International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
Geetanjali Chellani, Anshuman Kalla
Department of Electronics and Communication Engineering
Jaipur National University, Jaipur, Rajasthan, India
ABSTRACT
The Mobile Internet Protocol (Mobile IP) is an extension to the Internet Protocol proposed by the Internet
Engineering Task Force (IETF) that addresses the mobility issues. In order to support un-interrupted
services and seamless mobility of nodes across the networks (and/or sub-networks) with permanent IP
addresses, handover is performed in mobile IP enabled networks. Handover in mobile IP is source cause of
performance degradation as it results in increased latency and packet loss during handover. Other issues
like scalability issues, ordered packet delivery issues, control plane management issues etc are also
adversely affected by it. The paper provides a constructive survey by classifying, discussing and comparing
different handover techniques that have been proposed so far, for enhancing the performance during
handovers. Finally some general solutions that have been used to solve handover related problems are
briefly discussed.
KEYWORDS
Mobile IP, MIPv4, MIPv6, Hierarchical Mobile IP, Fast Handover.
1. INTRODUCTION
Foundation of today’s Internet architecture, based on TCP/IP, was laid during the days of
telephony when enabling communication between static end users was of prime importance and
mobility of users was least envisioned. But the advent of wireless technology gave rise to the
possibility of mobility and seamless connectivity. Among the several other solutions that have
been proposed so far, Mobile IP is the only widely deployed add-on solution for handling
mobility[1]. In the TCP/IP based Internet architecture a user node is assigned an IP address which
is in-fact a locator of user-node in network. As the node moves from one place to another, it
results in change of network and/or subnet which consequently results in change of IP address.
Since all the connections take IP address as a seed thus change in IP address means that all the
connections must be re-established which inevitably leads to interruption in on-going applications
and services. This issue of varying IP addresses when node is mobile is resolved by Mobile IP.
There are still some issues that need to be reconsidered looking at the enormous growth of mobile
users every-day-and-now. As discussed by J. Chandrasekarn [2] these issues are (i) Handover
Latency, (ii) Triangulation, (iii) Reliability and (iv) Security. In this paper we will discuss all
these issues.
The organization of the paper is as follow. Section II introduces current solutions for supporting
mobility in IPv4 & IPv6 and major differences between them. In section III, mobility
management and related components is presented. The network mobility for mobile networks is
introduced in section IV. In section V different handover techniques that improve handover
DOI : 10.5121/ijcnc.2013.5608
137
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
performance are discussed and a comparative study is done. Section VI introduces some general
techniques that are used to improve handover performance followed by the conclusion in section
VII.
2. MOBILE IP
Mobile IPv4 (MIPv4) is popular mobility internet protocol used in different IPv4 networks and
Mobile IPv6 (MIPv6) has emerged to deal with mobility for advanced version of IP i.e. IPv6.
2.1. MOBILE IPV4
Mobile IPv4 introduced four functional entities: (i) Home Agent (HA), (ii) Foreign Agent (FA),
(iii) Mobile Node (MN), (iv) Correspondent Node (CN). Each MN is resident in its home network
where it receives a permanent Home Address (HoA).When an MN moves out of its home
network and visits a foreign network, it obtains a temporary address which is known as Care-ofAddress (CoA) by the FA in that foreign network. When the MN moves from one foreign
network to another foreign network, it registers its new CoA to the HA that is located in the home
network. The HA keeps track of the HoA and CoA for all MN. A packet from CN destined to MN
is sent to HoA of MN. The HA intercepts all the IP packets destined to the MN and tunnels them
to the CoA of the MN [6].
2.1.1. Basic Mobile IPv4 Protocols Functioning
2.1.1.1. Agent Discovery - In order to discover prevailing agent i.e. home agent or foreign agent,
a mobile node invokes this mechanism. Two different types of messages used are:
2.1.1.1.1. Agent Advertisement - Home/foreign agent advertises its presence periodically by
broadcasting agent advertisement message with-in its network.
2.1.1.1.2. Agent Solicitation - MN can also issue a request message with-in the current network
in order to seek an agent advertisement message.
2.1.1.2. Registration - Mobile node visiting a foreign network informs about its current location
by initiating a registration procedure.
138
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
2.1.1.2.1. Registration Request - Visiting mobile node after fetching temporary CoA from
foreign network needs to convey this CoA to the home agent so it generates and sends a
registration request message.
2.1.1.2.2. Registration Reply - Upon reception of registration request, home agent verifies the
authenticity of mobile node. In case of authentic request, a mapping of CoA is established with
corresponding HoA of mobile node, by adding an entry in routing table. Finally an
acknowledgment is sent to MN in form of registration reply message.
2.1.1.3. Tunneling - Tunneling is used to forward IP datagram from a home address to a care of
address.
2.1.2. Issues in Mobile IPv4
2.1.2.1. Triangular Routing - Mobile IPv4 suffers from a long handover delay due to “triangular
routing”. As shown in figure 1, packets going from MN to CN follow direct route through
internet (i.e.4&5) but packets going from CN to the MN have to travel through HA when the
mobile node is away from home (i.e.1,2&3). This additional routing is called triangular routing.
2.1.2.2. Signalling Overhead - Large signalling overhead is due to large number of registration
updates. Every time a mobile node moves beyond the limit of link layer connectivity, a
registration update is required for the node with its home agent [7].
2.1.3. Solution of Mobile IPv4
Route optimization [8] was proposed to solve triangular routing problem. Messages from the CN
are routed directly to the MN’s CoA without passing through the HoA. The CN maintains a
binding cache that maps the HoA of the mobile node with their CoA. Binding cache needs four
additional messages which are as follows[9]:
2.1.3.1. Binding Request - In order to know the current location of MN, CN sends a binding
request to HA at home network.
2.1.3.2. Binding Update - HA replies to CN with a message that revels the current location of an
MN.
2.1.3.3. Binding Acknowledgement - CN acknowledges HA, the reception of binding update.
2.1.3.4. Binding Warning - This message is used to suggest a MN’s home agent that CN appears
to have either no binding cache entry or an out-of-date binding cache entry for some MN.
2.2. Mobile IPv6 - Mobile IPv6 (MIPv6) is the next generation internet protocol and offers a
number of improvements over MIPv4. MIPv6 supports mobility in both homogeneous (from one
LAN to another LAN) and heterogeneous media (node movement from LAN to 3G network). In
MIPv6, MN should assign three IPv6 addresses (i) Permanent home address, (ii) Current link
local address, (iii) Care-of-Address (CoA), which associated with the mobile node only when
visiting a particular foreign network [10]. MN’s CoA is co-located CoA in MIPv6 which allow
MN to encapsulate and decapsulate packets and connect to HA directly on any foreign link
without notifying FA. The FA function is not there in MIPv6. While the MN moves from one
network (or subnet) to another, CoA is automatically allocated to it in the foreign network due to
the address auto-configuration feature which are (i) Statefull Address Auto-configuration - MN
sends a CoA Request message to the local router and it allocates a new IPv6 address (ii) Stateless
Address Auto-configuration - MN combines IPv6-prefix which it received with its MAC address
139
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
to create new IPv6 address using neighbour discovery. The HA keeps a binding between MN’s
HoA and its CoA. The central data structure collected by each IPv6 node is used as a binding
cache. In MIPv6 route optimization is in-built function so MN periodically sends binding update
messages not only to the HA but also to CN. So, CN adds this binding to the binding cache and
thereafter CN directly sends packets directly to MN’s CoA indicated in the binding. In MIPv6,
DAD (Duplicate Address Detection) procedure is invoked to determine the uniqueness of the new
MN’s CoA in which a MN sends a neighbour solicitation message with a set timer to ask that this
address is being used or not. If no node replies with-in the set time then MN can assume that this
address is unique in that network and it could use this address.
2.3. Distinction Between MIPv4 and MIPv6
1. Route Optimization process is a fundamental operation in MIPv6. In MIPv4, this feature is an
extension which may not be supported by all nodes.
2. Address Auto-configuration is also basic part of the MIPv6 which leads to removal of FA
which is used in MIPv4.
3. Packets are tunnelled using a routing header in MIPv6 where as MIPv4 uses IP encapsulation
for all packets. Using routing header reduces overhead which requires less additional header bytes
to be added to a packet at the time of sending packets.
4. Security is the prime concern in MIPv6 which utilizes IP Security (IPsec), where as MIPv4
utilizes mobility security association and relies on its own security mechanism for all these
activities [11].
140
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
3. MOBILITY MANAGEMENT
Different components for mobility management are as follows:
3.1. Handover Management - Mobility support handover management reduces the service
interruption during the handover. In Mobile IP handover latency represent the time between the
last received packets from the old network until the first received packet from the new network
[12]. In case of high handover latency, large number of packet could be lost. Packet losses could
cause critical disruption for real time services. Thus packets should be routed with low latency as
possible by IP routing and thereby alleviating packet loss during handover [13].
3.2. Location Management - Location management is done by the network to find out the
current mobile node’s location and keep tracking its movement by using movement detection
algorithm[14]. Movement detection algorithms have a role of optimizing Mobile-IP handover by
reducing the registration delay. In Mobile-IP there are two types of movement detection
algorithms:
3.2.1. Advertisement Based Algorithm(ABS)[15] - This depends on periodic broadcasts from
mobility agents. ABS has two distinct algorithms are:
3.2.1.1. Lazy Cell Switching (LCS) expects that movement of MN is rare and thus it avoids
handover until it is absolutely necessary. Consequently LCS is always slow to adapt the mobility.
3.2.1.2. Eager Cell Switching (ECS) assumes frequent location changes and perform immediate
handover upon discovering a mobility agent thereby making movement detection time negligible.
Accordingly it is fast to adapt mobility.
3.2.2. Hint Based Algorithm(HBA)[14] - It requires information from the link layer termed as
hints in order to perform movement detection.HBA has two distinct algorithms are:
3.2.2.1. Hinted Cell Switching (HCS) is proposed to extend the amount of information
communicated from the link layer to MIP and to include information about the environment as
identity of the local mobility agent. So it reduces movement detection time and Mobile-IP
handover delay.
3.2.2.2. Fast Hinted Cell Switching (FHCS) allows link layer to send triggers to network layer
whenever handover occurs. So it is able to reduce handover latency by denying the need for
movement detection and identity of local mobility agent.
3.3. Multihoming - Multihoming is a special case of a mobility management in which the
mobile device can use many access networks for example GPRS and Wi-Fi to access the internet
and switch the network while moving[16]. Multihomed Mobile-IP provide MN to register
multiple CoA at the HA to achieve more reliable connectivity.
3.4. Security - Security needs are getting active attention as wireless environment is potentially
more vulnerable to attacks including passive eavesdropping, active reply attacks, insider attack
and Denial of Service (DoS) attacks [2] based on the Mobile-IP registration protocol. So key
management is strongly desired in order to preclude aforementioned attacks. In Mobile IPv4
mobility security association is considered while Mobile IPv6 uses IPsec.
141
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
4. NETWORK MOBILITY
Network Mobility (NEMO) is proposed to support mobility in mobile networks[17]. Two aspects
of mobile networks are host mobility and network mobility. Host mobility has a scope of only
single node which is connected and network mobility is concerned with entire network. NEMO
introduces an important device termed as Mobile Router (MR) which acts as a gateway for the
mobile networks to configure a connection to the mobile nodes. The mobile nodes are (i) Local
Fixed Nodes (LFN) which cannot move and have the same home agent as the MR has, (ii) Local
Mobile Nodes (LMN) which can move and belong to the mobile network as its home network,
(iii) Visiting Mobile Nodes (VMN) which do not belong to the mobile network and attached to
the mobile network as a temporary basis. IETF standard for NEMO is NEMO Basic Support
Protocol (BSP), has advantages like reduce signalling and increased manageability, but also have
disadvantages like inefficient route and increased handover latency. To solve the limitations of
the NEMO BSP a set of NEMO Route optimization schemes are introduced[18]. Route
Optimization (RO) is a solution for providing improved end-to-end path between CN and MN,
reduce signalling overhead and packet loss. In [19], number of RO schemes have been introduced
to overcome aforementioned disadvantages.
4.1. Delegation - In this RO scheme, prefix of the foreign network is delegated inside the
mobile network. Mobile Network Nodes (MNNs) obtain their CoAs from received prefixes.
Then the obtain CoA which send BUs (Binding Updates) to HAs and CNs. Therefore CNs have
BU of MNN’s CoAs, so packets are sent directly to the foreign network without considering HAs.
Delegation based approach provides optimal route with low header overhead[20].
4.2. Hierarchical - In this scheme a packet reaches the foreign network either from MNN’s HA
or carried through HA of MNNs and Top Level Mobile Router (TLMR)[21]. Packets sent by CN
to MNN, using MNN’s HoA, reaches MNN’s HA that tunnel packets to TLMR’s CoA or HoA.
Thus packets which are tunneled using CoA will directly go to corresponding foreign network
whereas the packets, which are tunneled to HoA will go to the TLMR’s HA and further TLMR
sends them to MNN using MRs that maintain a routing table which contains MNN’s prefix. In
this scheme one tunnel always exists between the TLMR and VMN’s HA, so it reduces signaling
and is easily deployable.
4.3. Source Routing - RO has been achieved through CN by inserting CoAs of MRs in the
packet header itself so that each packet knows the underlying network structure made-up of MRs.
Packet are sent from CN to TLMR without going through HAs using CoA of MRs which lies in
packet header, thanks to source routing. In this scheme memory requirement is low but header
overhead is increased[22].
4.4. BGP Assisted - This scheme of RO is originated in Border Gateway Protocol (BGP), in
this scheme BGP routers are always updated by using forwarding entries for the prefix of the
mobile network in the routing table when the mobile network moves. This information about the
mobile network moves, is flagged to few routers that swap the information containing routing
entries to forward packets to the mobile network with each other using routing protocol through
internet[23]. Signaling is reduced but scalability is increased in maintaining routing entries.
5. ENHANCED HANDOVER SCHEMES IN MIPV6
Handover Delay - Handover delay is considered as time taken to redirect the on-going
communication from previous to current point-of-attachment[3]. Moreover handover delay is
composed of two types of delays. (i) Registration Delay - This delay is considered as time taken
during the HA registration process, (ii) Resolution Delay - This delay is considered as time taken
142
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
[29], when MN configure a new location CoA, if it is in the foreign network. To overcome these
delays which provided interruption in communication many handover approaches has been
proposed by several authors which are described in next sections. Also a comparison between
these techniques based on many factors like handover latency, packet loss, signaling overhead etc
is presented at the end.
5.1. Hierarchical Mobile IP (HMIP)[24] - To address the problem of HA registration delay,
many hierarchical networks have been proposed in which internet is separated into different
administrative domains. Movement of MN with a single administrative domain is called micro
mobility while movement across different administrative domains is called macro mobility. In
hierarchical Mobile-IP Mobility Anchor Point (MAP) is used as a router that maintains the
binding process for the mobile nodes currently visiting its domain. The MAP is considered as a
HA of the MN. MAP intercepts the packets targeted to the MNs addresses inside the domain and
then tunnels them to the correspondent CoA of the MNs in their foreign network. When MN
moves inside the domain it register their CoA to MAP only, there is no need to inform the HA so
this is called Local Care of Address (LCoA) for inside domain movement but when MN moves to
a new MAP domain it obtains Regional Care of Address (RCoA) for outside domain movement
[25]. After obtaining address the MN sends a binding update to the MAP which will bind the
MN’s RCoA to its LCoA. MAP then sends binding acknowledgement to MN for informing
successful registration. One more binding update is sent to MN’s HA when MN changes the
entire MAP domain. So such network reduces signaling overhead as well as handover delay by
reducing home agent registration when MN moves inside the domain[26][37].
143
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
Various HMIP based mobility protocols are; (i) Paging HMIPv6 (PHMIP) [27], proposes paging
services in MAP domain which provides information to MAP domain about the MN when it
moves in an in-active mode (no active communication session) and determine exact location of
MN using paging criteria thus it reduces power consumption, (ii) Robust HMIPV6 (RHMIP) [28],
MN registers with two different MAPs known as Primary MAP (P-MAP) and Secondary MAP
(S-MAP) simultaneously. When MN or CN detects a failure of P-MAP, it changes its attachment
from P-MAP to S-MAP. Hence it is more robustness and resilient by improving failure recovery
time. On the other hand it results in increased signaling overhead, (iii) Mobility Based Binding
Update HMIPV6 (MBBUHMIP) [29], provides lifetime value of binding cache and
introduceslocation update of MN by reducing signaling cost, (iv) Multilevel Hierarchy HMIPV6
(MHHMIP) [30], uses tree structure hierarchy of MAP thus providing scalable service but suffers
from extra packet processing overhead, (v) FF-HMIP (FF-HMIP) [31], based on HMIP prevents
global handover signaling by appointing a MAP and uses a fast MIP that reduces handover
latency by link layer trigger. Hence, it achieves improved handover performance and signaling
overhead at the cost of additional tunneling header, (vi) HMIP over Multiprotocol Label
Switching (HMIP-MPLS) [32], provides mobility and multimedia service by merging radio
access network with HMIP without any alteration in HMIP protocol, so signaling overhead is
increased due to two merged protocols. Summary of comparison is given in Table-1.
Table 1. Comparison between significant techniques based on HMIP [33]
HMIPv6
Protocol
PHMIP
(2003)
RHMIP
(2003)
MBBUHMIP
(2003)
MHHMIP
(2004)
FF-HMIP
(2004)
HMIP-MPLS
(2007)
Vantages
Drawback
Signaling overhead and power
consumption is reduced.
Obtain robustness and fault
tolerance.
Reduce
signaling
overhead
adjusted MN’s lifetime using
MN’s mobility pattern.
Supports multi level hierarchal
structure.
Improves signaling overhead and
handover performance.
Supports multilevel protocol
switching over HMIPv6.
Increase handover latency due
to inter domain movement
Signaling overhead due to
multiple registration
Increase binding update and
signaling cost
Signaling
Overhead
Low
High
Low
Packet
processing
and
signaling overhead
Introduce tunneling overhead
High
Additional signaling overhead
High
Low
5.2. Fast Handover Mobile IP (FHMIP) - To address the problem of FA address resolution
delay, FHMIP has been proposed in which MN will pre-configure a new CoA when it moves
from old Access Router (oAR) to new Access Router (nAR). It has three different types are:
5.2.1. MN initiated handover – When fast handover is about to occur, it is MN that gets first
notification from link layer (L2) information. Accordingly MN sends a Router Solicitation for
Proxy (RtSolPr) message to oAR as well as to new access node. Along with RtSolPr message MN
send sent link layer address to new access node. In response, oAR sends the Proxy Router
Advertisement (PrRtAdv) message to MN, which provides information about the new access
node that includes link-layer address and prefixes. On receiving PrRtAdv message MN decides a
prospective CoA based on prefix of selected nAR. Further MN sends Fast-Binding Update
(FBU) to the oAR and in response oAR sends Handover Initiation (HI) message to nAR for
imminent handover [34][35]. After that nAR returns a Handover Acknowledgement (HAck)
message to oAR in order to establish a binding between old CoA (oCoA) to new CoA (nCoA). In
144
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
response of HAck, oAR sends duple Fast Binding Acknowledgement (F-BAck) to MN and nAR
for forwarding the MN’s traffic towards nCoA. The nAR buffers the packets until MN establishes
link connectivity with the nAR. The MN sends a Fast Neighbor Advertisement (F-NA) to inform
the nAR of its presence and finally nAR sends the buffered packets to MN.
5.2.2. Network Initiated Handover - In such type of handovers, networks are made capable of
initiating handovers. However process of message exchanging is slightly different. PrRtAdv
message is sent by an oAR in an unsolicited way to the MN which contain the information
(configuring CoA) about the new networks in absence of initial RtSolPr message[36].
5.2.3. Reactive Handover – Unlike MN initiated and network initiated handovers, the oAR does
not receive FBU from MN before connectivity ends. Therefore HI, Hack and F-BAck messages
are not present. MN sends FBU to nAR by encapsulating with the Fast Neighbor Advertisement
(FNA) message. Further, nAR send this FBU to oAR. oAR then allows to create a binding
between oCoA and nCoA. Further the oAR forwards the MN’s traffic to the nAR and nAR in
term send the traffic to MN. FHMIP uses wireless link layer (L2) trigger based information for
smoothing of handover procedure and minimizing the FA resolution delay [37].
Some of the important research done over FHMIP are discussed below: (i) Fast MIPv6 (FMIPv6)
[38], provides seamless handover by making use of layer-2 trigger to obtain new link CoA while
still being connected to the previous link in order to reduce packet loss, (ii) Simultaneously
Binding Fast Handover (SBFHMIPv6) [39], provides simultaneous binding function at the MN.
MN’s traffic is multi casted to current location as well as to the locations where MN could roam
in near future, (iii)Seamless Multicasting Fast Handover (SMFHMIPv6) [40],provides integrated
unicast and multicast handover with combination of fast handover that creates seamless multicast
handover, [41], (iv) Pre-Binding Fast Handover (PBFHMIPv6) presents a modified version of
FMIPv6 using extra binding updates such as pre-binding update and pre-binding
acknowledgement between nAR and oAR. Thus there is no need to established reverse tunneling
between nAR and oAR, (v) Early Binding Fast Handover (EBFHMIPv6) [42], provides EBFH in
which an MN completes its binding update with current access router before link-going-down
trigger (i.e. MN is close to handover), (vi) Simplified Fast Handover (SFHMIPv6) [43],
significantly increases the probability that the protocol can successfully perform the fast handover
procedure in predictive mode which MN cannot complete due to lack of time in FMIPv6 version.
SFHMIPv6 also reduces anticipation time. A tabular summary is given in Table- 2.
Table 2. Comparison between significant techniques based on FHMIP
FHMIPv6
Protocol
FMIPv6(2005)
SBFHMIPv6(2006)
SMFMIPv6(2006)
PBFHMIPv6(2006)
EBFHMIPv6(2006)
Vantages
Drawback
MN perform fast handover
in predictive mode, So no
packet loss
Additional signaling overhead
due to additional signaling
message are required for
handover
Protocol enables to decouple
L2 and L3 handover, so
signaling overhead
Additional signaling message
Provides
simultaneous
binding to reduce packet
loss
Packet processing overhead
reduce due to air interface
Remove tunneling
Provides fast handover for
fast moving nodes
Extra binding update create
signaling overhead
Consumes large amount of
network performance and
creates overhead
Handover
Delay
High
High
Still High
High
Comparative
Low
145
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
SFHMIPv6(2008)
Supporting high speed MN
movement in predictive
mode
Reduce signaling cost and
packet delivery cost
Low
5.3. Seamless Handover Mobile IP (SH-MIP) -Seamless handover is an improved version
handover which is based on hierarchical network and fast handover. The main aim of this
handover scheme is to reduce packet loss by using Synchronized Packet Simulcast (SPS) (packets
are broadcast on both oAR and nAR) and hybrid handover mechanism (tracking of MN’s current
location and its signal strength). In seamless Mobile-IP a new entity introduced is Design Engine
(DE) which have mainly four functions, (i) To control handover process, (ii) Take decision for
handovers at the network domain, (iii) Keeps location tracking of all the mobile nodes by
identifying the movement modes (linearly, stochastically, stationary), (iv) Offers load balancing
when MN connects with lower load access routers [44]. The seamless handover occurs when MN
wants to go to a new network. When MN receives beacon advertisement message from adjacent
nAR, then it sends RtSolPr message to oAR for initiating the handover. oAR then sends HI
message to adjacent nAR which contain nCoA and oCoA. In response, HAck is send by the nAR
to oAR for establishing a binding between oCoA to nCoA. Further oAR sends Carrying Load
State (CLS) message to DE periodically which indicates the number of MN’s related to the AR
and their IP addresses. MN also sends Current Tracking State (CTS) message to DE periodically
when it receives beacon advertisement message from nAR which indicates the signal strength of
nAR. After determining CLS and CTS messages and tracking the mobile node movement, DE
sends Handover Decision (HD) to all ARs, following which oAR sends Handover Notification
(HN) message (which is extracted from HD) to MN that indicates the MN to which nAR it must
handover. In response, MN sends F-BU to oAR in order to bind its link address with nCoA, after
that oAR send Simulcast (Scast) message to MAP which initiates simulcasting of packets (i.e.
duplication) and sending the packets to oAR and nAR’s cache buffer at the same time. oAR and
MAP sends F-BAck to both current and new networks for ensuring reception of its message. MN
sends F-NA message to nAR when it connects to the new link and nAR forwards packets to MN.
At same time oAR also forwards the packets to nAR. On completion of packet sending from oAR
to MN through MAP, nAR sends Simulcast off (Soff) message to the MAP and MAP forwards
this message to DE which indicates that MN does not execute another seamless handover process
until current handover process is not completed.
Significant work has been done an SH-MIP, some of them are summarized here: (i) Adaptive SH
over video streaming (ASHMIPv6-VS) [45], presents an adaptive mobile video streaming scheme
for dynamically establishing network conditions. MN always buffers frames for disruption in
connectivity during handover so it is easy for streaming media server to adapt the video being
streamed to MN during handover to support seamless mobility, (ii) SH for Proxy MobileIPV6
(SH-PMIPV6) [46], it is a network based approach to control mobility management on behalf of
the MN so that MN is not required in order to provide any information about the target network,
(iii) Optimized SHMIPv6 (OSHMIPv6) [47], uses dynamic distributed algorithm which belongs
to the b-matching problem to select regional MAP that achieves peer-to-peer communication
mode in handover process, (iv) SHMIPv6 based on cellular network (SH-CN) [48], allows MN to
utilize their oCoA on the new link. It provides not only expedited forwarding of packets to MN
but also accelerated forwarding packets to their correspondents, (v) SH for IP Multimedia
Subsystem over MobileIPv6 (IMS-SHMIPv6) [49], presents context transfer mechanisms based
on predictive and reactive schemes. It also provides QoS provisioning for improvement of the
service quality of IP Multimedia Subsystem(IMS), (vi) Secure Password Authentication
Mechanism for SHPMIPv6 (SPAM-SHPMIPv6) [50], introduces a modified version of SHPMIPv6 that provides high security, resists various attacks (forgery attack, reply attack, stolen
verified attack) and performs authentication procedure by using bi-casting scheme based on
piggy-backing technique to reduce packet loss. Table- 3 provides comparative summary of all
techniques under SHMI. Finally a comparison between all the broad categories is presented in
Table- 4.
146
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
Table 3. Comparison between significant techniques based on SHMIP
SHMIPv6
Protocol
ASHMIPv6VS(2006)
Vantages
SHPMIPv6(2008)
OSHMIPv6(2011)
SH-CN(2011)
IMSSHMIPv6(2012)
SPAMSHMIPv6(2013)
Drawbacks
Handover Delay
Support cross layer
approach to adapt the
changes in the network
condition
Avoids on-the-fly packet
loss while ensuring the
packet sequence
Extra frames are buffered
Handover delay is
minimized
Suffers from packet
buffering in order to
perform packet ordering
Reduce packet loss and
improve handover
performance
Provides Pre-configure bi
directional secure tunnels
to accelerate mobility
management
Introduce IP multimedia
subsystem for real time
application
Signaling cost is
increased
By using neighbor
discovery message
handover latency is
reduced
Reduce handoff
latency
Avoiding packet loss
problem and reduces
signaling overhead
Introduce tunneling key
overhead
Introduce signaling
message overhead for reregister and re-invite of
MN for re-establishment
of the session
Memory requirement is
increased due to
buffering
Handover delay is
reduced in both inter
domain and intra
domain movements
Reduce handover
latency
Handover latency is
minimized
Table 4. Comparison between significant Handover Techniques
Handove
r
schemes
MIPv4
MIPv6
HMIPv6
FHMIPv6
Handove
r
schemes
SHMIPv6
Handover
Latency
Packet
Loss
Signaling
overhead
Long
High
High
Lengthy
handover
delay
Moderate
Moderate
High
High
Low using L2
trigger
restricted
under
movement
speed of the
MN
Handover
Latency
Moderate
Low due
to
signaling
overhead
Low
Packet
Loss
Signaling
overhead
Low
Low
Low
Route
Optimizat
ion
Optional
process
In-built
process
In-built
process
In-built
process
Route
Optimizat
ion
In-built
process
Deployment
FA is deployed in
MIPv4
No extra
functional
component is used
Gateway foreign
agent and
Regional foreign
agent is used
No extra
functional
component is used
Deployment
Design Engine is
used
Packet
Bufferin
g
No
No
Yes
Yes
Packet
Bufferin
g
Yes
147
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
6. GENERAL SOLUTIONS
6.1. Buffering - Buffering is a general solution deployed for avoiding packet loss during
handover. In this scheme when handover occurs, all the packets which are destined to previous
FA for the MN are forwarded to new FA by using buffering which happens by notifying the
CoA of new FA [51].
6.2. HAWAII - Handoff Aware Wireless Access Internet Infrastructure (HAWAII) is a
solution for improving handover latency. It is very similar to the hierarchal network but in this
scheme packets are routed in intra-domain and route optimization strategies are also
implemented in order to reduce handover delay [52].
6.3. Exclusive Handover Message (EHM) - EHM is another solution for improving packet
loss. This scheme gives end-to-end approach and improves the bust effect of host mobility on
TCP performance in wireless network. It calculates timeout at the Base Station (BS) when
handover occurs. This information about the handover is easily acquired by receiving router
advertisement occurs. This information about the handover is easily acquired by receiving
router advertisement message from new base station so BS sends EHM to fixed node to avoid
retransmission of packets at fixed node[53].
6.4. Mobile IP Fast Authentication Protocol (MIFA)[54] - MIFA is used to solve
handover latency. This scheme is based on local authentication with the new FA and
independent of re-authorization with the HA. MIFA uses security associations like MN-HA,
MN-FA which adds extra security between the connections which enables the FA to
authenticate the MN. Unlike hierarchical MIP it does not require hierarchical of FA’s.
6.5. Enhanced Mobile IP (E-MIP) - E-MIP is a solution for improving handover latency
and packet loss. It improves handovers through link layer information which allows an MN to
predict the loss of connectivity before connection is lost. A forceful handover is made to new
network even before any mobility is detected at network layer [55]. So it reduces handover
latency by eliminating the time required for handover detection at the network layer when
mobility occurs. So fast and seamless handover is achieved if MN’s moving speed is not high or
not low .
7. SUMMARY
The luxury of seamless connectivity and interruption free access to the internet anytime and
anywhere to users requires network to ensure that mobile node remains attached with globally
known permanent IP address even on a move and packets are delivered correctly without loss
during transit. An overview and comparative study of Hierarchical Mobile IP, Fast handover,
Seamless handover is presented. The global aim of all techniques is to remove packet loss, end
to end delay, handover latency and signaling load resulting in smooth handover.
REFERENCES
[1]
C. Perkins, editor.“IP Mobility Support for IPv4.” RFC 3344, Internet Engineering Task Force,
August 2002.
[2]
J. Chandrasekaran.“Mobile IP: Issues, Challenges and Solutions.” available at: http://bcs.adishare.com /contents / Mobile_ip_ Issues _ challenges_and_solutions913.pdf. 2009.
[3]
M. Alnas, I. Awan, D.R Halton.“Handoff Mechanism in Mobile IP.” IEEE, 2009.
148
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
[4]
Q. Zhao, L. Feng.“Packet Loss Analysis in Mobile IP.” in Springer addition: HSNMC 2004 pp.-524532, 2004.
[5]
D. Lee, J. Kim.“Out-of-Sequence Packet Analysis in Mobile IP Handoff and its Enhancement.” Proc.
3G Wireless 2, 2002.
[6]
Yi-an Chen.“A Survey Paper On Mobile IP.”, available at: http://www.cis.ohio-state.edu/ jain/cis
788-95/mobile_ip, 1996
[7]
J. Schiller.“Mobile Communications.”, ISBN 978-81-317-2426-2, Pearson education, 1996.
[8]
H. Chen, L. Trajkovic.“Simulation of Route Optimization in Mobile IP.” Proc.LCN 2002, 27th
Annual IEEE Conference on IEEE, 2002.
[9]
C. Perkins, D.B Johnson.“Route Optimization in Mobile IP.” draft-ietfmobileip-optim-11.txt,
September, 2001.
[10] J. Xie, U. Narayanan.“Performance Analysis of Mobility Support in IPv4/IPv6 Mixed Wireless
Networks.” IEEE Transactions on Vehicular Technology, Vol.59, No.2, Feb 2010.
[11] F. Nada.“Performance Analysis of Mobile IPv4 and Mobile IPv6.” The International Arab Journal of
Information Technology, Vol.4, No.2, April 2007.
[12] L. Niesink.“ A Comparison of Mobile IP Handoff Mechanisms.” 6th Twente Student Conference on
IT, Enschede, Feb2, 2007.
[13] M. Alnas, I. Awan, R. Holton.“A Survey of handoff Performance in Mobile IP.” Third UKSim
European Symposium on Computer Modeling and Simulation, IEEE 2009.
[14] N. Fikouras, C. Gorg. “A Complete Comparison of Algorithm for Mobile IP Handoffs with Complex
Movement Pattern and internet audio.” 4th International Symposium on Wireless Personal
Multimedia Communications (WPMC),Aalborg, Denmark, Sept. 2001.
[15] A. Hasson, N. Ventura.“An Evaluation of Mobile IP Movement Detection.” 2004.
[16] X. Jin.“A Survey on Network Architectures for Mobility.” Sep.5, 2006.
[17] C.J. Bernardos, D. Hugo et.al.“NEMO: Network Mobility. Bringing ubiquity to the Internet access.”,
demonstration at IEEE Infocom, 2006.
[18] P.C. Newton, L. Arockiam.“Route Optimization Mechanisms for Internet Applications in Mobile
Networks: A Survey.” International Journal of Future Generation Communication and Networking
Vol. 3, No. 3, September 2010.
[19] A.Z.M. Shahriar, M. Atiquzzaman.“Route Optimization in Network Mobility: Solutions,
Classification, Comparison, and Future Research Directions.” IEEE Communications Surveys &
Tutorials, VOL. 12, NO. 1, First Quarter 2010.
[20] R. Qureshi, A.Z.M. Shahriar, M. Atiquzzaman. “Performance of Prefix Delegation-based Route
Optimization Schemes for NEMO.” Military Communications Conference, 2008. MILCOM 2008.
IEEE, pp.1-7. IEEE, 2008.
[21] M. S. Jeong, Y. H. Cho, J. T. Par.“Hierarchical mobile network binding scheme for route optimization
in NEMO.” Wireless Personal Communications 43.3 (2007): 975-995.
[22] Y. Kim et al. “Local source routing based route optimization in nested mobile networks.”
Computational Science and Its Applications-ICCSA 2006. Springer Berlin Heidelberg, 2006. 730739.
[23] F. Mimoune, et al. "Route optimization for large scale network mobility assisted by BGP." Global
Telecommunications Conference, IEEE, 2007.
[24] A.H.A Hashim, et al. “Mobility Issues in Hierarchical Mobile IP.”, 2005.
[25] J. Zhang. “Cross-Layer Analysis and Improvement for Mobility Performance in IP-Based Wireless
Networks.” Communications Research Group Department of Electronics University of York,
September 2005.
149
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
[26] T.C. Hung, et al.“Improving handover performance in Mobile IPv6.” Advanced Communication
Technology, 2008. ICACT 2008. 10th International Conference on. Vol. 3. IEEE, 2008.
[27] J.S. Lee, et al.“Paging extension for hierarchical mobile IPv6: P-HMIPv6.” Networks, 2003.
ICON2003. The 11th IEEE International Conference on. IEEE, 2003.
[28] T. You, S. Pack, and Y. Choi.“Robust hierarchical mobile IPv6 (RH-MIPv6): an enhancement for
survivability and fault-tolerance in mobile IP systems.” Vehicular Technology Conference, 2003.
VTC 2003-Fall. 2003 IEEE 58th. Vol. 3. IEEE, 2003.
[29] M. K. Yi and C. S. Hwang.“A Mobility-Based Binding Update Strategy in Hierarchical Mobile Ipv6.”
dpnm.postech.ac.kr/papers/DSOM/03/118 yunmnmayuy.pdf.
[30] S. Pack, M. Nam and Y. Choi.“A Study on Optimal Hierarchy in Multi-Level Hierarchical Mobile
IPv6 Networks.” p.p. 1290-1294, Vol. 2, IEEE GLOBECOM 2004.
[31] Y. Gwon, J. Kempf and A. Yegin.“Scalability and Robustness Analysis of Mobile IPv6, Fast Mobile
IPv6, Hierarchical Mobile IPv6, and Hybrid IPv6 Mobility Protocols Using a Large scale
Simulation.” p.p. 4087-4091, Vol. 7, Issue, 20-24 ICC.2004.
[32] V. Vassiliou, A. Pitsillides.“Supporting mobility Events within a Hierarchical Mobile IP-over-MPLS
Network.” p.p. 61-70, vol. 2, No. 2, Journal of Communications,2007.
[33] S. Saha, et al.“Analysis of Hierarchical Mobile IP Based Fast Mobility Management Schemes.”
Networks and Communications, 2009. NETCOM'09. First International Conference on. IEEE, 2009.
[34] M. Ergen, et al.“Fast handoff with position information for mobile IP.” University of California,
Berkeley, 2002.
[35] R. Schmidiger, G. Zographos.“Fast Handover in Mobile IPv4 and IPv6.” April 28, 2005. Seminar:
Mobile Systems Talk No.4.
[36] M. Dunmore and T. Pagtzis. “of Deliverable: Mobile IPv6 Handovers: Performance Analysis.” March
2005
[37] N.T. Thuy An.“Research Handover on Mobile IP.” Multidisciplinary Journals in Science and
Technology, Journal of Selected Areas in Telecommunications (JSAT), July Edition, 2012.
[38] R. Koodli.“Fast Handovers for Mobile IPv6.” IETF, RFC 4068, July 2005.
[39] K. E. Malki, H. Soliman. “Simultaneous bindings for mobile IPv6 fast handovers.” IETF, draftelmalki-mobileip-bicasting-v6-06-txt, July 2005.
[40] G.A. Leoleis, G.N. Prezerakos, I.S. Venieris. “Seamless multicast mobility support using fast MIPv6
extensions.” Computer Communications 29 (18) (2006) 3745–3765.
[41] H.F. Chen, J. Zhang, “Prep-binding of fast handovers for mobile IPv6” draft-chen-mipshop-fasthandovers-prep-binding-02.txt, IETF, April 2006.
[42] H. Kim, Y. Kim.“ An early binding fast handover for high-speed mobile nodes on MIPv6 over
connectionless packet radio link.” Seventh ACIS International Conference on Software Engineering,
Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD’06), 2006, pp.237242.
[43] N. V. Hanh, S. Ro, and J. Ryu.“Simplified fast handover in mobile IPv6 networks.” Computer
Communications 31.15 (2008): 3594-3603.
[44] K. E. Aldalaty.“Mobile IP Handover Delay Reduction Using Seamless Handover Architecture.”,
Blekinge Institute of Technology August, 2009.
[45] M. K Mathews and N. Ventura.“An Adaptive Scheme Supporting Seamless Handovers over Mobile
IPv6 for Video Streaming Applications.” network 15: 17., 2006.
[46] J.E. Kang, et al. “Seamless handover scheme for proxy mobile IPv6.” Networking and
Communications, 2008. WIMOB'08. IEEE International Conference on Wireless and Mobile
Computing,. IEEE, 2008.
150
International Journal of Computer Networks & Communications (IJCNC) Vol.5, No.6, November 2013
[47] L. Zhuang, C. Wang, and W. Song.“Research of an Optimized Mobile IPv6 Real-time Seamless
Handover Technology” Journal of Networks 6.9 (2011): 1373-1378.
[48] L. Zhang, L. J. Zhang and S. Pierre. “Performance analysis of seamless handover in mobile IPv6based cellular networks” Intech 7.1 (2011): 1-26.
[49] R. Farahbakhsh and N. Movahhedinia.“Seamless Handover for IMS over Mobile-IPv6 Using Context
Transfer” arXiv preprint arXiv:1208.1207, 2012.
[50] M.C. Chuang, J.F. Lee, and M.C. Chen. “SPAM: A Secure Password Authentication Mechanism for
Seamless Handover in Proxy Mobile IPv6 Networks” (2013): 1-1.
[51] B. Ayani.“Smooth handoff in mobile IP.” Diss. University of California, 2002.
[52] R. Ramjee, T.L. Porta, et.al.“IP Mobility support using HAWAII.” Internet Draft, IETF Mobile IP,
2000.
[53] P. Bonde, B. Jharia, A. K. Shrivastav.“Performance evaluation of exclusive handover message TCP in
mobile IP based networks during handover.” Advanced Communication Technology (ICACT), 2010
The 12th International Conference on. Vol. 1. IEEE, 2010.
[54] A. Diab.“Minimizing Mobile IP Handoff Latency.” 2004.
[55] M. Alnas, M. Mansour.“Enhanced Mobile IP Handover Using Link Layer Information.” International
Journal of Computer Science and Security4.6: 598., 2011.
AUTHORS
Geetanjali Chellani She completed her B.Tech Degree in Electronics and
Communication Engineering from Rajasthan Technical University, Kota in year 2010. She
is pursing M.Tech in Communication and Signal Processing from Jaipur National
University, Jaipur, Rajasthan, India. Her areas of interest includes Wireless Networking,
Micro-controller, Digital Signal Processing, Digital Electronics, Circuit Analysis.
Anshuman Kalla is at present working as an Assistant Professor at department of
Electronics and Communication Engineering, Jaipur National University. He did
Bachelor's of Engineering (B.E.) from Engineering College Bikaner (Rajasthan
University) in 2004. He has pursued two funded Masters; First from ISEP, Paris, France
in 2008 and second from University of Nice Sophia Antipolis, France in 2011. In addition,
he has completed two research based internship one at Alcatel Lucent Technologies and
another at Orange Labs, France. He has worked on Genetic Algorithms and its implementation in networks,
Peer-to-peer video streaming, Delay Tolerant Network and Content Centric Networking.
151