzzy
logic controller has been developed. The aim of that developed algorithm is to automati-
cally find out the suitable HOM and TTT. The presented results of the proposed algorithm
was compared with another four algorithms from the literature. The results show that the
proposed algorithm achieves some improvements in terms of handover as compared to
other algorithms.
Based on the presented studies, most of the proposed algorithms adjust the margin level
based on only a single parameter, such as distance [10, 13], service type [22], velocity [23–25],
traffic load [26], and Fuzzy Logic Controller [31, 32]. Since there are several influence factors
that can contribute for taking a proper HOD, such as the distance, channel condition, noise,
interferences, resource availability and UE’s velocity; therefore, estimating HOM margin level
in the perspective of single factor only leads to shortage for estimating a suitable level. That in
turn leads to increase the throughput degradation and outage probability. AHOM-CF consid-
ers multiple parameters for adjusting HOM level, but there are uninfluenced parameters need
to be ignored and other influence parameters to be considered. The uninfluenced parameter
such as service type can be ignored since all eNBs in LTE-Advanced network provides same
service type with same cost. Furthermore, HOD in horizontal handover is not taken based
on the service type, so it is normally taken either based on distance [1, 2], RSS [2–4], SINR
[5–8] or IINR [9]. Thus, there is no point of consideration of service type for adjusting HOM
level in a homogeneous network, while it can be considered in heterogeneous networks, as
each network can provide a different service type (i.e. Wifi provides internet, while LTE pro-
vides voice calls and broadband services) with a different cost. On the other hand, there are

I. Shayea et al.
1 3
influence parameters should be considered for adjusting HOM level in horizontal handover,
such as distance, channel condition, noise and interferences from the neighbours’ eNBs. These
parameters are influence factors as both the provided throughput and service continuity are
affected by them and handover decision can be taken based on one or more of these param-
eters. Furthermore, there is a lack of studies that are focused on AHOM based on multiple
factors with horizontal handover compare to vertical handover [33–37]. Adapting HOM level
based on a comprehensive algorithm considering different influence parameters is needed.
Moreover, all the AHOM algorithms in a horizontal handover [12] and vertical handover
[33–37], have not been formulated with any mathematical model for estimating the weight of
each normalize function that has been considered in their cost functions. Therefore, formulat-
ing a mathematical model for estimating the weight of each normalized function is required.
3
Proposed Adaptive HO Margin Algorithms
In this paper, a novel algorithm for adjusting margin level is proposed based on several influ-
ence factors such as distance, channel condition, noise, interferences, cell load, and user veloc-
ity. Since RSS is evaluated as a function of distance and channel condition, while SINR is
evaluated as a ratio of RSS to the interference plus noise ratio, SINR is thus as a factor which
will be suffice for estimating HOM level instead of distance, channel quality, noise and inter-
ferences. In terms of cell loads, the availability of resources at the target cell contributes for
performing successful handover. Also, it considers the traffic load for taking handover deci-
sion contribute for balancing loads between cells. That leads to enhanced user throughput and
reduced disconnection probability. It is considered an influential factor which should be taken
into account for adjusting margin level. According to the velocity, high movement speed of
users principals to increase the unnecessary handover rate [23, 24], which in turn leads to
degrade system performance. Thus, different UE’s velocities give different performance eval-
uations. Therefore, UE’s velocity needs to be considered for adjusting HOM level in order
to prevent the unnecessary handover, especially at the high movement speeds; which in turn
leads to enhanced user throughput and reduced outage probability.
Consequently, an HO algorithm is proposed to adjust the HOM level based on adaptive
function (fAHOM(SINR, TL, v))
, which automatically adjusts HOM level based on three func-
tions f(SINR), f(TL) and f(v)
, which are evaluated as functions of SINR, Traffic Load (TL)
and User’s velocity (v), respectively. The weight of each function is taken into account in order
to estimate an accurate margin level. However, the proposed function fAHOM(SINR, TL, v) can
be simplified by the following expression:
where MAvg represents the average HOM level, which is evaluated by
→ MAvg =
(
Mmax − Mmin
)
∕2 . 𝜔sinr , 𝜔TL and 𝜔v represent the weights of f(SINR) , f(TL)
and f(v) respectively. The value of these three functions is varied between {− 1} and {1},
while the weight of each function varies between {0} and {1}. The sum of all weights is
(1)
fAHOM =
⎧
⎪
⎪
⎨
⎪
⎪
⎩
MAvgx
�
𝜔sinrf(SINR) + 𝜔TLf(TL) + +𝜔vf(v)
�
if SINRT,S ≤ SINRThr
MAvgx
�
1 + 𝜔TLf(TL) + +𝜔vf(v)
� if SINRT SINRThr
SINRS ≥ SINRThr
MAvgx
�
−1 + 𝜔TLf(TL) + +𝜔vf(v)
� if SINRS SINRThr
SINRT ≥ SINRThr
⎫
⎪
⎪
⎬
⎪
⎪
⎭

1 3
equal to one. However, these functions and the weights of each function are explained in
details on the following two subsections respectively.
3.1 The Proposed Functions
3.1.1 A function based on SINR
A function of SINR represents the differences between the target and serving SINR ratios,
which may be expressed by the following formula:
where Max_SINR represents the maximum SINR that can be resulted at the UE. For sim-
plicity Max_SINR is set to 30 dB. SINRS and SINRT represent the SINR over the serving
and target CCs respectively. With the advent of CA technology in LTE-Advanced system
more than one CC can be paired to one UE simultaneously. One CC is configured as Pri-
mary Component Carrier (PCC) and one or more CCs can be configured as Secondary
Component Carriers (SCCs). So, SINRS represents the SINR over the serving PCC only,
while SINRT represents the SINR over the best selected CC among the available CCs.
Since the handover procedure can be occur between two CC in the same sector under the
same eNB to change the PCC, so the target CC can be the serving SCC. Thus, in this case
the SINRT will be the SINR over the serving SCC
(
SINRS−SCC
)

. On the other hand, if the
handover procedure is performed between two sectors under the same eNB or between two
different eNBs the target CC will be the best selected CC among the available CCs, which
can be CC1 or CC2. Thus, in this case the SINRT will be the SINR over the best selected
target CC
(
SINRbT−CC
)

. Therefore, for simplicity SINRT may be simplified by the following
expression:
where SecS and SecT represent the serving and target sectors respectively, while
eNBS and eNBT represent the serving and target eNBs respectively.
3.1.2 A function based on Traffic Loads
The function based on traffic loads is expressed by f(TL)
, which represents the differences
between the target and serving load ratio. The Target load ratio is defined as a ratio of the
occupant target eNB’s load to the maximum eNB’s traffic load capacity
(
TLmax
)
. Similarly,
serving load ratio is defined as a ratio of the occupant serving eNB’s load to the maxi-
mum eNB’s traffic load capacity
(
TLmax
)

. Thereby, the function based on traffic load ratios
(f(TL)) can be simplified by the following expression:
where TLT and TLS represent occupant target and serving traffic loads respectively.
(2)
f(SINR) =
(
SINRT
Max_SINR
)
−
(
SINRS
Max_SINR
)
=
SINRT − SINRS
Max_SINR
(3)
SINRT =
{
SINRS−SCC if eNBT = eNBS and SecT = SecS
SINRbT−CC if
(
eNBT ≠ eNBS or SecT ≠ SecS
)
}
(4)
f(TL) =
(
TLT
TLmax
)
−
(
TLS
TLmax
)
=
TLT − TLS
TLmax

I. Shayea et al.
1 3
3.1.3 A function based on Velocity
The function based on velocity is expressed by f(v)
, which is evaluated as a function of UE’s
movement speed v. Higher movement speeds (v) will lead to increase the f(v) maximum up
1, while the lower movement speeds lead to decrease the f(v) minimum to − 1. However, we
may simplify the function of velocity (f(v))(f(v)) using the following expression:
where vmax represents the maximum expected velocity by UE, which is assumed to be con-
stant (
vmax = 200 kmph) in this paper.
3.2 The Proposed Weight Model
In [12, 33–37] different adaptive handover algorithms have been proposed for adjusting
the handover margin level based on Weight Functions. Each weight function considers dif-
ferent normalized functions. The weight of each normalized function is considered in order
to increase the weight of the significant function to estimate an accurate margin level. The
authors did not formulate any mathematical expression to illustrate how the weight of each
normalized function is assigned. Moreover, the user velocity, eNB load and user SINR are
frequently changed. Therefore, there is a need for a mathematical model to estimate the weight
of each normalized function considered in the weight function. Because of that, in this paper
a mathematical model is formulated to meet the target, which is expressed by the following
formula:
where 𝜔n represents the weight of function n, which can be one of the functions:
SINR, TL or v . f(xn), is the corresponding function n that needs to evaluate its weight. It is
also can be one of the functions of SINR, TL or v . F is a metric’s factor, which represents
the total numbers of parameters that are considered for adapting HOM level. In this paper,
we set F = 3 because we considered only SINR, TL and v factors. f(xi) is the function of x
that corresponding to i, whereas i is varied from 1 to F. For simplicity, we define f(x1) as
f(SINR) , while f(x2) as f(TL) and f(x3) as f(v)
. For example, to evaluate the weight of
function f(SINR)
, it can be evaluated as: 𝜔SINR = 1−f(SINR)
(1−f(SINR))+(1−f(TL))+(1−f(𝜈))
.
Consequently, the HOM level may be adaptively estimated using the following expression:
4 System Model
4.1 System Layout Model
The LTE-Advanced system model is shown in Fig. 2 based on 3GPP specifications that
were introduced in [38]. The network consists of 61 macro-hexagonal cell layout models
with an inter-site-distance of 500 m for each cell. Every hexagonal cell contains one eNB
(5)
f(v) = 2log2
(
1 +
v
vmax
)
− 1
(6)
𝜔n =
1 − f
�
xn
�
∑F
i=1
�
1 − f
�
xi
��
(7)
HOM = MAvg + fAHOM(SINR, TL, v)

1 3
located at its centre and each cell divided into three sectors. Two contiguous CCs are con-
figured in each sector. Two CA Deployment Scenarios are considered, as defined by (1) CA
Deployment Scenario number one (CADS-1) as shown in Fig. 3a and (2) Coordinated Con-
tiguous—CA Deployment Scenario (CC-CADS) as shown in Fig. 3b [39–41]. In CADS-1
and CC-CADS both CCs are operating on contiguous bands with operating frequencies of
2 GHz and 2.0203 GHz for CC1 and CC2 respectively. The Frequency Reuse Factor (FRF)
is assumed to be one. In CADS-1, the antennas of both CCs are pointed toward the same
side of the hexagonal cell per Fig. 4a. The beam directions for both antennas in sectors 1,
2, and 3 are aimed with beam angles of 45°, 180° and 300°, respectively, as illustrated in
Fig. 4a. In CC-CADS, the antenna of each CC is pointed toward a different flat side of the
hexagonal cell as shown in Fig. 4b. Therefore, the main beam of each CC is directed in a
different direction. The beam directions for antenna 1 in sectors 1, 2 and 3 are aimed with
beam angles of 30°, 150° and 270°, respectively, and the beam directions for antenna 2 in
sectors 1, 2 and 3 are aimed with beam angles of 90°, 210° and 330°, respectively, as illus-
trated in Fig. 4b.
The transmitted power from the eNB over each CC is assumed to be the same. As
regards to the users, random numbers of UEs are generated and removed randomly at ran-
dom uniform positions in the serving and target cells. This random generation and removal
of UEs is intended to mimic a random generation of traffic in the simulation. The UEs’
directional movements are selected randomly with a fixed speed throughout the simulation,
which contains five different mobile speed scenarios (30, 60, 90, 120 and 140 km/h). The
mobility movement of all users is considered to occur in the first 37 cells only. Thus, when
the UE moves from the serving to the target eNBs, considering Random Waypoint Model,
it should be surrounded by six eNBs. These six eNBs are considered to be the stations that
cause the interference signals for the user. Moreover, the Adaptive Modulation and Coding
-4000 -3000 -2000 -1000 0 1000 2000 3000 4000
-4000
-3000
-2000
-1000
0
1000
2000
3000
4000
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
eNB-to-UE X location [m]
eNB-to-UE
Y
location
[m]
UE
eNB
Fig. 2 LTE-Advanced system model

I. Shayea et al.
1 3
(AMC) scheme is considered based on the sets of Modulation Schemes (MS) and Coding
Rate (CR) that were introduced in [42, 43]. In addition, detailed models for the handover
procedure for LTE, the Radio-Link-Failure (RLF) detection, the re-establishment proce-
dure, and the Non-Access Stratum (NAS) recovery procedure are considered through the
simulation in order to achieve accuracy in the high performance evaluation. The essential
parameters that are used in this paper are listed in Table 1 based on an LTE-Advanced sys-
tem profile that was defined by 3GPP specifications in [38, 42, 43].
F1 F2
F1 F2
(a) CADS - 1
(b) CC-CADS
Fig. 3 Contiguous CA Deployment Scenarios
Sector # 1
CC1
B - Angle = 2700
CC1
B - Angle = 300
CC2
B - Angle = 900
CC1
B - Angle = 1500
CC2
B – Angle = 3300
Sector # 2
Sector # 3
Sector # 1
CC1 and CC2
Beam Angle = 450
CC1 and CC2
Beam Angle = 3000
CC1 and CC2
Beam Angle = 1800 Sector # 2
Sector # 3
CC2
B - Angle = 2200
(a) (b)
Fig. 4 Beam direction of CC1 and CC2

1 3
4.2 Handover Scenario
The advent of CA technology in LTE—Advanced system (Rel.10 to Rel.13) has increased the
number of Handover Scenarios (HO-Ss) comparable to LTE Release 8 and 9 (Rel.8 Rel.9).
However, there are five handover scenarios which may occur through the users’ mobility in the
CA environment, as illustrated in Fig. 5. In more detail, these scenarios can be explained by:
(1) HO between CCs at the same sector and same eNB (2) HO between sectors at same eNB,
while the target and serving CCs are operating on the same frequencies, (3) HO between sec-
tors at same eNB, while the target and serving CCs are differentiated from each other, (4) HO
between eNBs, while the serving and target CCs are operating on the same frequencies and (5)
HO between eNBs, while the serving and target CCs are differentiated from each other. All
these handover scenarios are considered in these papers.
The handover decision is taken based on serving and target RSRPs qualities. Once the tar-
get RSRP becomes greater than the serving RSRP by the HOM level during the trigger period
of time (Time-To-Trigger (TTT)), the serving eNB performes a true handover decision and
sends the handover request message to the target eNB. Thus, the considered handover decision
in this paper can be expressed by the following:
(8)
RSRPT ≥
(
RSRPS + HOM
)
Table 1 Simulation parameters
Parameter Assumption
Cellular layout Hexagonal grid, 61 cell sites, 3 sectors
per cell site, 2 CCs per sector
Minimum distance between UE and eNB ≥ 35 m
Total eNB TX power 46 dBm per CC
Shadowing standard deviation 8 dB
White noise power density (Nt) − 174 dBm/Hz
eNBs noise figure 5 dB
Thermal noise power NP = Nt + 10 log (BW × 106) dB
UE noise figure 9 dB
Operation carrier bandwidth 20 MHz for each, carrier PCC and SCC
Total system bandwidth 40 MHz (2CCs × 20 MHz)
Number of PRB/CC 100 PRB/CC
Number subcarriers/RB 12 Subcarriers per RB
Number of OFDM symbols per subframe 7
Sub-carrier spacing 15 kHz
Resource block bandwidth 180 kHz
Q_rxlevmin − 101.5 dB
Measurement Interval 50 ms for PCC and SCC
Time-to-Trigger (TTT) 300 ms
Max HO margin 6 dB
Each X2-interface delay 10 ms
Each eNB process delay 10 ms
T311 10 s

I. Shayea et al.
1 3
Once the handover decision becomes true, the serving eNB starts for preparing hando-
ver by sending a Handover Request message to the target eNB; thus, the UE will enter the
handover procedure to establish connection with the target eNB. The handover procedure
is performed based on the handover procedure that has been introduced in LTE-Advanced
system in [44]. However, once the target eNB receives the Handover Request message, it
will start admission control. If the admission control decision is true, the target eNB will
send a Handover Request Acknowledge to the serving eNB, which in turn will begin DL
allocation. Thus, once the UE receives the Radio Resource Control (RRC)—Connection-
Reconfiguration message with the necessary parameters, it will begin to execute the hando-
ver to the target eNB.
5 Results and Discussion
In this section, the performance of the proposed AHOM-NWF algorithm is explained and
compared with other algorithms from the literature. The AHOM-NWF algorithm is com-
pared to (1) Fixed HOM (2) AHOM-D and (3) AHOM-CF. The AHOM-NWF and all the
comparative algorithms are implemented based on a conventional handover decision algo-
rithm →
(
RSRPT ≥
(
RSRPS + HOM
))
, where HOM represent the margin level, which is
the focus of this study as has been discussed in section II. The results are presented based
on two contiguous CA deployment scenarios (CADS-1 and CC-CADS) with different per-
formance metrics. Figures 6, 7 and 8 show the SINR, user’s cell edge spectral efficiency,
and outage probability, respectively, based on different handover margin algorithms with
two different CADSs. In Figs. 6 and 7, the results are presented as a Cumulative Distributed
Fig. 5 Handover Scenarios with the advent of CA technology

1 3
Probability Function (CDF), while in Fig. 8 user’s outage probability is presented versus
different mobile speed scenarios. The evaluation performances of SINR and spectral effi-
ciency are performed based on the evaluation that are analysed in [45], while the outage
probability is evaluated based on the evaluation method that is introduced in [46].
In Fig. 6a, AHOM-NWF achieves around 29.8, 14 and 6.3% as average gains of SINR
based on CADS-1 over the legacy decision algorithm based on Fixed-HOM, AHOM-D
and AHOM-CF respectively. While in Fig. 6b, AHOM-NWF achieves around 18.7, 2.7
and 2.3% as average gains of SINR based on CC-CADS over the legacy decision algorithm
based on Fixed-HOM, AHOM-D and AHOM-CF respectively.
In Fig. 7a, the cell edge spectral efficiency of AHOM-NWF based on CADS-1 can reach
up to 2.5 bps/Hz which shows significance improvement compare to Fixed-HOM, AHOM-
D and AHOM-CF with average gain of 30%, 4.4% and 3%, respectively. The same perfor-
mance is achieved for CC-CADS deployment, where AHOM-NWF has the higher spectral
efficiency of 3.25 bps/Hz among others algorithm as depicted in Fig. 7b. It achieves around
28.5%, 4.5% and 3.8% as average gains over the legacy decision algorithm based on Fixed-
HOM, AHOM-D and AHOM-CF respectively.
(a) (b)
CADS-1 CC-CADS
-20 -15 -10 -5 0 5 10 15 20 25
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
SINR [dB]
SINR
Probability[dB]
Empirical CDF
Fixed-HOM
AHOM-D
AHOM-CF
AHOM-NCF
-10 -5 0 5 10 15
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
SINR [dB]
SINR
Probability[dB]
Empirical CDF
Fixed-HOM
AHOM-D
AHOM-CF
AHOM-NCF
Fig. 6 SINR based on different handover algorithms with two different CADSs
CADS-1 CC-CADS
0 0.5 1 1.5 2 2.5 3
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Cell Edge Spectral Efficeincy [bps/Hz]
Cell
Edge
Spectral
Efficeincy
Probability
Cell Edge Spectral Efficeincy
Fixed-HOM
AHOM-D
AHOM-CF
AHOM-NCF
0.5 1 1.5 2 2.5 3 3.5
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Cell Edge Spectral Efficeincy [bps/Hz]
Cell
Edge
Spectral
Efficeincy
Probability
Cell Edge Spectral Efficeincy
Fixed-HOM
AHOM-D
AHOM-CF
AHOM-NCF
(a) (b)
Fig. 7 User’s cell-edge spectral efficiency

I. Shayea et al.
1 3
Figure 8a, b shows that the outage probability of all algorithms increased as the mobile
speed increase. The proposed algorithm, AHOM-NWF is less than FHOM algorithms with
63.7% and 65% as reduction gains of UE’s outage probability for CADS-1 and CC_CADS,
respectively. Compare to AHOM-D and AHOD-CF algorithms, the AHOM-NWF has the
reduction gains of only 9.2% and 14.7%, respectively for CADS-1 algorithm and 3% and
7%, respectively for CC-CADS algorithm.
Consequently, it can be stated that, AHOM-NWF achieves a significant enhancement
gains compare to the others algorithms. The total average enhancement gains that are
achieved by AHOM-NWF around 24.4, 14.6 and 17.9% over the legacy algorithm based
on Fixed-HOM, AHOD-D and AHOD-CF respectively. Thus, as these enhancements will
supports for service continuity and enhancing service quality. These achievable enhance-
ments by AHOM-NWF are mainly due to the three proposed functions, which are f(SINR) ,
traffic load, f(TL) and UE’s velocity, f(v)
. These parameters are influence factors, which
are contributing for taking a proper handover decision. The variations of these parame-
ters are mainly affecting the system performance inversely or extrusive. So that, adjusting
handover margin level based on these factors is particularly useful for taking the proper
HO decision. Thus, the effects of these three parameters on the estimated handover margin
level are further explained in the following paragraphs.
SINR is used as parameter for adjusting HOM level, which is considered sufficient as
a factor instead of channel condition, distance (d), noise and interferences. Since RSS is
evaluated as a function of channel condition and distance, while SINR is evaluated as a
ratio of serving RSS to the neighbours interferences plus noise ratio, thus adapting HOM
based on SINR parameter is more comprehensive than adapting HOM based on any single
parameters only. That contributes for estimating more a suitable HOM level than the other
algorithms. However, the formulated f(SINR) in (2), which is representing the difference
between the target and serving SINR ratios. f(SINR) is automatically varies between {− 1}
and {1} based on the serving and target SINR qualities. For the case of serving SINR is
better than the target SINR quality, the function of SINR will increased (f(SINR) 0)
whereupon the estimated HOM level by (7) become high, which in turn leads to prevent
the unnecessary handover. Because of that, the user’s connection is alaways connected with
the best serving eNB.In the other case, when the target SINR quality become better than
the serving SINR quality that leads to decrease f(SINR) 0
. Whereupon the estimated
CADS-1 CC-CADS
20 40 60 80 100 120 140 160
10
-2
10
-1
10
0
Mobile Speed [km/houre]
Outage
Probability
Fixed-HOM
AHOM-D
AHOM-CF
AHOM-NCF
20 40 60 80 100 120 140 160
10
-2
10
-1
10
0
Mobile Speed [km/houre]
Outage
Probability
Fixed-HOM
AHOM-D
AHOM-CF
AHOM-NCF
(a) (b)
Fig. 8 User’s Outage Probability versus Mobile Speed Scenarios

1 3
HOM level become small, which in turn leads to an advanced handover procedure to the
best target eNB.
Traffic Load is used as parameter for adjusting HOM level which is considered as a suf-
ficient factor since it contributes for balancing loads between CCs and neighbouring eNBs.
However, a mathematical function is simplified as a function of Traffic Load f(TL) in (4),
which is representing the difference between the target and serving loads ratios. f(TL) is
automatically varies between {−
1} and {1} based on the serving and target traffic loads.
In case if the traffic load of serving eNB is less than the traffic load of target eNB, that
leads to increases f(TL)
, which will become greater than zero, whereupon the estimated
HOM level by (7) will be high, which in turn leads to prevent unnecessary handover. This
leads to keep the user’s connection with the serving eNB, which is considered whereupon
the best eNB since it has more resource available. In the other case, when the target traf-
fic loads become less than the serving traffic loads that leads to decrease f(TL) , which
will become smaller than zero, whereupon the estimated HOM level will be small, which
in turn leads to take an early handover decision. This leads to an advanced handover pro-
cedure to the target eNB, which is considered as the best eNB as it enjoys more available
resources. Consequently, traffic load is a useful factor for adjusting margin level, contribut-
ing to an increase in HOM level when the load of target eNB is increased and it contributes
for decreasing HOM level when the load of target eNB is decreased. That leads to take a
proper handover decision, which in turn balancing the loads between CCs and eNBs as
well as the available resources between UEs. It also enhanced user’s SINR, spectral effi-
ciency, and reduced outage probability because the user’s connection with eNB that has
more resource available can be reduced.
UE’s velocity is considered in adjusting margin level. It is a useful consideration since
it can contribute to adjusting the margin level based on UE’s velocity, which can cause pre-
vention of unnecessary handover procedure that maybe occur at high UE movement speeds.
However, a mathematical expression is formulated as a function of UE’s velocity f(v) as in
(5), which can be decreased when the UE’s velocity is decreased and it is increased when
the UE’s movement speed is increased. In case of low mobility speeds a lower level of f(v)
is resulted, which leads to decrease the estimated margin level. That leads to perform an
early handover to the best target eNB when it is needed. On the other hand, when the UE’s
speed is increased f(v) will be increased, which in turn leads to increase HOM level. Thus,
the unnecessary handover that can be resulted by high movement speeds can be prevented.
That leads to prevent resource waste. Thus, considering UE’s velocity for adjusting the
margin level can contribute for enhancing user’s SINR, spectral efficiency and reducing the
user’s outage probability.
6 Conclusions
In this paper, AHOM-NWF has been proposed based on several parameters such as SINR
quality, traffic load and UE’s velocity. Mathematical expression has been formulated for
adjusting margin level based on three functions, which are evaluated as functions of SINR
quality, traffic load and UE’s velocity. Also, a mathematic model for estimating the weight
of each normalized function has been proposed. Simulation results demonstrate that the
proposed AHOM-NWF is optimal from the perspective of User’s SINR, cell edge through-
put enhancement and outage probability reduction compared to Fixed HOM, AHOM-D
and AHOM-CF through CA operation in LTE-Advanced system. Thus, AHOM-NWF

I. Shayea et al.
1 3
contributes to estimation of a suitable HOM level, which leads to a proper handover deci-
sion. That has allowed the UE to remain connected to the best eNB that either provides
a better SINR quality with availability of resources, or that has more resources available
with acceptable SINR level. Furthermore, it prevents unnecessary handover that may result
from a high UE’s velocity.
References
1. Sawant, S. S., Vernekar, N. K. (2013). Adaptive distance handover scheme in mobile WiMax. Inter-
national Journal on Advanced Computer Theory and Engineering (IJACTE), 2(3), 87–91.
2. Itoh, K. I., Watanabe, S., Shih, J. S., Sato, T. (2002). Performance of handoff algorithm based on
distance and RSSI measurements. IEEE Transactions on Vehicular Technology, 51(6), 1460–1468.
3. Halgamuge, M. N., Hai, V. L., Rarnamohanarao, K., Zukerman, M. (2005). Signal-based evaluation
of handoff algorithms. IEEE Communications Letters, 9(9), 790–792.
4. Pollini, G. P. (1996). Trends in handover design. IEEE Communications Magazine, 34(3), 82–90.
5. Kemeng, Y., Gondal, I., Qiu, B., Dooley, L. S. (2007). Combined SINR based vertical handoff algo-
rithm for next generation heterogeneous wireless networks. In Proc. of 7th Int. IEEE global telecom-
munications conference (GLOBECOM ‘07) (pp. 4483–4487), November 26–30, 2007.
6. Kemeng, Y., Qiu, B., Dooley, L. S. (2007). Using SINR as vertical handoff criteria in multimedia
wireless networks. In Proc. of IEEE int. conference on multimedia and expo (pp 967–970), July 2–5,
2007.
7. Ayyappan, K. Narasimman, K., Dananjayan, P. (2009). SINR based vertical handoff scheme for QoS
in heterogeneous wireless networks. In Proc. of 1st int. conference on future computer and communi-
cation (ICFCC 2009) (pp. 117–121), April 3–5, 2009.
8. Bathich, A. A., Baba, M. D., Rahman, R. (2011). SINR based media independent handover in
WiMAX and WLAN networks. In Proc. of IEEE int conference on computer applications and indus-
trial electronics (ICCAIE2011) (pp 331–334), December 4–7, 2011.
9. Hyun-Ho, C. (2010). An optimal handover decision for throughput enhancement. IEEE Communica-
tions Letters, 14(9), 851–853.

10. Huamin, Z., Kyung-sup, K. (2007). Performance analysis of an adaptive handoff algorithm based on
distance information. Computer Communications, 30(6), 1278–1288.

11. Lal, S., Panwar, D. K. (2007) Coverage analysis of handoff algorithm with adaptive hysteresis
margin. In Proc. of 10th. IEEE int conference on information technology, (ICIT 2007) (pp 133–138),
December 17–20, 2007.

12. Lee, D.-W., Gil, G.-T., Kim, D.-H. (2010). A cost-based adaptive handover hysteresis scheme to
minimize the handover failure rate in 3GPP LTE system. EURASIP Journal on Wireless Communica-
tions and Networking, 2010(1), 1–7.

13. Huang, Y.-F., Chen, H.-C., Chu, H.-C., Liaw, J.-J., Gao, F.-B. (2010). Performance of adaptive hys-
teresis vertical handoff scheme for heterogeneous mobile communication networks. Journal of Net-
works, 5(8), 977–983.

14. Zhu, H., Kwak, K. S. (2006). An adaptive hard handoff algorithm for mobile cellular communica-
tion systems. ETRI Journal, 28(5), 676–679.

15. Sinclair, N., Harle, D., Glover, I. A., Irvine, J., Atkinson, R. C. (2013). An advanced SOM algorithm
applied to handover management within LTE. IEEE Transactions on Vehicular Technology, 62(5),
1883–1894.

16. Ghanem, K., Alradwan, H., Motermawy, A., Ahmad, A. (2012) Reducing ping-pong handover
effects in intra EUTRA networks. In Proc. of 8th int. symposium on communication systems, networks
digital signal processing (CSNDSP2012) (pp. 1–5) July 18–20, 2012.

17. Yifan, Z., Muqing, W., Shunming, G., Linlin, L., Ankang, Z. (2012). Optimization of time-to-trigger
parameter on handover performance in LTE high-speed railway networks. In Proc. of 15th int. sym-
posium on wireless personal multimedia communications (WPMC-2012) (pp. 251–255), September
24–27, 2012.

18. Ewe, L., Bakker, H. (2011). Base station distributed handover optimization in LTE self-organiz-
ing networks. In Proc. of 15th int. symposium on personal indoor and mobile radio communications
(PIMRC 2011) (pp. 243–247), September 11–14, 2011.

19. GPP Team (2011). Evolved universal terrestrial radio access network; Self-configuring and self-opti-
mizing network (SON) use cases and solutions (Release 9). In TR 36.902 V9.3.1. http//:ww.3gpp.org/.

1 3

20. Legg, P., Gao, H., Johansson, J. (2010). A simulation study of LTE intra-frequency handover
performance. In Proc. of 72nd int. IEEE vehicular technology conference fall (VTC 2010-Fall) (pp.
1–5), September 6–9, 2010.

21. Zhenzhen, W. (2010). Mobility robustness optimization based on UE mobility for LTE system. In
Proc. about int. conference on wireless communications and signal processing (WCSP 2010) (pp.
1–5), October 21–23, 2010.

22. Haijun, Z., Xiangming, W., Bo, W., Wei, Z., Yong, S. () A novel handover mechanism between
Femtocell and Macrocell for LTE based networks. In 2010 Proc. of 2nd int. conference on commu-
nication software and networks, (ICCSN ‘10) (pp. 228–231), February 26–28, 2010.

23. Hao, C., Liu, H., Zhan, J. (2009). A velocity-adaptive handover scheme for mobile WiMAX.
International Journal of Communications Network System Sciences (IJCNS), 2(9), 874–878.

24. Anwar, M. I., Khosla, A., Sood, N. (2010). A mobility improvement handover scheme for
mobile-WiMAX. International Journal of Computer Applications, 11(3), 28–31.

25. Haijun, Z., Xiangming, W., Bo, W., Wei, Z., Zhaoming, L. (2009). A novel self-optimizing hand-
over mechanism for multi-service provisioning in LTE-advanced. In Proc. of IEEE int. conference
on research challenges in computer science (ICRCCS ‘09) (pp. 221–224), December 28–29, 2009.

26. Nasri, R., Altman, Z. (2007). Handover adaptation for dynamic load balancing in 3GPP long
term evolution systems. In Proc. of int. conference on advances in mobile computing multimedia
(MoMM’07) (pp. 145–153).

27. Yang, F., et al. (2015). Handover optimization algorithm in LTE high-speed railway environment.
Wireless Personal Communications, 84, 577–1589.

28. Park, M.-H., Joo, Y.-I. (2015). Efficient Handover Strategy for Inbound Mobility to LTE Small
Cell. Wireless Personal Communications, 82, 1435–1447.

29. Vu, T.-T., Decreusefond, L., Martins, P. (2014). An analytical model for evaluating outage
and handover probability of cellular wireless networks. Wireless Personal Communications, 74,
1117–1127.

30. Lim, J., Hong, D. (2013). Mobility and handover management for heterogeneous networks in
LTE-advanced. Wireless Personal Communications, 72, 2901–2912.

31. Munoz, P., Barco, R., de la Bandera, I. (2013). On the Potential of Handover Parameter Optimiza-
tion for Self-Organizing Networks. IEEE Transactions on Vehicular Technology, 62(5), 1895–1905.

32. Saeed, M., El-Ghoneimy, M., Kamal, H. (2017). An enhanced fuzzy logic optimization tech-
nique based on user mobility for LTE handover. In 2017 34th national radio science conference
(NRSC), Alexandria (pp. 230–237).

33. Nasser, N., Hasswa, A., Hassanein, H. (2006). Handoffs in fourth generation heterogeneous net-
works. IEEE Communications Magazine, 44(10), 96–103.

34. Zhu, F., Mc Nair, J. (2006). Multiservice vertical handoff decision algorithms. EURASIP Journal
on Wireless Communications and Networking, 2006, 1–13.

35. Dongyeon, L., Youngnam, H. Jinyup, H. (2006) QoS-based vertical handoff decision algorithm
in heterogeneous systems. In Proc. of IEEE 17th int. symposium on personal, indoor and mobile
radio communications (pp. 1–5), September 11–14, 2006.

36. SuKyoung, L., Sriram, K., Kyungsoo, K., Yoon, H. K., Golmie, N. (2009). Vertical handoff deci-
sion algorithms for providing optimized performance in heterogeneous wireless networks. IEEE
Transactions on Vehicular Technology, 58(2), 865–881.

37. Rizvi, S., Aziz, A. Saad, N. M. (2010) Optimizations in vertical handoff decision algorithms for
real time services. In Proc. of int. conference on intelligent and advanced systems (ICIAS 2010)
(pp. 1–4), June 15–17 2010.

38. GPP Team (2016). Evolved Universal Terrestrial Radio Access (E-UTRA); Radio Frequency (RF)
system scenarios (Release 11). In 3GPP TR 36.942 V13.0.0 (2016-01). http://www.3gpp.org/.

39. Iwamura, M., Etemad, K., Mo-Han, F., Nory, R., Love, R. (2010). Carrier aggregation framework in
3GPP LTE-advanced [WiMAX/LTE update]. IEEE Communications Magazine, 48(8), 60–67.

40. GPP Team (2010). Simulation assumptions for Mobility performance in Carrier Aggregation. In
R4-102114 NTT DOCOMO, http://www.3gpp.org/.

41. GPP Team (2010). Carrier aggregation deployment scenarios. In R2-102490 (pp. 1–3). http://
www.3gpp.org/.

42. GPP Team (2017). Evolved Universal Terrestrial Radio Access (E-UTRA); Physical channels and
modulation (Release 11). In 3GPP TS 36.211 V13.6.0 (2017-06) (pp. 1–171). http://www.3gpp.org/.

43. GPP Team (2017) Evolved Universal Terrestrial Radio Access (E-UTRA); LTE physical layer;
General description (Release 11). In 3GPP TS 36.201 V13.3.0 (2017-03). http://www.3gpp.org/.

44. GPP Team (2014). Evolved Universal Terrestrial Radio Access; Overall description (Release 11).
In 3GPP TS 36.300 V13.8.0 (2017-06). http://www.3gpp.org/.

I. Shayea et al.
1 3

45. Shayea, I., Ismail, M., Nordin, R. (2013) Downlink spectral efficiency evaluation with carrier
aggregation in LTE-advanced system employing adaptive modulation and coding schemes. In Proc.
of IEEE Malaysia int. conference on communications (MICC2013) (pp. 98–103), November 26–28,
2013.

46. Tjeng, T. T., Chin Choy, C., Xiaodai, D. (1997). Outage probability for lognormal-shadowed
Rician channels. IEEE Transactions on Vehicular Technology, 46(2), 400–407.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Ibraheem Shayea received the B.Sc. degree in Electronic Engineering
from the University of Diyala, Baqubah, Iraq, in 2004, and the M.Sc.
degree in Computer and Communication Engineering and the Ph.D.
degree in Mobile Communication Engineering from The National Uni-
versity of Malaysia, Universiti Kebangsaan Malaysia (UKM), Malay-
sia, in 2010 and 2015, respectively. Since the 1st of January 2011 until
28 February 2014 he worked as Research and Teaching Assistant at
Universiti Kebangsaan Malaysia (UKM), Malaysia. Then, from the 1st
of January 2016 until 30 Jun 2018, he joined Wireless Communication
Center (WCC), University of Technology Malaysia (UTM), Malaysia,
and worked there as a Research Fellow. He is currently working as a
Researcher Fellow at Istanbul Technical University (ITU), Istanbul,
Turkey, since the 1st of September 2018 until now.
Mahamod Ismail received the B.Sc. degree in Electrical and Electron-
ics from University of Strathclyde, U.K. in 1985, the M.Sc. degree in
Communication Engineering and Digital Electronics from UMIST,
Manchester U.K. in 1987, and the Ph.D. from University of Bradford,
U.K. He is currently a Professor with the Department of Electrical,
Electronics and System Engineering, and attach to the Center for
Information Technology as the Deputy Director (Research and Educa-
tion), Universiti Kebangsaan Malaysia. In 1997–1998 he was with the
team engineer building the first Malaysian microsatellite Tiungsat in
Surrey Satellite Technology Ltd., United Kingdom. He became the
Guest Professor in University of Duisburg Essen (formerly known as
Gerhard Mercator Universitat Duisburg), Germany in summer semes-
ter 2002. His research interests include mobile communication and
wireless networking. He published more than 200 technical papers in
journal and proceeding at local and international level. He is the past
chapter chair of IEEE Communication Society, Malaysia and Educa-
tional Activities chairman, IEEE Malaysia Section and currently the
committee member for the Joint chapter Communication and Vehicular Technology Society, IEEE Malay-
sia. He is also actively involved in conference and became the Technical Program Chairman, technical com-
mittee and paper reviewer.

1 3
Rosdiadee Nordin received the B.Sc. degree in Electrical and Elec-
tronics from Electrical and Electronics Engineering department, Uni-
versiti Kebangsaan Malaysia, Malaysia, July, 2001, and the Doctor of
Philosophy (Ph.D.), in Wireless Engineering from University of Bris-
tol, United Kingdom, January, 2011. He is currently a Ass. Professor
with the Department of Electrical, Electronics and System Engineer-
ing, Universiti Kebangsaan Malaysia.
Prof. Dr. Mustafa Ergen is professor of Electrical Engineering in Istan-
bul Technical University, president of venture funded Ambeent Inc.
focusing 5G and Artificial Intelligence plus Chief Technology Advisor
in Türk Telekom. Previously, Mustafa co-founded Silicon Valley
startup WiChorus Inc. to focus on 4G technologies and company is
acquired by Tellabs [now Coriant] for $200 M. Previously, he was a
National Semiconductor Fellow [now TI] at the University of Califor-
nia Berkeley, where he co-founded the Distributed Sensing Lab, focus-
ing on statistical sensor intelligence and vehicular communication.
Mustafa completed four programs from UC Berkeley: Ph.D. and MS
degrees in electrical engineering, MA degree from international stud-
ies and MOT program from HAAS Business School. His BS is from
electrical engineering as Valedictorian from Orta Dogu Technical Uni-
versity with 4.0/4.0 GPA. Prof. Dr. Ergen has more than 40 patent
applications, many publications and authored three books: Girisimci
Kapital: Silikon Vadisi Tarihi ve Startup Ekonomisi (2nd Edition—
KUY, 2017) 移动宽带系统—包括 WiMAX 和 LTE (PHEi, 2011),
Mobile Broadband: Including WiMAX and LTE (Springer, 2009), Multi Carrier Digital Communications:
Theory and Applications of OFDM (Springer, 2004). He is national delegate in 5G Infrastructure Associa-
tion and Horizon2020 ICT Funding Programs of European Union and advisor at Berkeley Program on
Entrepreneurship and Development. He is also an adjunct associate professor at Koç University. He also
served in the board of trustees of TOBB University of Economics and Technology and was cohost in TV
show on BloombergHT about entrepreneurship.
Dr. Norulhusna Ahmad graduated from Universiti Teknologi Malaysia
(UTM) in 2001 with B.Sc. in Electrical Engineering. She joined UTM
as a staff and later pursuing her study at the same university. She
received her Master degree of Electrical Engineering (Telecommuni-
cation) and Ph.D. in Electrical Engineering in 2003 and 2014, respec-
tively. Currently, she is a senior lecturer at Razak Faculty of Technol-
ogy and Informatics, UTM KL. Her expertise is on the area of signal
processing in wireless communication, iterative decoding and error
control coding. Her research interest is on emerging communication,
cognitive radio, internet of things, rural communication and communi-
cation in disaster management.

I. Shayea et al.
1 3
Nor Fadzilah Abdullah received an M.Sc. in Communications Engi-
neering from University of Manchester, UK and a B.Sc. in Electrical
and Electronics degree from Universiti Teknologi Malaysia, in 2003
and 2001 respectively. She has worked with major telecommunication
companies such as Ericsson Malaysia and Maxis Communications
Berhad, Malaysia between 2003 and 2008. She received a Ph.D. stu-
dent from the Centre for Communications Research at the University
of Bristol and sponsored by Malaysian Ministry of Higher Education
and Universiti Kebangsaan Malaysia in 2015. She is currently a Senior
lecturer with the Department of Electrical, Electronics and System
Engineering, Universiti Kebangsaan Malaysia.
A. Alhammadi received his BE in Electronic majoring in telecommu-
nications and M.S degree in wireless communication from Multimedia
University, Malaysia, in 2011 and 2015, respectively. He is currently
serving as a research scholar at Multimedia University since 2012. He
is the author of more than 20 papers in international journals and con-
ferences. His main research interests are in heterogeneous networks,
mobility management, D2D communication, cognitive radio networks,
localization, propagation modelling. He is a member of professional
institutes and societies such as IEEE, IEICE, IACSIT and IAENG. He
is also a member of more than ten program committees at international
conferences or workshops.
Hafizal Mohamad received the BE with First Class Honours and Ph.D.
in Electronic Engineering from University of Southampton, UK in
1998 and 2003, respectively. He has been a faculty member at the Mul-
timedia University, Malaysia from 1998. He served a short stint as a
visiting fellow at National Institute of Information and Communication
Technology (NICT), Yokosuka, Japan in 2005. Since 2007, he is a
Senior Staff Researcher at Wireless Communications Cluster, MIMOS
Berhad, where he leads a team of researchers working on cognitive
radio and mesh network. He has published over 50 journal and confer-
ence papers. He has 3 patents granted and 17 patents filed. He is a Sen-
ior Member of IEEE. He was the Vice Chair for IEEE Malaysia Sec-
tion (2013), and the Executive Committee of IEEE Malaysia Section
for Educational Activities (2011–2012). He was the Chair of IEEE
Communication Society and Vehicular Technology Joint Chapter,
Malaysia Section (2009–2011). He has been involved in organizing a
number of conferences since 2005 including; Technical Program Co-
Chairs for APCC 2012 (Jeju, Korea), APCC 2011 (Sabah), MICC
2009 (Kuala Lumpur) and Tutorial Chair for ICT 2007 (Penang).

1 3
Affiliations
Ibraheem Shayea1,2
· Mahamod Ismail1
· Rosdiadee Nordin1
· Mustafa Ergen2
·
Norulhusna Ahmad3
· Nor Fadzilah Abdullah1
· Abdulraqeb Alhammadi4
·
Hafizal Mohamad5
Mahamod Ismail
mahamod@ukm.edu.my
Rosdiadee Nordin
adee@ukm.edu.my
Mustafa Ergen
mustafaergen@itu.edu.tr
Norulhusna Ahmad
norulhusna.kl@utm.my
Nor Fadzilah Abdullah
fadzilah.abdullah@ukm.edu.my
Abdulraqeb Alhammadi
abdulraqeb.alhammadi@gmail.com
Hafizal Mohamad
hafizal.mohamad@mimos.my
1
Department of Electronics, Electrical and System Engineering, Universiti Kebangsaan Malaysia,
43600 Bangi, Selangor, Malaysia
2
Electronics and Communication Engineering Department, Faculty of Electrical and Electronics
Engineering, Istanbul Technical University (ITU), Istanbul, Turkey
3
Razak Faculty of Technology and Informatics, Universiti Teknologi Malaysia (UTM),
Kuala Lumpur, Malaysia
4
Multimedia University (MMU), Cyberjaya, Malaysia
5
MIMOS Berhad, Technology Park Malaysia, 57000 Kuala Lumpur, Malaysia

New weight function for adapting handover margin level over contiguous carrier aggregation deployment scenarios in lte advanced system

More Related Content

What's hot (20)

Similar to New weight function for adapting handover margin level over contiguous carrier aggregation deployment scenarios in lte advanced system (20)

More from Journal Papers (20)

Recently uploaded (20)

New weight function for adapting handover margin level over contiguous carrier aggregation deployment scenarios in lte advanced system