Fast Mobile IPv6 Handover Using Link and Location Information

Mahmud Mansour & Mohamed Alnas
International Journal of Computer Science and Security (IJCSS), Volume (9) : Issue (3) : 2015 174
Fast Mobile IPv6 Handover Using Link and Location Information
Mahmud Mansour mansour30@hotmail.com
Faculty of Information Technology
Department of Network Engineering
Tripoli University
Tripoli, Libya
Mohamed Alnas M.J.R.Alnas@bradford.ac.uk
School of Informatics, Mobile Computing
Network and Security Research Group
University of Bradford
Bradford, UK
Abstract
There are two causes of latency in mobile handover: the move detection latency and registration
latency. This delay is inherent in the round-trip incurred by Mobile IP as the registration request is
sent to the home agent and the response sent back to the foreign agent. Throughout the time
between Mobile Node (MN) leaving the old foreign network (oFN) and Home Agent (HA)
receiving the MN registration message, HA does not know MN's latest Care of Address (CoA),
and therefore it still forwards the packets destined for MN to the old foreign network. These
packets will be discarded and lost.
This paper present an improved link layer mechanism with Location information Provider. Global
position systems GPS used to assist FMIPv6 for fast handovers and reduced packet loss during
handover. We introduce a new link layer combined with Location information Provider signalling
in this algorithm accordingly. Further, we report the implementation details performed through
simulations.
Therefore, link layer information and Location information Provider allows an MN and FAs to
predict the loss of connectivity more quickly than the L3 advertisement based algorithm. The
simulations evidence performance improvements in terms of latency and packet loss. It is also
shown that by enabling Location information Provider inside the FA discovery method and
improving link layer event services, an MN can be well prepared for handover and perform faster
movements.
Keywords- Mobile IP, Link Layer Information, Global Position Systems, Fast Handover,
Handover Latency, Packet Loss.
1. INTRODUCTION
The primary purpose of IP is to keep data packets delivering between hosts in the Internet.
Mobile IP is an Internet standards protocol, proposed by the Internet Engineering Task Force
(IETF), which enhances the existing IP to accommodate mobility [1, 2].
The most important functions in mobile IP is the addressing. Because host in the Internet must
has a unique IP address, which species its location. Such an address consists of a network
address and a host address. Mobile IP allows a MN to communicate with other nodes after
changing its link-layer point of attachment from one Access point to another without changing the
MN’s address [2,3]. The MN perform handovers between access Points while still using the
preserving IP Address. Therefore, packets may be routed to it using this address regardless of
the MN’s current point of attachment to the Internet [1,4].

Mobile IPv6 [1] is a protocol proposed to develop as a subset of IPv6 to support mobile
connections; Mobile IPv6 allows mobile nodes to change their point of attachment whilst not
breaking existing application sessions. Each MN is always identified by its home address,
regardless of its current point of attachment to the Internet. While situated away from its home, a
MN is also associated with a Care-of-Address (CoA), which provides information about the MN's
current location. Packets destined to the MN's home address are transparently routed to its CoA.
When the MN changes its point of attachment to the Internet, a handover occurs. The handover
mechanism provided in the Mobile IPv6 causes latency, which makes the MN unreachable for a
period of time [16, 17].
The protocol allows IPv6 nodes to cache the binding of a MN’s home address with its CoA, and
then to send any packets destined for the MN directly to it at this CoA. MIPv6 offers a solution to
solve the IP mobility, but due to intolerable high data lost rate and long handover latency. A new
protocol, called Hierarchical Mobile IPv6 has been proposed by the RFC 4140 [3] document,
which spread out Mobile IPv6 to allow for both micro mobility and macro mobility handling.
Hierarchical Mobile IPv6 (HMIPv6) proposal suggests a Mobile Anchor Point (MAP) to act as a
local HA to reduce signalling delays in handovers. However, the handover delays still remain
unacceptable for some applications.
When a MN changes its point of attachment to the network, it moves from one network (Old
network) to another new network and this process is known as handover. During the handover
process, the MN usually has disconnected from the old network before connecting to the new
network and thus there is a time when the MN has lost connectivity to the Internet. During this
period time, it cannot send or receive IP packets to maintain existing application sessions,
because of the link switching delay and this time period known as handover latency, it is the
primary cause of packet loss.
The latency due to a handover using basic MIPv6 is directly proportional to the minimum round-
trip time necessary for a binding update (BU) to reach either the home agent (HA), the
correspondent node (CN) or old location in case forwarding from old location is allowed. The
interruption time starts in the moment that the mobile node (MN) does not listen to the old location
anymore and finishes when the first packet arrives via the new location either from the HA, CN or
old location [2].Therefore, these packets may be lost and need to be retransmitted [3, 4].
There are two causes of latency in mobile handover: the move detection latency and registration
latency. This delay is inherent in the round-trip incurred by Mobile IP as the registration request is
sent to the home agent and the response sent back to the foreign agent. Moreover, there is a
high Mobile IP handover delay because of the agent discovery and registration periods,
eventually Mobile IP handover can cause significant performance degradation, especially in
large-scale mobility environments. Mobile IP can use link layer information to force a handover to
a new access network before any mobility at the network layer detected [2]. The Handover
decisions based on movement calculation eliminate the need to wait for beacon signals from
other FAs. In this paper, we propose the use of link-layer information combined with the global
position systems (GPS) in every FA (Location Information of FAs) that can detect the direction of
the MN to the new network agent. The link-layer trigger and the Location information of FAs
enhance the overall performance of the Mobile IP handover [9].
2. MOVEMENT DETECTION
Movement detection is one of the most important operations performed by the MN, because it is
used to discover the handover. To achieve this goal the MN will use any mechanism to detect its
movement from one link to another. The Standard movement detection mechanism defined in
Mobile IPv6 uses services defined in IPv6 Neighbor Discovery. Additional information provided by
other mechanisms can be used besides the one provided by the standard mechanism in order to
facilitate the movement detection [2]. The movement of the MN can be detected by using
Location information that install inside all FAs. Therefore, FAs can discover the direction of the

mobile node and the address of the new foreign agent that MN will move. This will reduce the
delay of the registration [10].
3. HANDOVER LATENCY
The Handover latency is the most important issue in mobility network. It refers to the ability of the
network to allow a call in progress to continue as the MN continues to travel and change its point
of attachment. The handover refer to the time between the reception of the last packet through
the old FA (oFA) and reception of the first packet through the new FA (nFA) [5, 6, 7]. During this
time the MN, start disconnecting from the old network and start new registration with the new
network while packet still forwards to the old network; these packets will be discarded and lost.
The packet losses could cause impossible disruptions for real-time services, degrade the QoS
and lead to severe performance deteriorations of upper layer protocols, especially when the
handover is frequent and the distance between MN and the HA is great [ 8,9,10].
In general, handover can be classified as either proactive or reactive. Proactive handovers utilize
link layer triggers to support the MN in determining that a handover is about to happen and
establish packet flow to the target access point prior to the handover event, i.e. requires link layer
coupling. This covers a hybrid of mobile assisted and mobile controlled handover types. Reactive
schemes only follow the base mobile IP movement methods [13].
4. LINK LAYER INFORMATION AND LOCATION INFORMATION PROVIDER
The main reason to use the link layer to improve the handover delay and packet lost. This can be
achieved by providing the information of the link layer; the MN can predict its connectivity more
quickly than Network Layer advertisement-based algorithms. Therefore, it used this information to
predict the breakdown of the link layer before is broken. This facilitates the execution of the
handover, and the elimination of the time to detect handover. [11,12].
The Location information Provider built inside of FAs is now being used in most of the mobile
networks to determine the location of any FAs address. GPS is used in different areas and is
becoming more commonly used because it is integrated in various devices. Moreover, it can also
be used in MN and other wireless access devices to facilitate good handover due to its accurate
location-trace [18].
Building Location information Provider in the MN and FAs means the MN and FAs are able to
track the position constantly. By using L2 and Location information Provider in MN and FAs, it is
possible to decrease latency and packet loss. Handover decisions based on movement
calculation remove the need to wait for beacon signals from other FAs and to discover handover
target areas in advance. The link layer and GPS information Provider used in this paper to reduce
the delay and Packet lost [20].
5. RELATED WORKS
In the past few years, different proposals have been presented to minimize the handover delay in
Mobile-IPv6 networks. Many of the proposed methods require modification of the Access Routers
(ARs). Two slightly different handover solutions using multicast routing are presented in [7] and
[8].
The Post-Registration proposal involve Link layer [L2 triggers]. [15] The handover method is
based on a network-initiated model of a handover, which does not require any MN involvement
until the actual Layer 2 (L2) connection with the new Foreign Agent (nFA) is completed. Such a
trigger is a signal related to the L2 handover process. Two types of triggers can be received: a
source trigger at the oFA (L2-ST) and a target trigger at the nFA (LS-TT). The first trigger that is
used is an early notice of an upcoming change in the L2 point of attachment of the MN, referred
to as anticipation trigger. A second trigger, the Link Down trigger (L2-LD), indicates that the L2
link between the MN and the oFA is lost. The Link Up trigger (L2-LU) occurs when the L2 link

between the MN and the new FA is established. A trigger initiated at the old FA is referred as a
source trigger and a trigger initiated at the new FA is referred as a target trigger.
This approach uses Bi-directional (BET) edge tunnels to perform low latency change in the L2
without the MN involvement. A handover occurs when the MN moves from the oFA, Where the
MN performed a Mobile IP registration to nFA. The MN delays its registration with the nFA, while
maintaining connectivity using the BET between the oFA and nFA.
The other proposal is Pre-Registration [15], realizes an anticipated L3 handover. This handover
method allows the MN to communicate with the new Foreign Agent (nFA) while still connected to
the Old Foreign Agent (oFA). This way, the MN is able to pre-build its registration state on the
nFA prior to an underlying L2 handover.
The network assists the MN in performing an L3 handover before the L2 handover is completed.
Both the MN (mobile-initiated) and the FAs (network-initiated) can initiate a handover. A mobil-
initiated handover occurs when the L2 anticipation trigger is received at the MN informing it that it
will shortly move to the nFA. The L2 trigger contains information such as the nFA’s IP address
[14].
The standard Mobile IPv6 procedures have to deal with the same handover latency problem as
Mobile IPv4. In [3], Koodli species a protocol to improve handover latency in Mobile IPv6 as [2]
does for Mobile IPv4.The Fast Handover method is an extension proposed for Mobile IPv6 and
resembles a combination of Pre-Registration and Post-Registration. The Fast handovers for
Mobile IPv6 [FMIPv6] Handover can be either Network-Initiated or Mobile-Initiated, depending on
whether one of the ARs or the MN initiates the handover. The two main possibilities are router
discovery performed by MN on Layer 3 and a link-specific event (L2 trigger) occurring in the MN
or in the network. In [8], HMIPv6 a proposal suggest to an extension to Mobile IPv6, which aims
to reduce the amount of signalling between the MN and its CNs during a handover, and to
improve the performance in terms of handover speed.
In an IETF draft, which expired in April 2006 [19], Jung et al. propose a combination of the Fast
Handovers and Hierarchical Mobile IP extensions to Mobile IPv6. The scheme is called Fast
Handover for Hierarchical Mobile IPv6" (F-HMIPv6). The MN enters a new MAP domain, it first
performs the HMIPv6 registration procedures with HA and MAP. Later, when the MN moves from
a PAR to a NAR within the MAP domain, it will follow the local Binding Update (BU) Procedure of
F-HMIPv6. During the handover, data packets sent by CNs will be tunneled by the MAP toward
the NAR via a bi-directional tunnel, similarly to the FMIPv6 procedure. Optionally, the MAP may
start bi-casting packets to PAR and NAR simultaneously. It should be noted that no bi-directional
tunnel is established between PAR and NAR.
6. PROPOSED ALGORITHM
The Predictive handover for FMIPv6 (P-HMIPv6) provides a different approach for resolving the
timing ambiguity problem. Link layer information such as signal strength is continuously available,
providing important information about the availability of new links, and the FAs will use the
location information’s of MNs and all neighbours (FAs) to facilitate handover. Therefore, the
handover can be predict in advance before MN moves out of the coverage area of the oFA. The
main idea behind the proposal is to apply link layer information and Location information of the
FAs to predict a breakdown wireless link before the link is broken. The use of Proactive will
significantly reduce handover latency and reduces packet loss in handover.
The proposal will consider the handover to start when it is predict that the link layer association to
the oFA will lost. The handover will completed when the registration reply message received from
the HA to the MN. Figure 1 describes the overall P-Mobile IP protocol message flow.

FIGURE 1: P-Mobile IP Protocol Message Flow.
1. The handover process starts when the MN become leavening the oFA and entering the
overlap area between nFA and oFA.
2. The Foreign Agent receiving an L2 trigger informing that MN is about to move from oFA
to nFA.
3. The oFA provided location information for nFA include MN home Agent address and that
include by oFA, which include the direction that based on Layer 2 information.
4. The oFA sends a handover Initiation (HI) message containing the MNs home Agent
address.
5. The nFA sends a handover Acknowledgement (H-Ack) message to oFA.
6. Tunnelling then establish between oFA and nFA.
7. The oFA will forward all packets received through the tunnel to the nFA.
8. MN then will receive fast acknowledgment (FB-Ack) via nFA.
9. The nFA will buffer packets that received from the oFA, and this packets will forwarding
after the MN send FNA message to ask nFA to forward all buffered packets,
7. SIMULATION SCENARIO and CONFIGURATION
In this section, we evaluate the performance proposed for the FMIPv6 using the link and location
information algorithm. We compare our algorithm against a Mobile IPv6 and Mobile IP We
assume that the MN on area (A) and start handover to (B or E) the area for the overlapping is
25m, the Handover decisions based on movement calculation eliminate the need to wait for
beacon signals from other FAs and also to discover handover target areas in advance.
We use network simulator CIMS NS-2 version ns-allinone-2.31 as a simulation tools in order to
simulate FMIPv6 handover [21, 22]. The simulator is modified to emulate IEEE 802.11 infra-

structured behaviours with multiple disjoint channels. This modification forces L2 handover
operations, where stations only receive data packets via one FA at a time.
The network features three MNs connected to it; the first will move sequentially from oFA to nFA,
starting at overlap of the nFA1, performing handovers at a rate of a 30 handovers/min. In each
test, the MN1 will be the receiver of a CBR or FTP traffic source, generating either UDP or TCP
packets. This traffic originates from the CN1 outside the network, or inside the domain from CN2.
All presented results are taken as the average of multiple independent runs, coupled with a 95%
confidence interval. The best possible handover point occurs at position A, as shown in Figure 2.
FIGURE 2: Overlapping Coverage Area.
8. PERFORMANCE ANALYSIS and EVALUATION
In our simulation, we use a 500m × 500m and a 1000m × 1000m area with a 3 to 7 MNs [5, 11].
The network bandwidth is 2 Mbps and the medium access control (MAC) layer protocol is IEEE
802.11 [19]. The packet size is 10p/s which will generate enough traffic when we increase the
number of connections for example at 40 connections of source-destination pairs, it will generate
400 packets per second for whole scenario. Other simulation parameters are shown in Table1.
These parameters have been widely used.
TABLE 1: Simulation Parameters.
Simulation parameter Value
Simulator Ns-allinone-2.31
Network range 600m×600m and
1000m×1000m
Transmission range 25m
Mobile nodes 3 and 5
Traffic generator Constant bit rate
Bandwidth 2Mbps
Packet size 512 bytes
Packet rate 10 packet per second
Simulation time 750s and 1100s

Figure 3 and 4 showing the relation between the handover latency and packet loss, as we
observe that the P-FMIPv6 performs better in terms of handover latency and packet loss
compared to the others, although the fast handover protocol is proposed and design to minimize
the latency and the packet loss during a handover while the worst case observed Mobile IP and
Mobile IPv6.
FIGURE 3: Impact of Handover Latency.
We observe in figure 4, that the number of packet loss increase with increase of CN, the P-
FMIPv6 showing better performance comparing to the wars case of MIP and MIPv6.
FIGURE 4: Impact of Packet Loss.
The overall throughput graph showing in Figure 5. The figure shows that as the number of
sending rate increase the throughput increase. The P-FMIPv6 performs better than all other
proposal. The reason for the throughput increase is that more packets are sent overall, although
the number of packets lost increase as the sending rate increase. The P-FMIPv6 slightly performs

well compared to the other three proposals. We can see that at rate of 10 the MIPv6, FMIPv6 and
P-FMIv6 very close.
FIGURE 5: Throughput Versus Rate.
Obviously, the loss in the buffer increases when the buffer size is increased. The number of
packets lost depends both on the size of the buffer used to store packets for potential handovers
and the sending rate as seen in figure 6. The number of packets lost increases for Mobile IP
since no buffer is used and increases as the sending rate increase since more packets are sent,
while MN is unable to receive them during handover. While on the other hand, the number of
packets lost decreases as buffer size increase for P-MIPv6 and FMIPv6.
FIGURE 6: Packet Loss vs Buffer Size.

The result graph shows the uplink of MN to CN transmission behaviour with sixe handover in unit
time of all four schemes. The result graph shows the transmission bit rate of each handover
protocol. The MIPv6 and Mobile IP receive less data than other schemes because their time
period take to finish the registration, while the FMIPv6 and P-FMIPv6 shows the highest
transmission rate.
FIGURE 7: Handoff Behaviour.
9. CONCLUSION
In this paper, we developed and analyzed the proposed scheme of the P-Mobile IPv6 handover
using link layer and location information scheme. The performance study in this paper indicate
that the use of link layer and location information helps to minimize packet loss and improver the
throughput. In our scheme, we analysed the performance by simulating the proposed scheme in
ns-2 to get fast mobile handover for FMIPv6.
We then compared the experimental results with the results of the Mobile IP and Mobile IPv6 and
FMIPv6. The performance study in this paper indicates that the use of link layer information with
location information helps to minimize packet loss and improve the throughput of Mobile IP
handover. We have seen that the starting point for packet loss could happen in two ways: first,
packets may get lost in the oFA when the forwarding buffer overflows and secondly, packets may
get lost in the nFA when, upon their arrival, the ReRep from the HA has not arrived in the nFA.
The first reason for loss may be avoided by appropriately dimensioning the forwarding buffer.
This buffer should be able to store arriving packets at least during a time equal to the delay on the
nFA and oFA path. The second loss is more difficult to deal with. It is determined by the
difference between the delays of the paths oFA, nFA and nFA, HA.
In addition, we evaluated the impact of L2 setup on different performance measures of Mobile IP,
together with handover latency, packet loss and throughput. The simulation results show that P-
Mobile IPv6 handover latency is not too sensitive to L2 setup latency and beacon periods
compared to the other schemes of Mobile IP. Moreover, P-Mobile IPv6 can achieve a fast and
seamless handover if MN’s moving speed is not too high, but is within reasonable limits.
10. REFERENCES
[1] D. Johnson, C. Perkins, and J. Arkko. ”Mobility Supportin IPv6”. Internet Engineering
Task Force (IETF), RFC-3775, June 2004.

[2] C. E. Perkins, “Mobile Networking through Mobile IP”, Internet Computing, IEEE, vol. 2,
pp. 58–69, 1998.
[3] H. Soliman, C. Castelluccia, K. E. Malki, and L. Bellier.”Hierarchical Mobile IPv6 Mobility
Management (HMIPv6)”. Internet Engineering Task Force (IETF), RFC-4140, August 2005.
[4] P. Bhagwat, C. Perkins and S. Tripathi, “Network Layer Mobility: an Architecture and
Survey”, IEEE Personal Communications, vol. 3, pp. 54–64, 1996.
[5] H. Balakrishnan, V. N. Padmanabhan, S. Seshan and R. H. Katz, “A Comparison of
Mechanisms for Improving TCP Performance over Wireless Links”, IEEE/ACM
Transactions on Networking, vol. 5, pp. 756–769, 1997.
[6] S. Mohanty and I. F. Akyildiz, “A Cross-layer (Layer 2 + 3) Handoff Management Protocol
for Next-Generation Wireless Systems”, Transactions on Mobile Computing, vol. 5, pp.
1347–1360, 2006.
[7] I. F. Akyildiz, X. Jiang and S. Mohanty, “A Survey of Mobility Management in Next-
generation All-IP-Based Wireless Systems”, IEEE Wireless Communications, vol. 11, pp.
16–28, 2004.
[8] I. F. Akyildiz, “Mobility Management for Next Generation Wireless Systems”, Proceedings
of the IEEE, vol. 87, no. 8, pp. 1347–84, August 1999.
[9] J. Puttonen, “Using Link Layer Information for Improving Vertical Handovers”, 16
th
International Symposium on Personal, Indoor and Mobile Radio Communications IEEE,
2005.
[10] H. Chung-Ming, C. Meng-Shu and L. Jin-Wei, “A Link Layer Assisted Fast Handoff
Scheme Using the Alternative Path Approach”, 20
th
International Conference on
Advanced Information Networking and Applications, 2006.
[11] F. Fang and D. S. Reeves, “Explicit Proactive Handoff with Motion Prediction for Mobile
IP”, 2004 IEEE Wireless Communications and Networking Conference, WCNC 2004,
vol. 2, pp. 855–860, 2004.
[12] S. Oh, H. Song and Y. Kim, “Seamless Fast Handover in Mobile IPv4 Using Layer-2
Triggers”, 2
nd
International Conference on Systems and Networks Communications,
ICSNC 2007, pp. 16–16. 2007.
[13] S. Thalanany, “Low Latency Handoffs in Mobile IPv4”, draft-ietf-mobileip-lowlatency-
handoffs-v4-04.txt, June 2002.
[14] K. El-Malki and H. Soliman, “Fast Handoffs in Mobile IPv4”, Internet draft, draft-emalki-
mobileip-fast-handoffs-03.txt, September 2000.
[15] S. Oh, H. Song and Y. Kim, “Seamless Fast Handover in Mobile IPv4 Using Layer-2
Triggers,” in Systems and Networks Communications, ICSNC 2007, 2
nd
International
Conference, pp. 16-16, 2007.
[16] R. Koodli and C. E. Perkins, “Mobile IPv4 Fast Handovers”, Internet draft, Internet
Engineering Task Force, draft-ietf-mip4-fmip, February 2006.
[17] R. Hsiehet, “S-MIP: a Seamless Handoff Architecture for Mobile IP”, Proceedings of
INFOCOM 2003, March 2003.
[18] G. Dommety and T. Ye, “Local and Indirect Registration for Anchoring Handoffs”, draft-
dommety-mobileip-anchorhandoff-01.txt, July 2000.
[19] Columbia University, Columbia IP Micro-Mobility Software,
http://www.comet.columbia.edu/micromobility/.index.html.
[20] G. Pollini, “Trends in Handover Design”, IEEE Communications Magazine, 34, 3, 80–90,
March 1996.

[21] S. Goswami, “Simultaneous Handoff of Mobile-IPv4 and 802.11”, Internet Draft, IETF,
draft-goswami-mobileip-simultaneous-handoff-v4- 02.txt, February 2003
[22] H. Chung-Ming, C. Meng-Shu and L. Jin-Wei, “A link layer assisted fast handoff scheme
using the alternative path approach,” in Advanced Information Networking and
Applications: 20
th
International Conference, pp. 5, 2006.

Fast Mobile IPv6 Handover Using Link and Location Information

More Related Content

What's hot (20)

Similar to Fast Mobile IPv6 Handover Using Link and Location Information (20)

Recently uploaded (20)

Fast Mobile IPv6 Handover Using Link and Location Information