Handover Challenges in Wireless Communications and

Their Solutions
Sudesh Pahal*, Priyanka Nandal
*Department of ECE, Maharaja Surajmal Institute of Technology, Delhi, [email protected]
Department of CSE, Maharaja Surajmal Institute of Technology, Delhi, [email protected]

ABSTRACT
The significant expansion of mobile communication over the last few years allows the users to
communicate ubiquitously without individual geographical coverage constraints on wireless
networks. Mobility is the major motivation for the extensive development and deployment of
mobile wireless networks. In order to provide global connectivity to mobile users with reliable
communication channels and to meet the requirement of seamless handover for real-time and
multimedia applications, handover management deserves careful attention. This paper presents a
comprehensive review of handover challenges in wireless networks and their solutions. This paper
offers a detailed study of the handover process, potential open issues, and future directions.

Keywords: Cross layering, Handover, Mobility Management, Mobility Protocol, Signal

Strength

1 BACKGROUND AND INTRODUCTION

To fulfill the needs of users, the increasing evolution of access technologies necessitates
interoperability and enhanced techniques for mobility management. Based on coverage
area and bandwidth, most of the wireless access technologies can be divided into three
categories: Wireless Local Area Networks (WLANs), Wireless Metropolitan Area
Networks (WMANs) and Wireless Wide Area Networks (WWANs) as depicted in Figure
1. The technologies with smaller coverage, high power consumption and high bandwidth
are covered under WLAN. The coverage area of ten meters to few hundred meters can be
served by these networks. WLAN standard (Deng and Yen, 2005) mainly comprises of the
IEEE 802.11 series and ETSI’s (European Telecommunications Standards Institute)
HiperLAN. WMANs provide wireless connectivity to users between various locations within a
metropolitan area. It enables users to roam within a range from hundreds of meters to a few
kilometers. Broadband Wireless Access (BWA) networks are also becoming popular as they offer
high speed Internet access to users without incurring the extra cost of leasing lines and laying fiber
or copper cable. The standards being researched and developed to support WMANs (Chang,2005)
are High Performance Radio Metropolitan Area Network (HiperMAN), High Performance Radio
Access (HiperACCESS) and IEEE 802.16 series. HiperACCESS, an interoperable standard,
provides broadband access to a lesser coverage area and backhaul for mobile systems (e.g.
W−CDMA, CDMA2000, GSM and GPRS) with a data rate of approximately 100 Mbps.

Electronic copy available at: https://ssrn.com/abstract=4749805

HiperMAN is designed for providing a broadband wireless solution for MANs. The air interface
allows Point−to−Multi Point (PMP) configurations as well as flexible mesh network.
HiperACCESS operates at
high frequency bands (40.5 – 43.5 GHz) while HiperMAN is a fixed broadband wireless access
system that operates at radio frequencies ranging from 2 to 11 GHz.

CDMA
IEEE
IEEE
802.16
802.11
X
WAN
PAN LAN MAN
IEEE ETSI’s ETSI’s
802.15
HiperLAN HiperMAN
X
3G UMTS

GSM

Figure 1 Hierarchy of wireless networks

WiMAX (Wireless Interoperability for Microwave Access), an IEEE 802.16 standard, is a BWA
technology that offers scalable options with high bandwidth and cheap cost as an alternative to
wired access links. The IEEE 802.16e standard, often referred as mobile WiMAX, supports
mobility and has been shown to be effective whenever mobility services are expected for users
travelling at fast speeds. To encourage compatibility across different vendors, the IEEE 802.16
standard (IEEE Std 802.16, 2006) includes PHY and MAC layers. WiMAX uses two frequency
bands: unlicensed (2 to 11 GHz) and licenced (2 to 11 GHz) (10 to 66 GHz).
WWAN has a coverage area from a few kilometers to tens of kilometers. Cellular networks
including Global System for Mobile Communication (GSM), Code Division Multiple Access
(CDMA), General Packet Radio Services (GPRS), High Speed Downlink/Uplink Packet Access
(HSDPA/HSUPA), UMTS (Universal Mobile Telecommunication System), and LTE (Long Term
Evolution) are examples of these access networks (Berezdivin et. al., 2002). These wireless
networks were created to meet certain service needs including coverage area, data rate, and delay.
The wireless networks, comprised of different radio access technologies including WLAN, GSM,
CDMA, 3G UMTS and WiMAX, are known as heterogeneous wireless networks while
homogeneous wireless networks comprise of the same kind of access networks. Users will be able

Electronic copy available at: https://ssrn.com/abstract=4749805

to select the most suitable wireless network interface for their requirements in the next generation
of networks.
In upcoming wireless networks, it is vital to keep user’s overall satisfaction at a very high level by
providing them high QoS at affordable cost. The contribution of this paper are as follows:
• Section 1 presents background and introduction to handover in wireless networks.
• In section 2, handover management, an important aspect in the context of providing QoS,
is discussed in detail including its classification.
• In section 3, different criteria used for taking decision regarding handover are discussed.
The rational for designing intelligent and dynamic handover schemes is explained.
• Afterwards in section 4, the handover challenges and their possible solutions are
elaborated.
• The incorporation of cross layer concepts is emphasized to provide ubiquitous services to
mobile users in section 5 followed by open issues in section 6. Then the paper is concluded,
followed by future research directions.

2 THE CONCEPT OF HANDOVER

Mobility is the term related to movement of users from one place to another, which leads to
variation in Point of Attachment (PoA) to the access network. Location management and handover
management are two aspects of mobility management (Chi et.al., 2007). The former one monitors
Mobile Node (MN) for successful exchange of information and determines its current PoA while
the handover management keeps undisrupted connections for moving mobile nodes. Handover can
either be forced or unforced. The former is caused by changes in conditions of network such as
signal strength subsequent from the movement of user. The latter refers to a user's modification of
the access network to increase performance.

2.1 CLASSIFICATION OF HANDOVERS

The handovers can be classified into various categories (Naseer et.al., 2006) as shown in Figure 2.
Based upon the kind of networks, the handover can be classified into two categories: Horizontal
Handover (HHO) or Intra−domain handover and Vertical Handover (VHO) or Inter−domain
handover [Figure 3]. Another basis of classification is the number of active connections, which
defines two types of handovers: Hard handover and Soft handover. The handover can be initiated
and controlled by either MN or Base Station (BS). So further classification, based upon handover
decision, describes following types of handovers: network initiated, mobile initiated, network
controlled, mobile controlled, mobile assisted and network assisted handover.

Electronic copy available at: https://ssrn.com/abstract=4749805

 Handover Classification

Type of Number of
networks active Handover decision
involved connections

Horizontal vertical Hard Soft Network Mobile Network Mobile

handover handover handover handover assisted assisted controlled controlled

Figure 2 Classification of handovers

Figure 3 Horizontal and Vertical handovers

Horizontal handover or Intra−domain handover/Vertical handover or Inter−domain

handover: Horizontal handover takes place when the MN moves between BSs supporting same
network technology. In other words, horizontal handover is implemented between two neighboring
cells of homogeneous networks, e.g. handover between two WLAN cells. On the other hand,
vertical handover or Inter−domain handover is accomplished between BSs belonging to diverse
wireless access technologies. The handover between an Access Point (AP) of WLAN and BS of
WiMAX or 3G UMTS network, shown in Figure 1.3, is an example of vertical handover. Vertical
handover is more complicated than horizontal handover because Various characteristics of the
related networks differ, including operating frequency, bandwidth, and modulation schemes.

Electronic copy available at: https://ssrn.com/abstract=4749805

Hard and soft handover: Hard handover is based on the BreakBeforeMake (BBM) principle,
which states that an MN can only be connected with one BS at a time. To put it another way, the
MN can only be linked to the new link once the prior link has been broken. In the event of a hard
handover, the service is temporarily disrupted. In the case of a soft or make-before-break handover,
the MN can communicate with numerous access networks at the same time. Because the new
connection is formed before the previous one is released, services are not disrupted during a gentle
handover.

Mobile initiated/Network initiated handover: If the MN is taking the decision to initiate

handover, it is termed as mobile initiated handover and if the handover is initiated by the network
then it is said to be a network initiated handover. In case of weakening of the signal strength
received from the Previous BS (PBS), the handover is originated by mobile terminal. On the other
hand; a BS can also induce a handover to poise the traffic distribution to prevail crowding on a
specific router.

Mobile controlled/Network controlled handover: The mobile has prime control over the
handover process in case of mobile-controlled algorithms while the network is responsible for the
same in case of network-controlled handover.

Mobile assisted/ Network assisted handover: In case of mobile assisted handover, the core
network executes the handover using information from the MN while a network assisted handover
is performed by the mobile with the assistance of the network.

3 CRITERIA FOR HANDOVER

The networks have distinct features in terms of bandwidth, modulation and coding schemes,
spectral efficiency and data rate. These characteristics or parameters have been considered as
basis/criteria for handover decision. Some of these important parameters are listed below.

1. Received Signal Strength (RSS): The signal strength received at the MN from Current
BS (CBS) and Target BS give a measure of link quality. Most of the existing handover
algorithms consider RSS as the main criteria for taking handover decisions (Gudmunson,
1991; Vijayan and Holtzman, 1991; Zonoozi et. al., 1997). The RSS decreases as the
distance between BS and MN increases due to the fact that path losses introduced in the
signal are directly proportional to the distance.
2. Quality of Service (QoS): The QoS professed by the user can be measured in terms of
jitter, Bit Error Rate (BER), throughput, and Packet Loss Rate (PLR). The requirements of
QoS vary with type of application services such as voice, data and streaming, etc. The
packet loss should be minimized in order to guarantee successful delivery of packets. To
maintain QoS, this attribute is to be kept into consideration while performing handover.
Data rate is a feature desired mainly in multimedia applications like video streaming. The
end-to-end delay is defined as the time taken for packets to reach from source to destination
address. The variation in delay due to multipath is termed as jitter, which is not tolerable
in some delay sensitive applications including conversational and interactive services.

Electronic copy available at: https://ssrn.com/abstract=4749805

 3. Available bandwidth: The available bandwidth, a measure of available network
resources, is also an important measure to be well-thought-out in handover decision. The
requirement for bandwidth varies according to the demands of users and type of service.
4. Energy efficiency: The energy efficiency can also be considered as a handover criteria
while deciding for a handover. The simultaneous running of various applications demands
more energy efficient networks to be available for use.
5. Monetary cost: The cost of different networks varies according to the different charging
policies. While making handover decision, the cost of the network can also account for a
network selection.
6. Security: Security level provided by access network plays the most significant role while
selecting network for some specific applications. Security refers to the integrity or
confidentiality of the transmitted data which is desired at the highest priority in the military
like applications.
7. Number of active users: The number of users, accessing a particular network
simultaneously, is the major criteria in case of network-initiated handover. The system
controller tries to avoid network congestion by balancing the load on co−existing networks.

4 HANDOVER CHALLENGES AND OPPORTUNITIES

To achieve internetworking among different networks, various issues need to be resolved. In this
section, a brief discussion on handover challenges and opportunities is given.

1. Handover rate: The handover rate is referred as the total number of handovers an MN
experiences while drifting from one cell to another. There should only be one handover,
which should take place at the cell's edge. However, due to RSS oscillations, the MN may
experience many handovers.
2. The probability of handover failure: The handover failure is the situation when signal
strength received from the serving network, reduces below a threshold level but the
handover procedure is not completed. Thus, the probability of handover failure is stated as
the probability of occurrence of handover at a point where sufficient signal strength is not
received by MN from CBS and it has not been connected to the target network. Due to late
initiation of handover, there may be a high probability of handover failure.
3. Resource utilization: If the handover is initiated too early, the resources of serving
network will not be utilized at its best level and the possibility of occurrence of this
situation is known as the probability of false handover initiation. When the cell size gets
smaller, the problem of erroneous handover initiation gets worse. Wireless system cell
sizes are shrinking, allowing capacity and data rate to improve. As a result, in advanced
wireless systems, it's critical to choose the right trigger time to limit the chance of a false
handover.
4. Probability of unnecessary handovers: The ambiguity in handover trigger time may lead
to enhanced handover rate, which will lead to wastage of network resources and enhance
the signaling load on the system unnecessarily. It is mainly caused when handover
requirement is not estimated properly.
5. Ping−pong effect: Ping-pong effect is caused by frequent handovers between access
networks due to fluctuations observed in RSS. This frequent changing of access
technologies leads to unstable connection because MN is oscillating between two

Electronic copy available at: https://ssrn.com/abstract=4749805

 networks. Ping-pong effect results in an increase in handover rate, therefore, need to be
mitigated by using efficient handover algorithms.

5 THE CONCEPT OF CROSS LAYER DESIGN IN HANDOVER

PROCESS
Next-generation wireless systems must provide high-speed data, real-time and non-real-time
multimedia capabilities to mobile customers at all times. A layered approach is used in the majority
of current work in the field of wireless network protocol analysis and design. A TCP/IP protocol
stack typically has five layers; Application layer, Transport layer, Network layer, Data link and
Physical layer. Each layer of mobile protocol stack contributes to various functions related to
mobility. In layered architecture, each layer has been assigned its individual function which is to
be performed independently. It allows for layer transparency and versatility. However, the layered
approach does not perform well in advanced wireless systems as Recent wireless networks employ
an all-IP architecture. As a result, when building handover algorithms, it is critical to include
factors from higher layers. In addition, the performance of one layer influences the performance
of others and ultimately the overall system performance of wireless networks. For instance, poor
link quality measurements at physical layer can lead to packet losses and finally deteriorates the
overall network performance (liu, et. al., 2009As a result, it is necessary to include a mechanism
for communicating important information between layers that can be used to improve system
performance. Cross Layer method is the concept of sharing information between layers (Chen et.
al.,2011). The majority of cross-layer topologies are classified as either direct or explicit cross-
layer communications or indirect or implicit cross-layer communications (Wang, 2003). The first
category, direct communications, should be used when only a single cross layer optimization is
expected. The combination of the cross layer signaling methods has been optimized using CLASS
(Cross−Layer Signaling Shortcuts) based upon multilayer mobility management architecture.
CLASS aids as a general interface to empower the direct interactions among non−neighboring
layers using light weighted messages. Indirect communications, on the other hand, are
implemented with a shared cross-layer entity, where handover decisions can be made based on
data acquired from multiple layers and coordinated by a well-designed cross layer interface
manager (Mingxin, 2007). Some popular cross layer signaling methods are ICMP (Internet Control
Message Protocol) messages, packet headers, local profiles and network service method [18].
According to the packet header method, the IPv6 packet headers are extended to store cross layer
information and to process them layer by layer. In the second method, the information across
different layers is propagated through ICMP messages. IP packets encapsulate these messages,
which must flow via the network layer. In the local profile method, separate profiles are designed
to store information abstracted from each layer. In network service method, an entity referred to
as Wireless Channel Information (WCI) server is responsible for gathering and managing
information from lower layers. Concerned applications access to the WCI server to fetch required
information. Due to several beneficial features of cross layer technique, it can be used to meet the
challenges that mobility poses. The utilization of cross layer information in handover decision
making process may allow mobile terminals to roam smoothly across wireless access networks
while maintaining access to the services they have subscribed for on their home networks. The
Media Independent Handover (MIH) defined by IEEE 802.21 standard (IEEE 802.21 MIHS),
enables interaction among different layers by introducing a technology independent function,
MIHF. By sharing event information among entities from various technologies, the MIH facilitates

Electronic copy available at: https://ssrn.com/abstract=4749805

faster handover, assuming MN's ability to support multiple interfaces (Taniuchi and Corporation,
2009; Kim et. al., 2009). The three functional components of MIHF to be used cooperatively to
assist with seamless handover are MIH Event Service (MIES), the MIH Command Service (MICS)
and MIH Information Service (MIIS). MIES is in capable of sensing and reporting events via local
and remote interfaces. This type of service is to provide information like link weakening and link
lost from lower layers to upper layers. MICS, on the other hand, defines instructions for higher
levels to govern lower layers when it comes to handovers. MICS is responsible for the
configuration of network devices as well as the scanning of potential networks. In contrast to
events, commands follow a top-down approach and allow higher layers to manage lower layer
behaviour relating to handover, connectivity, and mobility. MIIS provides a framework for
retrieving information about the serving and surrounding networks' characteristics and services to
aid in the handover decision. Static link layer parameters, such as information about channel
quality or the AP's Medium Access Control (MAC) address, can be accessed via a query/response
procedure from both the bottom and higher levels.
The link triggers like Link Up (LU), Link Down (LD) and Link Going Down (LGD) triggers are
defined by MIH, which play a significant role in taking a handover decision (Mhatre and
Papagiannaki , 2006; Akyildiz and Wang, 2002; Baroudi and Al-naseer, 2011) . LU and LD
triggers are related to the events where the link to the access network has been established and
broken respectively. The LGD trigger indicates that a broken link is impending. The trigger can
be fired based upon the conditions associated with single or multiple criteria.

6 OPEN ISSUES AND POSSIBLE SOLUTIONS

In the past few years, momentous research has been carried out to perform handover seamlessly
during movement of MN across different cells. To achieve a fast and smooth handover, the
conventional layered architecture seems to be inefficient as the functions performed at various
layers are interrelated. Some useful information among different layers is required to be exchanged
to expedite the handover process. The capabilities of MIH give a drive to exchange required
information such as parameters measured, the link triggers, etc. from different layers to take
decision about the handover. Also, predefined thresholds of parameters such as RSS, BER and
distance do not perform well in dynamic environments. Dynamic environments correspond to
changes in numerous parameters over time like condition of wireless channel, user preferences,
the MN speed and time taken for execution of a handover etc. It demands for handover mechanisms
which are adaptive to dynamic environments. The inappropriate handover trigger generation can
lead to too early or too late initiation of handover. If the trigger is generated too early, then there
will be wastage of network resources while too late handover trigger may result into disruption in
services. Hundreds of millions of multimedia applications have been released as a result of rapid
advancements in wireless technologies. As a result, a significant shortage of wireless
communication resources has emerged. Effective mechanisms to generate handover triggers need
to be incorporated for best utilization of network resources. Afterwards, the selection of target
network (to which handover is to be performed) is also a crucial task. To choose a network with
the best features, the handover decision is to be taken based on multiple criteria like RSS, QoS,
data rate, security, cost, power consumption. The priorities to these criteria may vary with time,
user’s choice, type of applications and situational context. To maximize end user's satisfaction,
different weights to handover parameters should be assigned based on their preferences, context
and applications.

Electronic copy available at: https://ssrn.com/abstract=4749805

In most of the existing multi−criteria handover algorithms, fixed weights to handover design
parameters are given irrespective of the changes in environments which could degrade system
performance. Moreover, the handover algorithms, which give crisp value to weights, ignore the
fuzziness and vagueness associated with characteristics of networks. Therefore, keeping
uncertainty of network information in mind, fuzzy logic can be incorporated into the weight
assignment technique.
The procedure for handover execution differs by type of the serving and target networks. For
horizontal handovers, handover is performed at link layer while in case of vertical handovers;
Layer 3 (L3) handover is invoked after Layer 2 (L2) handover. The vertical handover incurs more
delay comparatively as reconfiguration of IP addresses is postulated. Many mobility management
protocols (Manner, 2002) such as MIPv6 (Mobile Internet Protocol version 6), Session IP (SIP),
Hierarchical MIPv6 (HMIP), and Fast handover for Mobile IPv6 (FMIPv6) have been proposed
to enhance the performance of handover execution phase. Due to high delay associated with the
execution of vertical handover, service disruption time is high. To address these issues, cross layer
mechanism should be incorporated in handover process. The overlapping of functions of L2 and
L3 can reduce handover delay to the remarkable extent.

7 CONCLUSION AND FUTURE RESEARCH DIRECTIONS

It may be concluded from this research that cross layer concept plays a significant role to deal with
handover related problems and for performance enhancement of three stages of handover:
handover initiation, handover decision and handover execution stage. In addition, efficient
handover initiation algorithms can be designed by optimizing handover parameters including
threshold signal strength and transmitter power. These algorithms may result into reduced
handover rate and probability of call interruption as compared to RSS based algorithms. An
appropriate trigger generation for initiating the handover procedure would decrease the probability
of unnecessary handover, the probability of handover failure and false handover initiation. Correct
estimation of the handover requirement supports minimum wastage of network resources. Thus
the parameters for determination of the handover requirement need to be investigated correctly.
The selection of target network selection should be dynamic in nature which would be capable of
selecting the most suitable target network, ac/cording the user requirements at any moment which
in turn may improve QoS as perceived by the user. In addition, handover latency and signaling
load may be reduced by executing certain steps of the handover procedure of layer 2 and layer 3
simultaneously.

In this section, the scope for further improvement, in different aspects of the handover schemes, is
discussed. Some of the major issues that can be of interest for further research related to the
handover challenges discussed in the preceding sections are as follows:

• The interfacing among different layers needs more exploration to make best use of the MIH
functionality for implementing more efficient trigger methods for handover.
• The selection of the target network is one aspect of the handover decision to provide
ubiquitous access to mobile users. To provide continuous service, network selection can
be integrated with effective channel assignment mechanisms. The history of user profile

Electronic copy available at: https://ssrn.com/abstract=4749805

 can be a useful measure of expected service requirements. The past samples about user
information can be utilized to take a handover decision in advance.
• Furthermore, several artificial intelligent techniques such as neural networks, Fuzzy logic
etc. can be utilized to achieve better performance.
• The complexity of the different handover algorithms can be analyzed to determine
feasibility of the algorithm in real time scenario.
• Better visualization may be achieved using network simulators and test−based
measurements utilizing real wireless network conditions.

REFERENCES

1. Adachi, F. (2001). Wireless Past and Future--Evolving Mobile Communications Systems-

-. IEICE transactions on fundamentals of electronics, communications and computer
sciences, 84(1), 55-60.
2. Akyildiz, I. F., & Wang, W. (2002). A dynamic location management scheme for next-
generation multitier PCS systems. Wireless Communications, IEEE Transactions on, 1(1),
178-189.
3. Akyildiz, I. F., Xie, J., & Mohanty, S. (2004). A survey of mobility management in next-
generation all-IP-based wireless systems. Wireless Communications, IEEE, 11(4), 16-28.
4. Baroudi, U., & Al-Nasser, F. (2011, January). Performance evaluation study on a multiple-
parameter handoff algorithm. In Information Networking (ICOIN), 2011 International
Conference on (pp. 273-277). IEEE.
5. Berezdivin, R., Breinig, R., & Topp, R. (2002). Next-generation wireless communications
concepts and technologies. Communications Magazine, IEEE, 40(3), 108-116.
6. Chang, C. K. (2005, June). A mobile-IP based mobility system for wireless metropolitan
area networks. In Parallel Processing, 2005. ICPP 2005 Workshops. International
Conference Workshops on (pp. 429-435). IEEE.
7. Chen, L., Low, S. H., & Doyle, J. C. (2011). Cross-layer design in multihop wireless
networks. Computer Networks, 55(2), 480-496.
8. Chi, C., Cai, X., Hao, R., & Liu, F. (2007, November). Modeling and analysis of handover
algorithms. In Global Telecommunications Conference, 2007. GLOBECOM'07. IEEE (pp.
4473-4477). IEEE.
9. Deng, D. J., & Yen, H. C. (2005). Quality-of-service provisioning system for multimedia
transmission in IEEE 802.11 wireless LANs. Selected Areas in Communications, IEEE
Journal on, 23(6), 1240-1252.
10. Gudmundson, M. (1991, May). Analysis of handover algorithms [microcellular radio].
In Vehicular Technology Conference, 1991. Gateway to the Future Technology in Motion.,
41st IEEE (pp. 537-542). IEEE.
11. Gwo-Hshiung, T. (2010). Multiple attribute decision making: methods and
applications. Multiple Attribute Decision Making: Methods and Applications.
12. IEEE LAN/MAN Standards Committee. (2006). IEEE Standard for local and metropolitan
area networks Part 16: Air interface for fixed and mobile broadband wireless access
systems amendment 2: Physical and medium access control layers for combined fixed and
mobile operation in licensed bands and corrigendum 1. IEEE Std 802.16-2004/Cor 1-2005.
13. Kim, M., Moon, T. W., & Cho, S. J. (2009, February). A study on IEEE 802.21 MIH
frameworks in heterogeneous wireless networks. In Advanced Communication

Electronic copy available at: https://ssrn.com/abstract=4749805

 Technology, 2009. ICACT 2009. 11th International Conference on (Vol. 1, pp. 242-246).
IEEE.
14. Li, M., Chen, S., & Li, Y. (2007, September). A cross-layer based on share database
mobility management architecture in heterogeneous networks. InWireless
Communications, Networking and Mobile Computing, 2007. WiCom 2007. International
Conference on (pp. 1869-1872). IEEE.
15. Liu, C. H., Gkelias, A., Hou, Y., & Leung, K. K. (2009). Cross-layer design for QoS in
wireless mesh networks. Wireless personal communications,51(3), 593-613.
16. Manner, J., Toledo, A. L., Mihailovic, A., Munoz, H. L. V., Hepworth, E., & Khouaja, Y.
(2002). Evaluation of mobility and quality of service interaction.Computer
Networks, 38(2), 137-163.
17. Mhatre, V., & Papagiannaki, K. (2006, June). Using smart triggers for improved user
performance in 802.11 wireless networks. In Proceedings of the 4th international
conference on Mobile systems, applications and services (pp. 246-259). ACM.
18. Mohamed, L., Leghris, C., & Abdellah, A. (2012, January). A survey and comparison study
on weighting algorithms for access network selection. InWireless On-demand Network
Systems and Services (WONS), 2012 9th Annual Conference on (pp. 35-38). IEEE.
19. Nasser, N., Hasswa, A., & Hassanein, H. (2006). Handoffs in fourth generation
heterogeneous networks. Communications Magazine, IEEE,44(10), 96-103.
20. Piri, E., & Pentikousis, K. (2009). IEEE 802.21: media independent handover services. The
Internet Protocol Journal, 12(2), 7-27.
21. Saaty, T. L. (1999). Decision making for leaders: the analytic hierarchy process for
decisions in a complex world (Vol. 2). RWS publications.
22. Sgora, A., Gizelis, C. A., & Vergados, D. D. (2011). Network selection in a WiMAX–WiFi
environment. Pervasive and Mobile computing, 7(5), 584-594.
23. Taniuchi, K., Ohba, Y., Fajardo, V., Das, S., Tauil, M., Cheng, Y. H., ... & Famolari, D.
(2009). IEEE 802.21: Media independent handover: Features, applicability, and
realization. Communications Magazine, IEEE, 47(1), 112-120.
24. Vijayan, R., & Holtzman, J. M. (1993). A model for analyzing handoff algorithms [cellular
radio]. Vehicular Technology, IEEE Transactions on,42(3), 351-356.
25. Wang, Q., & Abu-Rgheff, M. A. (2003). A multi-layer mobility management architecture
using cross-layer signalling interactions. 237-241.
26. Zonoozi, M., Dassanayake, P., & Faulkner, M. (1997, May). Optimum hysteresis level,
signal averaging time and handover delay. In Vehicular Technology Conference, 1997,
IEEE 47th (Vol. 1, pp. 310-313). IEEE.
27. Yadav, D., & Azad, P. (2019, March). Experimental analysis of power generation for ultra-
low power wireless sensor nodes using various coatings on thermoelectric energy
harvester. In 2019 6th International Conference on Signal Processing and Integrated
Networks (SPIN) (pp. 13-17). IEEE.
28. Pahal, S., Rathee, N., & Singh, B. (2023). A Deep Learning-Based Model for Link Quality
Estimation in Vehicular Networks. IETE Journal of Research, 69(8), 5159-5168.

Electronic copy available at: https://ssrn.com/abstract=4749805