Content uploaded by Prithiviraj Ananth
Author content
All content in this area was uploaded by Prithiviraj Ananth on Aug 20, 2022
Content may be subject to copyright.
Electronic copy available at: https://ssrn.com/abstract=3123532
Proceedings of the International Conference on Intelligent Computing Systems (ICICS 2017 – Dec 15th -16th 2017)
organized by Sona College of Technology, Salem, Tamilnadu, India
Elsevier’s SSRN eLibrary – Journal of Information Systems & eBusiness Network -ISSN: 1556-5068
80
Genetic based approach to Optimize the Vertical
Handover performance among Hetrogenous
Network
M.Saravanan
Dr.A.Prithiviraj
Dept of Computer Science and Engineering
Dept of Computer Science and Engineering
Sona College of Technology
Sona College of Technology
saravananmsv44@gmail.com
prithivi@sonatech.ac.in
Abstract—Mobile terminals typically can be connected
to multiple wireless networks that offer different degrees of
suitability for different classes of service. A genetic algorithm
(GA) - based on a supervised approach - is used to assess the
likelihood of a successful handover in heterogeneous networks to
increase capacity and network performance. Traditional handoff
schemes are prone to ping-pong and corner effects, and the
development of an optimized handover scheme for a seamless,
faster and less energy-consuming handover decision is a
challenge. In order to ensure service continuity and to maintain
the promising QoS, the handover decision should be made
appropriately. The most general criteria for handover is the
received signal strength, which is not sufficient in advanced
networks. The GA scheme can effectively optimize the soft
handoff decision by using the best-fit network for the mobile
terminal (MT), taking into account the quality of service
requirements (QoS), the network parameters, and the user
preference for the cost of the various connection points for the
MT is selected. The mathematical model-based approach to
vertical handoff prediction involves a well-defined objective
function taking into account signal strength, user equipment
speed, load and cost per user bandwidth. The system performance
can be improved according to the user settings by adjusting the
weights in the defined objective function. Due to the high
bandwidth and easy networking, the system consider the wireless
access standards: UMTS, Wi-Fi, LTE and Wimax. The decision
problem is formulated as several objective optimization problems
and simulated using a monitored genetic algorithm. The
simulation result shows that the number of handovers can be
minimized when using optimized network parameter values.
Keywords— Genetic Algorithm, Supervised algorithm, Mobile
Terminal, Quality of Service, Energy-Consuming.
I. INTRODUCTION
The wireless networks that use wireless data links to
connect to network nodes, thereby eliminating the costly
process of inserting cables into a building or connecting
different equipment locations with households,
telecommunications networks, and enterprise (business)
installations. Optimal network selection is a key issue in the
vertical handover decision phase. This multi-criteria nature of
the algorithm allows for the simultaneous consideration of several
significant aspects of the vertical handoff process to improve
system performance in accordance with the defined
heterogeneous network goals. The single criterion vertical
rendering algorithm only reflects the specified network
characteristic that is insufficient to characterize the network QoS.
In this section, various handoff decision algorithms are introduced
to reduce the undesirable number of handovers and improve the
quality of service (QoS) for various networks such as UMTS, Wi-
MAX, Wi-Fi, LTE. The proposed Vertical Handoff (VHO)
algorithm for a heterogeneous network architecture that integrates
both a cellular network and a wireless local area network
(WLAN). Using Markov Decision Process (MDP) to optimize
HO performance. In the RSS-based single-attribute HO decision
is taken into account. The preferred network selection is
determined by connection life using MDP. The awarded total
rewards will be maximized and the number of handcuffs will be
minimized. Another MDP-based cell selection method using the
value iteration algorithm (VIA) is proposed in which dynamic
channel loading and link quality have been considered to reduce
the number of handover and signaling overheads. The vertical
handoff decision algorithm provides support for multiple types of
services (video, voice, file transfer protocol) with different
priorities.
The GA-based specific processing steps used in this
work are summarized below:
(1) Evaluation and Suitability Assignment the objective
function values of the candidate solutions in the current
population are evaluated. The algorithm uses the objective
function values to determine the fitness values of the candidate
solutions in the current population.
(2) Selection The algorithm selects members called parents
based on their suitability. The main idea of the selection is to
favor better solutions for worse ones. There are several
strategies to choose the individuals to copy to the next
generation.
An important issue in breaking down large datasets, both in
terms of dimension and size, is choosing a subset of the original
features. Preprocessing the data to obtain a smaller set of
representative features while maintaining the optimum salient
features of the data not only reduces processing time, but also
Electronic copy available at: https://ssrn.com/abstract=3123532
Proceedings of the International Conference on Intelligent Computing Systems (ICICS 2017 – Dec 15th -16th 2017)
organized by Sona College of Technology, Salem, Tamilnadu, India
Elsevier’s SSRN eLibrary – Journal of Information Systems & eBusiness Network -ISSN: 1556-5068
81
results in greater compactness of the learned models. If class
labels of the data are available, we use a monitored feature
selection, otherwise an unattended function selection is
appropriate. Conventional feature selection techniques include
evaluating various feature subsets using an index and selecting
the best ones among them. The index usually measures the
ability of the respective subsets to classify or group depending
on whether the selection process is supervised or unattended.
In the GA framework, solutions are transformed into coded
forms called chromosomes. As with other optimization
methods, the viability of the candidate solution is assessed by
an objective function. In the GA approach, each solution is
tested by a fitness function that reflects its strength among all
other solutions in the population. Therefore, there is a need to
develop a new optimized handover technique for improved
service delivery in cellular networks. Traditional handover
optimization techniques are not reliable and inexpensive. They
present implementation difficulties and are only suitable for
static situations.
II. RELATED WORK
A. Vertical Handoff Scheme Concering Mobility in the Two-
hierarchy Network
Shuhui Liu, Yongyu Chang , Guangde Wang, Dacheng
Yang was said the Traditional VHO in heterogeneous cell
networks is on the basis of RSS or signal to interference plus
noise ratio (SINR) comparisons, as discussed. However, there are
various other factors need to be considered due to the overlay
nature of heterogeneous networks. The vertical handoff must
evaluated additional factors such as cost, services, network
conditions and user performances too RSS. In the proposed a
VHO function that simply estimates the service quality for
available networks and selects the best one. Whereas, the above
traditional VHO schemes cannot achieve the optimization
because they don’t consider the influence on the other handoff
UEs when implement the vertical handoff decision (VHD) for a
certain UE. For example, at some situations, many mobile devices
may try to handoff to the same femto-cell or macro-cell since the
traditional schemes only consider the individual UE irrespective
of the other handoff requests. Thus, the collision among these
UEs will result in a high block rate and more power consumption
at the mobile devices due to increased congestion delays and
dropped call probability. Hence the goal of our scheme is to
facilitate the optimization of the overall performance in the macro
and femto-cell integrated access networks. The proposed VHD
scheme needs a central mobility control entity to collect
information of the current link condition, traffic load, network
capacity for the handoff decision. Given a large number of femto-
cells, multiple mobility control entities can be introduced and
distributed in the overall cellular coverage to implement the
VHD. Each mobility control entity provides the VHD function for
a region covering one or multiple femtocells and/or BSs, such as
a hotspot or a company building.
B. Efficient handoff algorithm for inbound mobility
in hierarchical macro/femto cell networks
Jung-Min Moon, Dong-Ho Cho was said the recent
development of hierarchical macro/femto cell networks is a
realistic way of providing better quality of service to indoor
mobile users. In these emerging networks, many low-power
femto base stations (f-BSs) are implemented within the coverage
of macro BSs (m-BSs) that typically uses large transmit power for
covering a wide geographic area. One challenge here is to support
the successful inbound mobility that corresponds to the handoff
from the m-BS to the f-BS. To achieve this purpose, we are
interested in designing an efficient handoff algorithm to be used
in the hierarchial macro/femto cell networks. A variety of handoff
algorithms based on received signal strength (RSS) with
hysteresis and threshold have been studied. The threshold sets a
minimum RSS from a dealing BS and the hysteresis adds a
margin to the RSS from the dealing BS over that from a target BS
Although their efficiency has been verified in many previous
works, the performance in the hierarchical macro/femto cell
networks was not evaluated. Therefore, we propose a new RSS-
based handoff algorithm that is suitable for the hierarchical
macro/femto cell networks. The handoff scenario considered in
this paper is the inbound mobility from the m-BS to the f-BS.
C. Handover Study Concerning Mobility in the Two Hierarchy
Network
Wu Shaohong, Zhang Xin, Zheng Ruiming, Yin Zhiwei,
Fang Yinglong,Yang Dacheng was said an auto-configuration
method due to signal control is proposed to reduce the
macrocell users’ call dropping by auto-conditioning
femtocell’s transmission and pilot power levels. Simulation
results show the handover probability as a function of the
distance from a femtocell for the users traveling from the
macrocell to the femtocell. To evaluate different handover
algorithms including soft handover in a picocell, while
provides a analytical model for handover schemes. In a
moving mobile is taken into account and a handover algorithm
based on the estimates of location and velocity of the mobile is
brought forward. A real GSM system is used to evaluate the
algorithm and the results show an obvious cut of unnecessary
handovers with the proposed scheme. The studies the optimal
moving speed access and suggests a mobility management
scheme for multi-hop networking in hybrid networks. To
describes a coverage adaptation method by maintaining a
certain number of mobility events around the femtocell so as
to optimize the femtocell coverage. However, these papers
haven’t combined signal and mobility together when
discussing handover issues in the two-hierarchy network.
D. Combined SINR Based Vertical Handoff Algorithm for
Next Generation Heterogeneous Wireless Networks
Kemeng Yang, Gondal, I., Bin Qiu, Dooley, L.S was said To
provide seamless handover between WLAN and WCDMA, a
SINR based vertical handoff that can support multimedia QoS
with adaptive data rate is desirable. The new vertical handoff
Electronic copy available at: https://ssrn.com/abstract=3123532
Proceedings of the International Conference on Intelligent Computing Systems (ICICS 2017 – Dec 15th -16th 2017)
organized by Sona College of Technology, Salem, Tamilnadu, India
Elsevier’s SSRN eLibrary – Journal of Information Systems & eBusiness Network -ISSN: 1556-5068
82
algorithm not only can support the user with multimedia QoS and
allow them achieving the maximum throughputs during vertical
handoff, but also makes the load balancing between WLAN and
WCDMA systems practical. . In order to have a consolidated
radio resource management strategy for the all kind of wireless
network, it is essential to also have a SINR based vertical
handoff. The proposed novel vertical handoff algorithm using
SINR instead of RSS as the handoff criterion. Here combined
effects of both the SINR from WLAN and WCDMA are being
considered to decide on the handoff. Analysis results show that
SINR based vertical handoff gives us a higher average throughput
for end users comparing with the RSS based vertical handoff with
various thresholds settings, and also it can adapt to different
network conditions, such as different noise level and load factor.
Simulation results further confirm that the SINR based vertical
handoff improves the overall system throughputs. The paper is
organized as follows.
III. PROPOSED SYSTEM
The proposed goal is to improve the efficiency of
decision-making. There is a need to select an optimal destination
network using an efficient decision algorithm and perform the
necessary handovers in vertical communication. In terms of
performance, it is necessary to minimize power consumption by
keeping the vertical handover decision simple and the bandwidth
overhead must minimize the amount of additional network
traffic used to implement handovers by achieving an optimal
network. Each application requires different QoS, so network
selection may vary accordingly. The proposed GA-based
framework addresses multiple wireless networks including
WiMAX, WLAN, 3G UMTS, and 4G LTE. The main focus is
on optimizing soft handoff so that the MT has the ability to scan
and find the best available access networks in a timely manner.
To achieve this goal and to select the best network for a mobile
terminal when switching from one network to another requires a
good decision algorithm that determines the best network for a
specific application that the user is based on the QoS parameter
needed.
A. System Design
It is expected that the optimized handover algorithm will
make a faster handover decision to reduce latency and service
degradation. The framework for optimizing the soft handoff
decision is based on the use of the GA technique, as illustrated
in FIG.1, and shows the flowchart showing required GA
operations and termination criteria for the framework. In
addition, a suitability function is developed to ensure a high
handoff probability of success, which can lead to timely
network discovery and handover decisions.
Figure 1: System Architecture
IV. METHODOLOGY
A. Supervised GA-based Handoff Optimization
To optimize the artificial neural network by the genetic
algorithm, each interconnected weight between input and hidden
or between hidden and output layer nodes is binary coded as a 16-
bit chain in a chromosome, and the chromosome is developed
through crossover and mutation operations , Each input layer
node and its associated weights of trained monitored genetic
elements are systematically omitted once (the self-depleted
weights), and the corresponding weight adjustments due to the
omission are calculated to keep the overall network behavior
unchanged. The primary feature rank index, defined as the sum of
the self-deprived weights and the corresponding calculated
weighting corrections, is found to be able to separate good from
bad features for some artificial data sets from tested known
feature scores. The final feature indexes used to rank the records
are calculated as a sum of the weighted frequencies of each
feature that is classified in a particular rank for each record that is
divided into numerous clusters.
Pattern recognition genetic algorithm (this type of algorithm
depends on the type of label output, whether the learning is
supervised or unattended, and whether the algorithm is statistical
or non-statistical), position-aware algorithm, received signal
strength (RSS) based algorithm (In this type of algorithm, RSS is
the primary handoff decision criterion.) These algorithms use the
RSS of the current network and the Candidate Network), artificial
intelligence based algorithm.
B. Handoff Decision in Wireless Networks
The handover may be initiated for reasons related to the
quality of service of the MT, such as: For example, channel
interference, user behavior changes, capacity shortage in the
current cell, and call termination prevention. During handover,
packet loss is likely to occur and one of the goals of mobility
management is to minimize packet loss. The various techniques
Proceedings of the International Conference on Intelligent Computing Systems (ICICS 2017 – Dec 15th -16th 2017)
organized by Sona College of Technology, Salem, Tamilnadu, India
Elsevier’s SSRN eLibrary – Journal of Information Systems & eBusiness Network -ISSN: 1556-5068
83
for controlling the handover decision may be divided into
mobile-initiated, network-initiated, mobile-controlled,
network-controlled, mobile-assisted and network-assisted
handover. In a hand initiated handover, the mobile node makes
the initial handover decision as the network decides to hand
over in the network initiated handover. In mobile handover,
the mobile node has primary control over the handover
process, while the network has primary control over the
handover process during network handover. Finally, with
mobile assisted handover, information and measurements from
the mobile node are used by the core network to decide on the
execution of handoff, while network assisted handover means
that the core network provides information that can be used by
the mobile node a handover decision.
During the handover detection phase, the framework activates
the wireless interfaces on the MT within an interval and looks
for available networks that overlap the current network. The
activation uses an adaptive detection scheme adopted by the
MT to ensure that a potential wireless network is ready for
handover, based on the list stored in the database, which
contains Rmax, Rmin and the location of the base station.
The interval for activation is evaluated using equation as:
------(1)
where, Tinterval is the interface activating interval, Tmax is the
upper bound of Tinterval, Tmin is the lower bound of Tinterval, Rmax
is the radius of the ideal coverage, Rmin is the radius of the
minimum coverage, Rcurrent is the distance between base station
and MT.
C. The Proposed Vertical Handsoff Decision algorithm
The proposed vertical handover decision algorithm consists of
three-level data acquisition, data processing and handover
decision. The first step is to collect decision parameters that
include terminal parameters, network parameters, user
parameters, and service parameters. This algorithm creates QoS
criteria from the network-related, terminal-related, user-related
and service-related attributes and determines attribute weight
values using the analytical hierarchy process and extracts the
main component using a principal component analysis. The
optimization target network selection is as follows.
------ (2)
The algorithm can guarantee the accuracy of the vertical
handoff. The available vertical multi-handover methods,
which only consider the network-related parameters, cannot
meet the user requirements because user satisfaction is most
important to the network QoS. If more attributes are selected
in the vertical multi-attribute handling decision algorithm, the
selection process will be more complex and the computational
cost will be higher. Here the less independent synthetic
attributes are used to replace the dependent handoff attributes
with principal component analysis.
V. EXPERIMENTAL RESULTS
The proposed protocol and its two associated
protocols were simulated in Network Simulator2, an event-
driven and modular simulation framework. At the beginning
of the simulation model, the input parameters are required on
which the decision is calculated, such. The network condition
and they also identify that the new network may handle the
handover or not.
Table 1: Simulation Parameters
In Figure 2, increasing the number of channels for
communication in the network allows more handover requests
to be accepted as the handover probability of success
increases. Thus, the algorithm helps the system maintain a
90% constant handoff probability of success if the availability
of the channels increases from 5 and higher. Where a plot of P
(SH) versus traffic intensity is presented, the likelihood of
success of handover requirements decreases with increasing
traffic intensity. The GA has still optimized the soft handoff
decision by ensuring a 75% probability of success, even at
high traffic intensity, ensuring drastic bridging latency
reduction. A diagram with the best individual values and the
parameters with GA for optimization.
Figure 2. Parameters vs. Number of Variables for Best Individual
Proceedings of the International Conference on Intelligent Computing Systems (ICICS 2017 – Dec 15th
-16th 2017) organized by Sona College of Technology, Salem, Tamilnadu, India
Elsevier’s SSRN eLibrary – Journal of Information Systems & eBusiness Network -ISSN: 1556-5068
84
As expected, the current best person should have
the ability to maintain some connection or traffic intensity
and have a reasonable number of channels. In addition,
the optimal system should have an almost non-existent
lowered call rate and be able to allow long call durations.
It is expected that the selected base station will have the
closest value to the global optima of the handoff
probability model.
VI. CONCLUSION
The optimized handoffs are extremely important
in heterogeneous networks because of the cellular
architecture used to maximize spectrum utilization. The
proposed GA-based framework addresses multiple
wireless networks including WiMAX, WLAN, 3G
UMTS, and 4G LTE. The main focus is on optimizing
soft handoff so that the MT has the ability to scan and
find the best available access networks in a timely
manner. To achieve this goal and to select the best
network for a mobile terminal when switching from one
network to another requires a good decision algorithm
that determines the best network for a specific application
that the user is based on the QoS parameter needed. This
process shows that with high traffic intensity on at least
two channels, the system could achieve a success rate of
about 85% and 93% respectively, indicating that a
telecommunications system can be planned to achieve the
desired maximum handover success rate. In summary, the
monitored GA scheme is simple, efficient, generic and
suitable for a vertical handover decision and can be
adapted to all other wireless network types upon receipt
of the required simulation parameters.
REFERENCES
[1] Shuhui Liu, Yongyu Chang , Guangde Wang, Dacheng Yang,
“Vertical
Handoff Scheme Concering Mobility in the Two-hierarchy
Network, ” IEEE GLOBECOM 2011,pp.237-241,2011.
[2] Jung-Min Moon, Dong-Ho Cho, “Efficient handoff algorithm for
inbound mobility in hierarchical macro/femto cell networks,”
IEEE
Communications Letters, vol.13, no.10, pp. 755-757, 2009.
[3] Wu Shaohong, Zhang Xin, Zheng Ruiming, Yin Zhiwei, Fang
Yinglong,Yang Dacheng, “Handover Study Concerning Mobility
in the TwoHierarchy Network”, IEEE 69th VTC Spring 2009,
pp.1-5, 2009.
[4] Kemeng Yang, Gondal, I., Bin Qiu, Dooley, L.S.,” Combined
SINR
BasedVerticalHandoffAlgorithmforNextGeneration
HeterogeneousWireless Networks “, IEEE GLOBECOM 2007, pp.
4483-4487, 2007.
[5] N. Nasser, A. Hasswa, and H. Hassanein, “Handoffs in fourth
generation heterogeneous networks,” IEEE Commun. Mag., vol.
44, no. 10, pp. 96–
103, Oct. 2006.
[6] Fang Zhu, McNair, J., “Optimizations for vertical handoff decision
algorithms”, IEEE WCNC 2004, vol.2, pp. 867-872, 2004.
[7] R. Fletcher and S. Leyffer, “Numerical experience with lower
bounds for MIQP branch-and-bound,” SIAM J. Optim., vol. 8, no.
2, pp. 604–616, 1998.
[8] S. Moghaddam, V. Tabataba, and A. Falahati, “New handoff
initiation algorithm (optimum combination of hysteresis and
threshold based methods),” in Proc. IEEE Vehic. Technol. Conf.,
Sept. 2000, pp. 1567– 1574.
[9] Lee, Kotikalapudi Sriram,Kyungsoo Kim, Yoon Hyuk Kim,
and
NadaGolmie,“Vertical handoff decision algorithm for providing
optimized performance in heterogeneous wireless networks” IEEE
Transactions On VehicularTechnology, vol. 58, no. 2, Feb2009.
[10] X. Yan, Y. A. S¸ekercio ù glu, and S. Narayanan, “A survey of vertical
handover decision algorithms in fourth generation heterogeneous
wireless networks,” Computer Networks, 54(11):1848-1863, August
2010
[11] X. Yan, N. Mani, Y.A. Sekercioglu, “A travelling distance
prediction based method to minimize unnecessary handovers from
cellular networks to WLANs,” IEEE Communications Letters,
Vol. 12, No. 1, pp. 14–16, 2008.
[12] A. H. Zahran, B. Liang, A. Saleh, “Signal threshold adaptation for
vertical handoff in heterogeneous wireless networks,” Springer
Mobile Networks and Applications, Vol. 11, no. 4, pp. 625-640,
August 2006.
[13] M. Miyim ,Mahamod Ismail, Rosdiadee Nordin "Vertical
Handover
Solutions Over LTE-Advanced Wireless Networks: An Overview"
springer Wireless Personal Communications V 77,( 4) pp 3051-
3079,2014.
[14] Vahid Sattari Naeini " Performance analysis of WiMAXbased
wireless mesh networks using an M/D/1 queuing model" Int. J.
Wireless and Mobile Computing,Vol. 7, No.1, 35-47, 2014.
[15] Vahid Sattari Naeini " Performance analysis of WiMAXbased
wireless mesh networks using an M/D/1 queuing model" Int. J.
Wireless and Mobile Computing,Vol. 7, No.1, 35-47, 2014.
[16] D. Ma and M. Ma, “A QoS-based vertical handoff scheme for
interworking of WLAN and WiMAX,” in Proceedings of the
IEEEGlobal Telecommunications Conference (GLOBECOM’09),
pp. 1–6, Honolulu,Hawaii, USA, December 2009.
[17] Jiang, D., Xu, Z., & Lv, Z. (2015). A multicast delivery approach
with minimum energy consumption for wireless multi-hop
networks. Telecommunication Systems, 62(4), 771–782.
[18] Bushra Naeem,” Seamless Vertical Handover in Wi-Fi and
WiMAX Networks using RSS and Motion Detection: An
Investigation” The
Pacific Journal of Science and Technology Volume 12. Number 1.
May 2011 (spring).
[19] P. Kavitha and M. Santhalakshmi, “Enhanced QOS Factors for
Load Balamced Handoff WiMax Net”, International Journal of
Computer Application, Vol. 43, No. 24, April, 2012.
[20] Ji, Xiaolong, Jing Zhang, and Sulei Zhu. "A Novel Vertical
Handoff Algorithm for UMTS and WiMAX Heterogeneous
Overlay Networks." Information Science and Control Engineering
(ICISCE), 2015 2nd International Conference on. IEEE, 2015
