Mechanism Activation

Below, Figure 4.8 describes the summary of relationships among problems, activation strategies, patterns of different roles, and even the mobile host sets.

For MCM Problem, whenever the mobile host of FCM set makes an Inter-domain Mobility and registers at the home registrar, the prediction results in terms of sequential calling patterns are sent from the home registrar to the current domain SIP server along with the authentication information. Therefore, once the user makes another Inter-domain Mobility, the calling patterns will assist the previous domain in intercepting and forwarding media packets to the predicted domain(s) in case of packet loss before delayed re-Invite reaches the corresponding host. With these calling portfolios, the reduction amount of packet loss increases when the mobile host becomes more far away from the corresponding host becomes in distance.

For HRFR Problem, as soon as the home registrar gets crashed, the SIP server will query the predicted domains of those mobile users of FM set, recovered from backup, in terms of sequential moving patterns incrementally. The mission is to restore current & complete location information in case of obsolete location information to any upcoming call request before the mobile user makes another Inter-domain Mobility and location update.

For PCM Problem, whenever the mobile host of FM set makes an Inter-domain Mobility and registers at the home registrar, the prediction results in terms of sequential moving patterns are sent from the home registrar to the current domain SIP server along with the authentication information. Therefore, once the user makes another Inter-domain Mobility, the moving patterns will assist the previous domain in forwarding call requests to the predicted domain(s) in case of

call failure before delayed re-Register arrives the home registrar. With these moving portfolios, the decrement of call failures increases while the mobile host becomes more distant from the corresponding host.

Figure 4.8 : Problem-Oriented Activation Strategies

Chapter 5.

Evaluation

In Chapter 4, we have presented our method, STAMP, which solves well-known SIP mobility problems (Mid-Call Mobility & Pre-Call Mobility) and notable mobility database failure problem (Home Registrar Failure Restoration). In STAMP, the preprocessing phase is efficient because it dynamically selects high risk users to do further STAMP algorithm; the rest phases after preprocessing provide adaptive solutions for each problem.

For MCM & PCM Problem, STAMP mobility scheme is mainly evaluated by the mathematical analysis model introduced from Shadow Registration in [7]. On the other hand, for HRFR Problem, our scheme certainly performs a better quality of service compared with the original condition without any preventives.

Before going into more details, there are some assumptions on WLAN deployment for STAMP. The reason is: for MCM & PCM Problem, the SIP Server and In-bound Router should know when the user makes Inter-domain Mobility so as to activate STAMP and start forwarding packets or requests! The WLAN assumptions and mobility cases are basically depicted in Figure 5.1.

Assumptions

z Only each access point (i.e., AP) under border subnet such as “b-Subnet 3” or “b-Subnet 1” in Domain A, the AP will inform the In-bound Router and SIP Server if the user leaves.

z All APs have to inform the domain In-bound Router and SIP Server if any user moves into the coverage of this AP.

Two Cases

z Upper case (Inter-domain Mobility):

Absence information is sent to the In-bound Router and SIP Server under Domain A from AP 3-2, but no AP sends any presence information of this user within a tolerable interval. Thus, STAMP is activated!

z Lower case (Intra-domain Mobility):

Absence information is sent to the In-bound Router and SIP Server under Domain A from AP 3-2, and AP 3-1 sends presence information subsequently. Thus, STAMP is not activated!

Figure 5.1 : Inter-domain Mobility: STAMP activation time

Based on the above mentioned, the evaluation for each problem is as the following.

5.1 Evaluation of Problem1. Mid-Call Mobility

For MCM Problem, we apply STAMP to Original SIP and SIP with Shadow Registration by two schemes (i.e., SIP with STAMP as well as SIP with STAMP &

Shadow Registration, accordingly), as depicted in Figure 5.2. Firstly, SIP with

STAMP outperforms Original SIP by 2tmc. The RTP translator is activated after being informed by any access point so as to start intercepting and forwarding media packets according to the prediction result from STAMP. Then the user starts receiving media packets while the authentication retrieval reaches the new domain.

Figure 5.3 and Figure 5.4 display more detail operations. Secondly, SIP with STAMP & Shadow Registration also outperforms SIP with Shadow Registration by 2tmc. The operation is almost the same as SIP with STAMP. The only difference lies in the fact that authentication is retrieved at local (new) domain because of Shadow Registration. Figure 5.5 and Figure 5.6 illustrate the operations in detail.

Moreover, Figure 5.7 displays the mathematical model in Shadow Registration and compares our contribution with Shadow Registration.

Figure 5.2 : MCM Problem Evaluation – Disruption Time Comparison

Figure 5.3 : Operation Sequence: Original SIP

Figure 5.4 : Operation Sequence: SIP with STAMP

Figure 5.5 : Operation Sequence : SIP with Shadow Registration

Figure 5.6 : Operation Sequence: SIP with STAMP & Shadow Registration

Figure 5.7 : Contribution Comparison of STAMP vs. Shadow Registration

5.2

5.3

Evaluation of Problem2. Home Registrar Failure Restoration

For original HRFR Problem, some locations are updated until the user makes any Inter-domain mobility; STAMP restores the location information actively as soon as possible instead. Comparatively speaking, STAMP shortens the interval and probability of call failures, evidently improving the service availability.

Evaluation of Problem3. Pre-Call Mobility

In regard to PCM Problem, STAMP reduces the interval of service unavailability by 2(th－tf). The procedure is as the following: (1) the call requests are intercepted at the old domain and then forwarded to the predicted domain according to STAMP once the In-bound Router knows the absence of the mobile user; (2) the call request starts being received by the mobile user once the authentication is retrieved, according to the sequence operation diagrams depicted in Figure 5.8 and Figure 5.9.

Figure 5.8 : Operation Sequence: Original SIP registration

Figure 5.9 : Operation Sequence: SIP registration under STAMP

Chapter 6.

Conclusion

As the mobile communication technologies become widespread and prevalent, a vast majority of portable devices providing a series of comprehensive real time applications grow rapidly. As a result, seamless terminal mobility plays a significant role in the quality of most mobile services. In this thesis, we focus on the SIP terminal mobility issues via Mid-Call Mobility, Home Registrar Failure Restoration, and Pre-Call Mobility Problems. We devise a mobility approach, STAMP, to solve all of the three problems in an adaptive manner. Considering both AAA functionality and RTP Translator, firstly we analyze and compare the Inter-domain handoff disruption time of Shadow Registration versus STAMP for MCM Problem.

The fact that STAMP shows less disruption time than Shadow Registration proves that STAMP is not only fast but smooth handoff scheme itself. Moreover, STAMP could even enhance Shadow Registration more efficiently with the knowledge of moving patterns. Then, for HRFR and PCM Problem, STAMP reveals shorter interval of service unavailability; especially for PCM Problem, the contribution of STAMP increases when the mobile user happens to be far away from the home network. In brief, STAMP contributes to service continuity as well as service availability.

Future Work

We have already evaluated STAMP on the MCM, HRFR, and PCM Problems in a mathematical manner, based on the analysis model presented in [7]. Moreover, experiments are under progress. The simulation results are expected to prove the contribution of STAMP in a statistical way.

Reference

[1] N. Akhtar, M. Georgiades, C. Politis, and R. Tafazolli, “SIP-based End System Mobility Solution for All-IP Infrastructures.” IST Mobile & Wireless Communications Summit, 2003

[2] N. Banerjee, S. K. Das, and A. Acharya, “SIP-based Mobility Architecture for Next Generation Wireless Networks.” Proceedings of the 3rd IEEE Int’l Conf. on Pervasive Computing and Communications, PerCom 2005

[3] A. Dutta, S. Madhani, W. Chen, and H. Schulzrinne, “Fast-handoff Schemes for Application Layer Mobility Management.” IEEE PIMRC, 2004

[4] H. Fathi, R. Prasad, and S. Chakraborty, “Mobility Management for VoIP in 3G System:

Evaluation of Low-Latency Handoff Schemes.” IEEE Wireless Communications, 2005 [5] J.-M. Gil, C. Y. Park, C.-S. Hwang, D.-S. Park, J. G. Shon, and Y.-S. Jeong, “Restoration

Scheme of Mobility Database by Mobility Learning and Prediction in PCS Networks”, IEEE Selected Areas in Communications, 2001

[6] A. B. Johnston, “SIP－Understanding Session Initiation Protocol.” 2nd Edition, Artech House, 2004

[7] T. t. Kwon, M. Gerla, and S. Das, “Mobility Management For VoIP Service: Mobile IP vs.

SIP.” IEEE Wireless Communications, 2002

[8] S. Ma, S. Tang, D. Yang, T. Wang, and J. Han, “Combining Clustering with Moving Sequential Patterns Mining：A Novel and Efficient Technique”, National High Technology Development 863 Program of China, National Grand Fundamental Research 973 Program of China, Foundation of the Innovation Research Institute of PKU－IBM [9] W.-C. Peng and M.-S. Chen, “Developing Data Allocation Schemes by Incremental

Mining of User Moving Patterns in a Mobile Computing System.” IEEE Tran. Knowledge

and Data Eng., 2003

[10] W.-C. Peng and M.-S. Chen, “Shared Data Allocation in a Mobile Computing System:

Exploring Local and Global Optimization.” IEEE Tran. Parallel and Distributed Systems, VOL. 16, NO. 4, APRIL 2005

[11] H. Schulzrinne and E. Wedlund, “Application-Layer Mobility Using SIP.” Mobile Computing and Communications Review, Volume 1, Number 2

[12] H. Schulzrinne and E. Wedlund, “Application-Layer Mobility Using SIP.” Mobile Computing and Communications Review, Volume 4, Number 3

[13] H. Sinnreich and A. B. Johnston, “Internet Communications Using SIP－Delivering VoIP and Multimedia Services with Session Initiation Protocol.” Wiley Computer Publishing, 2001

[14] S.M. Tseng and W.C. Lin, “Mining Sequential Mobile Access Patterns Efficiently in Mobile Web Systems”, Advanced Information Engineering and Applications 2005

[15] E. Wedlund and H. Schulzrinne, “Mobility Support Using SIP.” ACM/IEEE WoWMoM’99

[16] G. Yavas, D. Katsaros, O. Ulusoy, and Y. Manolopoulos, “A data mining approach for location prediction in mobile environments.” Data & Knowledge Engineering ELSEVIER 2005