Design and Performance Evaluation of an Efficient Home Agent Reliability Protocol

Description

Mobile IPv6 will be an integral part of the next generation Internet protocol. The importance of mobility in the Internet gets keep on increasing. Current specification of Mobile IPv6 does not provide proper support for reliability in the mobile network and there are other problems associated with it. This paper proposes “Virtual Private Network (VPN) based Home Agent Reliability Protocol (VHAHA)” as a complete system architecture and extension to Mobile IPv6 that supports reliability and offers solutions to other related problems. The key features of this protocol over other protocols are: better survivability, transparent failure detection and recovery, reduced complexity of the system and workload, secure data transfer and improved overall performance.

Shared by: ides.editor

Tags

Mobility Agents, VPN, VHAHA, Faulttolerance and Reliability.

Stats

views:: 16
posted:: 11/29/2012
language:
pages:: 7

Public Domain

Document Sample

ACEEE International Journal on Signal and Image Processing Vol 1, No. 2, July 2010

Design and Performance Evaluation of an
Efficient Home Agent Reliability Protocol
Rathi S1, and Thanushkodi K2
1
Government College of Technology/Department of Computer Science and Engg, Coimbatore, India
Email: sai_rathi@yahoo.com
2
Akshaya College of Engg. and Technology, Coimbatore, India
Email: thanush@rediffmal.com

Abstract— Mobile IPv6 will be an integral part of the next is known as “Mobility Binding”. In the scenario of the
generation Internet protocol. The importance of mobility in Mobile IP described above, it can be observed that the
the Internet gets keep on increasing. Current specification HAs are the single point of failure. Because all the
of Mobile IPv6 does not provide proper support for
communication to the MN is through the HA, since
reliability in the mobile network and there are other
problems associated with it. This paper proposes “Virtual
the Correspondent Node (CN) knows only the Home
Private Network (VPN) based Home Agent Reliability Address. Hence, when a particular HA is failed, all
Protocol (VHAHA)” as a complete system architecture and the MNs getting service from the faulty-HA will be
extension to Mobile IPv6 that supports reliability and offers affected. According to the current specification of
solutions to other related problems. The key features of this Mobile IP, when a MN detects its HA is failed, it has
protocol over other protocols are: better survivability, to search for some other HA and recreate the bindings
transparent failure detection and recovery, reduced and other details. This lacks the transparency, since
complexity of the system and workload, secure data everything is done by the MN. Also, this is a time
transfer and improved overall performance.
consuming process which leads to the service
Index Terms—Mobility Agents, VPN, VHAHA, Fault- interruption.
tolerance and Reliability. This paper proposes “Virtual Private Network
(VPN) based Home Agent Reliability Protocol
I. INTRODUCTION (VHAHA)” as a complete system architecture and
extension to Mobile IPv6 [1] that supports reliability
As mobile computing has become a reality, new and offers solutions to other related problems. The
technologies and protocols have been developed to key features of this protocol over other protocols are:
provide mobile users the services that already exist for better survivability, transparent failure detection and
non-mobile users. Mobile IP [1, 2] one of those recovery, reduced complexity of the system and
technologies that enables, a node to change its point of workload, secure data transfer and improved overall
attachment to the Internet in a manner that is transparent performance.The remainder of this paper is organized
to the application on top of the protocol stack. Mobile IP as follows: Section II discusses the other related work
based system extends an IP based mobility of nodes by done in this area. Section III, introduces the proposal.
providing Mobile Nodes (MNs) with continuous network Section IV presents the performance analysis and
connections while changing their locations. In other comparisons. Finally, conclusion is made in section
words, it transparently provides mobility for nodes while V.
backward compatible with current IP routing schemes by
using two types of Mobility Agents (MA), Home Agent II. RELATED WORK
(HA) and Foreign Agent (FA).
While HA is responsible for providing permanent Several solutions have been proposed in this area.
location to each mobile user, the FA is responsible for The proposals that are found in [3-8] are for Mobile
providing Care-Of-Address (COA) to each mobile user IPv4 and [9-15] are for Mobile IPv6 based networks.
that visits the Foreign Network. Each HA maintains a The architecture and functionality of Mobile IPv4
Home Location Register (HLR), which contains the and Mobile IPv6 are entirely different. Hence, any
MN’s Home Address, current COA, secrets and other solutions that are applicable for Mobile IPv4 can not
related information. Similarly, FA maintains Visitors be applicable for Mobile IPv6. This is due to the
Location Register (VLR) which maintains information following reasons. In mobile IPv4, single HA at the
about the MNs for which the FA provides services. Home Link would serve the MN. This makes the
When the MN is within the coverage area of HA, it gets Mobile IPv4 prone to single point of failure problems.
the service from the HA. If the MN roams away from the To overcome this problem, the Mobile IPv4 solutions
coverage of HA, it has to register with any one of the propose HA redundancy. But in Mobile IPv6, instead
FAs around to obtain the COA. This process is known as of having single HA to serve MNs, multiple HAs
“Registration” and the association between MN and FA from the Home link would serve the MNs. So, in

26
© 2010 ACEEE
DOI: 01.ijsip.01.02.05
ACEEE International Journal on Signal and Image Processing Vol 1, No. 2, July 2010

mobile IPv6 the HA will not be a single point of failure.

Figure 1. VHAHA Architecture

Solutions for Mobile IPv6 based networks are A.. Architecture Design
given in [9, 10, 11, 12]. In [9], one primary HA will The architecture of the proposed approach is given in
provide service to the MNs and Multiple HAs from Fig. 1. As part of Mobile IPv6, multiple Home Links are
different Home Links are configured as Secondary available in the network and each Home Link consists of
HAs. When the primary HA failed, the secondary multiple HAs. In this approach, one HA is configured as
HA will be acting as Primary HA. But, the Active HA, some of the HAs are configured as Backup
registration delay is high and the approach is not HAs and few other HAs are configured as Inactive HAs
transparent to MNs. The approach [10] is similar to [12] from the Home Link. The Active HA provides all
[9], but here all redundant HAs are considered from Mobile IPv6 services, the Inactive HA provides minimal
the same domain. The advantages of this approach set of services and Backup HA provides mid range of
are registration delay and overhead are comparatively services. VHAHA requires that for each MN there should
less. The drawback of this approach is that the Home be at least two HAs (one active HA and the other could
link is the single point of failure. be any one of the backup HA) holding its binding at any
The approaches [11] and [12] are similar to [10], instance of time. The functionalities of these HAs are
but they deal with load balancing issues also. In [13], given below:
the reliability is provided by using two HAs (Primary Active HA: There must be a HA on the Home Link
HA and Secondary HA) in the Home link. The two serving as the Active HA. Only one HA could act as
HAs are synchronized by using transport layer Active HA at any instance of time. The active HA
connections. This approach provides transparency maintains the Binding cache, which stores the mobility
and load balancing. Also, registration delay and bindings of all MNs that are registered under it. This will
service interruptions are less. But, if the Home Link hold [0-N] mobility bindings. This is responsible for data
is failed then the entire network will be collapsed. delivery and exclusive services [12]. The exclusive
To overcome the disadvantages of above said services mainly include Home Registration, De-
approaches, a novel architecture based on interlink registration, Registration, Registration-refresh, IKE and
HA redundancy is proposed in this paper. Since the DHAD. Besides these, it provides regular HA services
backup HAs are considered from different Home such as Tunneling, reverse Tunneling, Return Routability
links, single point of failure is not an issue here. and IPv6 neighbor discovery.
Moreover, the proposed architecture is compatible Backup HA: For each MN, there will be at least two
with the existing Mobile IP architecture. HAs will be acting as backup HAs (no limits on
maximum no. of HAs). The purpose of Backup HA is to
III. PROPOSED APPROACH provide continuous HA services in case of HA failures or
This proposed approach provides reliability by overloading. The back up HA could hold [1-N] bindings
introducing extension to the overall functionality and in its binding cache. This provides all the services of
operation of current Mobile IPv6. The advantages of Active HA except the exclusive services.
this protocol are reliable Mobile IPv6 operations, Inactive HA: Inactive HAs will not hold any Mobility
better survivability, transparent failure detection and Bindings and it provides only limited services from
recovery, reduced complexity of the system and Backup HA services since any HA in the Home Link can
workload, secure data transfer and improved overall act as Inactive HA.
performance.

27
© 2010 ACEEE
DOI: 01.ijsip.01.02.05
ACEEE International Journal on Signal and Image Processing Vol 1, No. 2, July 2010

The VHAHA is configured with static IP address MN appropriately using base Mobile IP. The various
that is referred as Global HA Address. The Global steps in forwarding the data packets are given in Fig. 2.
HA address is defined by the Virtual Identifier and a The packet formats are shown in Fig. 3.
set of IP addresses. The VHAHA may associate an
Active HA’s real IP address on an interface with the
Global HA address. There is no restriction against
mapping the Global HA address with a different
Active HA. In case of the failure of an Active HA,
the Global address can be mapped to some other
backup HA that is going to act as active HA. If the
Active HA becomes unavailable, the highest priority
Backup HA will become Active HA after a short
delay, providing a controlled transition of the
responsibility (Regular HA services and exclusive
services) with minimal service interruption. Besides
minimizing service interruption by providing rapid
transition from Active to Backup HA, the VHAHA Figure 2. VHAHA Scenario.
design incorporates optimizations that reduce
protocol complexity while guaranteeing controlled
HA transition for typical operational scenarios. The
significant feature of this architecture is that, the
entire process is completely transparent to MN. The
MN knows only the Global HA address and it is
unaware of the actual Active HA. It also does not
know about the transition between backup and active
HAs.
B. VHAHA Scenario
The steps to configure the Virtual Private
Networks are given in procedure 1. Two or more Figure 3. Packet Formats
HAs (One active HA and minimum of one backup
HA) from each Home Link are selected. Then VPN As in Mobile IPv6 [1], the CNs and MNs only know
[16] is constructed among the selected HAs through about the Global HA address. The packet (Fig. 3a)
the existing internetworking. This VPN is assigned addressed to the MN from CN (Fig. 2. Step 1) will be
with Global HA address and it will act as Global HA. directed to the Home Network using the Global HA
HAs of the VPN will announce their presence by address (Fig. 2. Step 2) of the MN. Here, the Home
periodically multicasting Heart Beat [12] messages Network refers to the VPN that is constructed by using
inside the VPN. So, each HA will know the status of the above procedure.
all other HAs in the Private network. Once the packet reaches the Global HA address, all
HAs that belong to Global HA address will hear the
_______________________________________
packet and the one which is closer to the MN and has less
___ workload will pick up (Fig. 2. Step 3) the packet (Fig.
Step 1: Select few HAs (≥2) from each Home Link.
Step 2: Set up VPN among the selected HAs. 3b) using the internal routing mechanism [15]. Then the
Step 3: Periodically multicast heartbeat messages among the HAs packet will be routed to the Active HA and this Active
of VPN. HA will do the required processing and tunnel the packet
(Fig. 3c) to the COA of the MN (Fig. 2. Step 4). Finally,
Procedure 1: VPN Configuration
the COA decapsulate and send the packet (Fig. 3d) to the
MN using base Mobile IPv6 (Fig. 2. Step 5).
The scenario of VHAHA protocol is given in Fig.
2. The protocol works at layer 3. In this approach, the C. Fault detection and Tolerance
HAs are located in different Home Links still share In contrast to Mobile IPv6 and other approaches, failure
the same subnet address as in [12, 15]. The shared detection and tolerance is transparent to the MN. Since
subnet address is known as Global HA address and the MN is unaware of this process, over-the-air (OTA)
the HAs in inter home link are identified by using messages are reduced, the complexity of the system is
Local HA addresses. reduced and the performance is improved. The failure
The data destined to the MN will be addressed to detection and recovery algorithm is illustrated in
the Global HA address of the MN, which will be procedure 2.
decapsulated by the Active HA and forwarded to the

28
© 2010 ACEEE
DOI: 01.ijsip.01.02.05
ACEEE International Journal on Signal and Image Processing Vol 1, No. 2, July 2010

The workload of each HA in the VPN is IV. PERFORMANCE ANALYSIS
calculated based on the number of mobility binding
The proposed protocol will introduce certain amount
associated with each HA. This workload is used for
of overhead in the network to construct the Virtual
setting priority for the HAs. The priority is
Network and to provide reliability. Hence, the
dynamically updated based on the changes in the
performance of the proposed approach depends on two
number of mobility bindings.
overheads: (1) Time and (2) Control message overhead.
Similar to [15], heartbeat messages are exchanged
In the proposed approach, these two overheads depend on
among the HAs at a constant rate. These heartbeat
the following four factors: (a) VHAHA configuration (b)
messages are used for detecting the failure. When
Home Registration (c) failure detection and recovery and
any HA fails, it will not broadcast the heartbeat
(d) Over-the-air communication between MNs and
message and all other HAs will not receive the
Mobility Agents.
heartbeats from the faulty one. Hence, the failure of
a. VHAHA configuration: The VHAHA is configured
the faulty HA can be detected by all other HAs that
only during the initialization of the network and it will be
are part of the VPN.
updated only when the inactive HA fails. This happens to
Once the failure is detected, the entry of that
be a rare case, since most of the implementations will not
faultyHA from the database will be deleted from all
take any action if the Inactive HA fails and let the
other HAs that are part of Global HA subnet. Then, if
Inactive HA to heal automatically because it will not
the faultyHA is Active HA, based on the priority of
affect the overall performance. Hence, this can be
backup HAs anyone of the backup HA with the
considered as one time cost and it is negligible. The Time
highest priority will be mapped to Active HA by
complexity and message complexity introduced to the
activating its exclusive services. Now, the new
over all systems are negligible.
Active HA will be the owner of the Global HA
b. Home Registration: This factor depends on the
address. If the faultyHA is a backup HA then anyone
total numbers and locations of Active, Backup and
of the Inactive HA will be set as the corresponding
Inactive HAs that are part of VHAHA network. The
backup HA by activating the required services and
registration messages include the number of messages
acquiring binding cache entries from the Primary
required for the MN to get registered with the Active HA
HA. If the Inactive HA is failed then nothing needs
and the control messages required by the Active HA to
to be done. But if it permanently goes off, then any
update this information in all other backup and Inactive
other HA from the link will be set as Inactive HA.
HAs of the MN.
__________________________________________
In the proposed approach, the Initial registration of
_Begin
the MN should take place with the Global HA address
Calculate priority for HAs that are part of Virtual Private instead of with a single HA. Hence, this delay will be
Network
high when compared to the normal Mobile IP
workload(HAi)  (Current mobility bindings of HAi x Current
Throughput) / (Maximum no. of mobility bindings of registration. The initial registration delay includes the
HAi x Maximum Throughput) time taken by the MN to get registered with the Active
Priority(HAi)  1/workload(HAi) HA and the time taken by the Active HA to update this
If(HAs failed to receive heartbeats from HAx) information in all other backup HAs. The Time
HAx  FaultyHA Complexity is O (D log3k) and Message Complexity is O
(|1| + klog3k). Where ‘D’ is the diameter of VHAHA and
If(HAx == Faulty) Then
Delete entries of HAx from the tables of all HAi, where ‘k’ is number of active, backup and Inactive HAs of the
1≤i≤n, i≠x MN.
If(HAx == Active HA) Then c. Failure detection and Recovery overhead: The failure
Activate exclusive services of Backup HA
is detected when heartbeats are not received from a
Active HA  Backup HA with highest priority
Backup HA  Select_Backup_HA (Inactive HA with
particular HA for a particular period of time (T). The
highest priority), heartbeat is actually multicasted using the multicast
activate the required services and acquire the binding address. Number of heartbeats exchanged depends on
details from primary HA to synchronize with it ‘T’ and the time taken to detect the failure depends on
If(HAx == Backup HA) Then
Backup HA  Select_Backup_HA(Inactive HA with
the speed of the underlying wired network. After the
highest priority), failure is detected, it requires just a single message to
activate the required services and acquire the binding switch over to the backup HA and the time taken is
details from primary HA to synchronize with it. negligible. The Time Complexity is O (D log 3n) and the
If(HAx == Inactive HA) Then
Do nothing till it recovers, if it permanently goes off;
Message Complexity is O (|L| + nlog3n). Where ‘D’
select an Inactive HA from the Home Link of HAx is the diameter of VHAHA, ‘n’ is number of HAs that
End are part of VHAHA and ‘L’ represents the number of
_______________________________________________ links that constitute VHAHA.
_ d. Over-the-air messages: This is very important
Procedure 2: Failure detection and Recovery factor because it is dealing with the air interface which is

29
© 2010 ACEEE
DOI: 01.ijsip.01.02.05
ACEEE International Journal on Signal and Image Processing Vol 1, No. 2, July 2010

having less bandwidth. When OTA messages are [12], and TCP [13]. The comparison results are given in
increased performance of the system will be Table 1. From the comparisons, it can be observed that
degraded. But in the proposed approach, the MN is VHAHA is persistent and has less overhead.
not involved in failure detection and recovery Simulation experiments are performed to verify the
process, so no OTA messages exchanged during this performance of the proposed protocol. It is done by
process. The time and message complexity extending the Mobile IP model given in ns-2[17]. In the
introduced by this factor is Nil. extended simulation model, there are 20 HAs, 10,000
From the above description, it is observed that the MNs and 5 Home links. Each home link consists of 4
performance of VHAHA is directly proportional to HAs. The Home link that constitutes the Active HA is
the speed of the wired network because the proposal configured as Primary Home link. All other Home links
only involves the wired backbone operations. constitute a backup HA and one or more Inactive HAs.
Actually, this is not a fair constraint because The VPN is constructed among the Home links. The
bandwidth of the network is very high thanks to the traffic is generated randomly and assigned to MNs. The
high speed and advanced networks. failures are randomly set to Active HA, Backup HA and
Inactive HA failures. The simulation time is set to 250
A. Simulation results and Comparisons
seconds.
The proposed approach is compared with Simple
Mobile IPv6 [1], HAHA [9], HARP [11], VHARP

TABLE 1. COMPARISON BETWEEN VHAHA AND OTHER APPROACHES
Metrics MIPv6 HAHA HARP VHARP TCP VHAHA
Recovery overhead High No No No No No
Fault tolerance
No MN initiated MN initiated HA initiated HA initiated HA initiated
mechanism
Fault tolerant Covers entire Limited to Limited to Limited to Covers entire
No
Range range Home Link Home Link Home Link range
Transparency No No No Yes Yes Yes
OTA messages
exchanged for More More Less Nil Nil Nil
recovery

The following parameters are used to evaluate the 120
Failure detection & Recovery time (sec).......

110
performance. (1) Failure detection and Recovery 100

time when a HA fails in the Home Link (2) Failure
90
80
MIPv6

detection and Recovery time when entire Home Link 70
60
T CP
VHARP

fails (3) Registration delay (4) Packet loss (5) No. of 50
40
VHAHA

messages exchanged during registration and (6) 30
20

Failure detection and Recovery messages. 10
0
00

00
00

00
0

1. Failure detection and Recovery time when a HA
10

40
30

No.of MNs

fails in the Home Link(TFD_R)
This factor is calculated by using (1). The Fig. 4 Figure 4. Comparison of Failure detection and recovery time,
when a HA fails in the Home Link
shows the TFD_R of VHAHA and other protocols.
Base Mobile IPv6 does not take any action for failure 2. Failure detection and Recovery time when entire
detection and Recovery of HAs. This needs to be Home Link fails
handled by MN itself. Because of that, the time taken
Failure detection & Recovery time (sec).........

120

for failure detection and Recovery is very high. This 110
100

causes service interruption to MNs that are affected 90
80
by the faulty HA. Other schemes like TCP, VHARP 70
MIPv6
TCP
60
and VHAHA handle the problem and almost take 50
VHARP
VHAHA

same amount of time for failure detection and
40
30

Recovery. 20
10
0
00

00
0

T FD =3T H  propagationdelayOfVHAHA No.of MNs

(1) Figure 5. Comparison of Failure detection and recovery time,
when the entire Home Link fails

30
© 2010 ACEEE
DOI: 01.ijsip.01.02.05
ACEEE International Journal on Signal and Image Processing Vol 1, No. 2, July 2010

This situation is represented in Fig. 5, where This is illustrated in the Fig. 8. From the Figure, it is
VHAHA’s Recovery time is almost equal to 1. But, found that the number of messages exchanged in
TCP and VHARP approaches fail to handle the VHAHA is somewhat high when compared to base
situation and Recovery time is very high which is protocol but it is comparable with the VHARP protocol.
equal to that of base MIPv6.
80000
3. Registration delay
70000

No. of msgs exchanged during Home Registration...........
130 60000
120
110 50000
MIPv6
100
Registration time (sec)...

TCP
90 40000
VHARP
80 MIPv6
VHAHA
70 T CP 30000
60 VHARP
50 VHAHA 20000
40
30 10000
20
10 0
0 250s--->
10000
1000

2000

3000

4000

5000
6000

7000

8000

9000
0

Simulation time

No.of MNs

Figure 8. Comparison of no. of messages exchanged
Figure 6. Comparison of Registration delay during Home Registration

This is calculated by using (2) and the values are 6. Failure detection and Recovery messages
given in Fig. 6. This factor depends on the total This is represented in Fig. 9. Here, also the
number and locations of HAs that are part of complexity of the VHAHA is approximately equal to
VHAHA network. This includes the time taken by that of TCP and VHARP.
the MN to get registered with the Primary HA and
the time taken by the Primary HA to update this 250000

information in all other backup and Inactive HAs.
No. of Failure detection & Recovery msgs exchanged...........

200000

regdelay=regdelay  propagationdelayOfVHAHA
Active− HA
(2) 150000 MIPv6
T CP
VHARP

4. Packet loss 100000 VHAHA

The packet losses of the compared protocols are
50000
represented in Fig. 7. From the Figure, it is inferred
that packet loss in the proposed approach is very less 0

when compared with MIPv6, TCP and VHARP, 250s--->
Simulation time
because it is able handle both intra link and interlink
failures. Figure 9. Comparison of no. of failure detection
and Recovery messages
10

From the results and analysis, it is observed that the
VHAHA outperforms all other reliability mechanisms
Packet Loss(pkts/sec)......

because it survives even when the entire Home link fails.
1
MIPv6
TCP
The overhead and complexity introduced by the
1000 5000 10000 VHARP
VHAHA
proposed approach is almost negligible when compared
to other existing recovery mechanisms.

0.1
V. CONCLUSION
No. of MNs
In this paper, a fault-tolerant mechanism based on
inter home link HA redundancy scheme is introduced.
Figure 7. Comparison of Packet Loss This approach does not require any changes in the
5. No. of messages exchanged during registration current Mobile IP architecture. The performance analysis
and the comparison results show that the proposed
This includes number of messages required to
approach has less overhead and the advantages like,
register with the Active HA and Binding Update
better survivability, transparent failure detection and
messages to the backup HA during the Initial
recovery, reduced complexity of the system and
Registration, FA Registration and deregistration.
workload, secure data transfer and improved overall

31
© 2010 ACEEE
DOI: 01.ijsip.01.02.05
ACEEE International Journal on Signal and Image Processing Vol 1, No. 2, July 2010

performance. The formal load balancing of workload based Networks” in Proc. International conference on Wireless
among the HAs of the VHAHA is left as future Networks, Communications, and Mobile Computing,
work. WirelessCom 2005, Hawaii, USA, June 13-16, 2005.
[13] Adisak Busaranun1, Panita Pongpaibool and Pichaya
Supanakoon, “Simple Implement of Home Agent Reliability
REFERENCES for Mobile IPv6 Network”, Tencon, November 2006.
[1] C. Perkins, D. Johnson, and J. Arkko, “Mobility [14] J. Faizan, H. El-Rewini, and M. Khalil, “Towards Reliable
Support in IPv6,” IETF Draft, draft-ietf-mobileip-ipv6-24 Mobile IPv6” Southern Methodist University, Technical Report
August 2003. (04-CSE-02), November 2004.
[2] C. Perkin RFC 3344: “IP Mobility Support for IPv4”, [15] J. Faizan, H. El-Rewini, and M. Khalil, “Problem
august 2002. Statement: Home Agent Reliability,” IETF Draft, draftjfaizan-
[3] B. Chambless, and J. Binkley, “Home Agent mipv6-ha-reliability-00.txt, November 2003.
Redundancy Protocol,” IETF Draft, draft-chambless- [16] RFC 2764, A Framework for IP Based Virtual Private
mobileip-harp- 00.txt, October 1997. Networks. B. Gleeson, A. Lin, J. Heinanen, G. Armitage, A.
[4] R. Ghosh, and G. Varghese, “Fault Tolerant Mobile Malis. February 2000. ftp://ftp.isi.edu/in-notes/rfc2764.txt.
IP,” Washington University, Technical Report (WUCS-98- [17] NS -2 , http://www.isi.edu/nsnam
11),1998.
[5] J. Ahn, and C. S. Hwang, “Efficient Fault-Tolerant
Protocol for Mobility Agents in Mobile IP,” in Proc. 15 th Rathi S is a senior Lecturer in Department of
Int. Parallel and Distributed Processing Symp., California, Computer Science and Engg, Government
College of Technology, Coimbatore,
2001.
Tamilnadu, India. Her fields of interests are
[6] K. Leung, and M. Subbarao, “Home Agent Computer Networks, Mobile Computing,
Redundancy in Mobile IP,” IETF Draft, draft-subbarao- Wireless Security and Fault Tolerant system
mobileipredundancy-00.txt, June 2001. Design. She received her Master degree, in
[7] M. Khalil, “Virtual Distributed Home Agent Computer Science and Engineering from
Protocol(VDHAP),” U.S.Patent 6 430 698, August 6, Government College of Technology,
2002. Coimbatore.
[8] J. Lin, and J. Arul, “An Efficient Fault-Tolerant She is a member of IEEE and ISTE. She also leads and teaches
modules at both B.E. and M.E. levels in Computer Science. She has
Approach for Mobile IP in Wireless Systems,” IEEE
published 15 technical papers in national and international conferences
Trans. Mobile Computing, vol. 2, no. 3, pp. 207-220, July- and journals.
Sept. 2003.
[9] R. Wakikawa, V. Devarapalli, and P.Thubert, “Inter
Home Agents Protocol (HAHA),” IETF Draft, draft- Thanushkodi K is working as Principal in
wakikawamip6- nemo-haha-00.txt (work in progress), Akshaya College of Engineering and
October 2003. Technology, Coimbatore, Tamilnadu, India.
[10] Deng, H. Zhang, R. Huang, X. and K. Zhang, “Load His research interests are in the area of
Computer Modeling and simulation,
balance for Distributed HAs in Mobile IPv6”, IETF Dreaft,
Computer Networking, Network security and
draft-wakikawa-mip6-nemo-haha-00.txt, October 2003. Power Systems and Design. He received the
[11] F. Heissenhuber, W. Fritsche, and A. Riedl, “Home B.E. in Electrical and Electronics
Agent Redundancy and Load Balancing in Mobile IPv6,” Engineering, M.Sc. (Engg) from Madras
in Proc. 5th International Conf. Broadband University, Tamilnadu.
Communications, Hong Kong, 1999. He did his PhD in Electrical and Electronics Engineering from
[12] J. Faizan, H. El-Rewini, and M. Khalil, “VHARP: Bharathiar University, Coimbatore. He has published 26 technical
Virtual Home Agent Reliability Protocol for Mobile IPv6 papers in National and International Journals.