International Advisory Committee• Byeong Gi Lee, Seoul National Univ., Korea• Nim Cheung, ASTRI, China• Chul Hee Kang, RAPA, Korea • Zygmunt J. Haas, University of Texas at Dallas, USA• Kyung Sup Kwak, Inha Univ., Korea• Ramjee Prasad, Aalborg Univ., Denmark• Chuwhan Yim, Korea Univ., Korea• Wu Hequan, Chinese Academy of Eng., China• Bijan Jabbari, George Mason Univ., USA• Iwao Sasase, Keio Univ., Japan• Jinwoo Park, Korea Univ., Korea• Douglass Zuckerman, IEEE ComSoC• Jaiyong Lee, Yonsei Univ., Korea• Naohisa Ohta, Keio Univ., Japan• Pascal LORENZ, Univ. of Haute Alsace, France• Zhisheng Niu, Tsinghua Univ., China• Dong Ho Cho, KAIST, Korea• Ilyoung Chong, HUFS, Korea • Zhen Yang, NUPT, China

Steering Committee• Masahiro Umehira, Ibaraki University, Japan• Yeong Min Jang, Kookmin Univ., Korea• Sungchang Lee, Korea Aerospace Univ., Korea• Jiandong Li, Xidian Univ., China• C. K. Toh, National Tsing Hua Univ., Taiwan• Seong-Ho Jeong, HUFS, Korea• Zary Segall, KTH, Sweden• Seung Hyong Rhee, Kwangwoon Univ., Korea• Xin Wang, Fudan Univ., China• Sang-Jo Yoo, Inha Univ., Korea• Honggang Zhang, Zhejiang Univ., China• Nguyen Huu Thanh, HUST, Vietnam• Joel Rodrigues, IT, Univ. of Beira Interior, Portugal• Myungsik Yoo, Soongsil Univ., Korea• Ki-Hyung Kim, Ajou Univ., Korea• Sunghyun Choi, Seoul National Univ., Korea• Juan Carlos Cano, Technical Univ. of Valencia, Spain• Dong Seog Han, Kyungpook National University, Korea

Organizing CommitteeHonorary Conference Co-chairs• Ramjee Prasad, Aalborg Univ., Denmark• Sang Hong Lee, IITP, KoreaOrganizing Co-Chairs• Yeong Min Jang, Kookmin Univ., Korea• C. K. Toh, National Tsing Hua Univ., Taiwan• Masahiro Umehira, Ibaraki University, JapanWorkshop Co-Chairs• Sungrae Cho, Chung-Ang Univ., Korea• Kyomin Jung, Seoul National Univ., KoreaSpecial Session Chairs• Kyung-Joon Park, DGIST, Korea • Wan-Sup Cho, Chungbuk National Univ., KoreaPublication Co-Chairs• Young-June Choi, Ajou Univ., Korea • Hwangnam Kim, Korea Univ., Korea Publicity Co-Chairs• Joon Yoo, Gachon Univ., Korea• Young-Ho Jung, Korea Aerospace Univ., Korea • Xuejun Sha, Harbin Institute of Tech., China• Timo Sukuvaara, FMI, Finland• Jyh-Cheng Chen, National Chiao Tung Univ., Taiwan• Carlos T. Calafate, Technical Univ. of Valencia, Spain International Liaison Chairs• Jangwon Lee, Yonsei Univ., Korea• Jaewoo So, Sogang Univ., Korea • Jiman Hong, Soongsil Univ., Korea International Journal Co-Chairs• Myungsik Yoo, Soongsil Univ., Korea• Joon Sang Park, Hongik Univ., KoreaRegistration Co-Chairs • Hyunggon Park, Ewha Womans Univ., Korea• Hongseok Kim, Sogang Univ., Korea• Sunwoong Choi, Kookmin Univ., Korea Patronage Co-Chairs • Seong-Choon Lee, KT, Korea• Myung Hyun Yoon, KETI, Korea• Won Lyu, ETRI, KoreaFinance Co-Chair• Sangjoon Park, ETRI, KoreaLocal Arrangement Co-Chairs • Takeo Ohgane, Hokkaido Univ., Japan• Toshihiko Nishimura, Hokkaido Univ., Japan• Junbeom Hur, Chung-Ang Univ., Korea• Jaesang Cha, SNUT, KoreaWeb Co-Chair• Yonghoon Choi, Kwangwoon Univ., KoreaCoordinators • Dongkyun Kim, Kyungpook National Univ., Korea • Sangheon Pack, Korea Univ., Korea

Technical Program CommitteeTPC Co-Chairs• Eui-Nam Hur, Kyunghee Univ., Korea• Xin WANG, Fudan Univ., China• Takeo Fujii, Univ. of Electro-Comms, Japan• Masaki Aida, Tokyo Metropolitan Univ., Japan• Kun Yang, Univ. of Essex, UK• Edmund Yeh, Northeastern Univ., USA TPC Vice Co-Chairs• Francisco Martinez, Univ. of Zaragoza, Spain• Sang-Chul Kim, Kookmin Univ., Korea • Macos Katz, Univ. of Oulu, Finland• Panos Papadimitratos, KTH, Sweden• Sangwhan Lee, Kookmin Univ., Korea

Submission Guidelines: All papers must be submitted electronically, in PDF format, and uploaded on EDAS. The direct link for paper submission is http://edas.info/N18535. There are two options for the submissions, either a full paper or a short paper. We decided to allow the submission of short papers in order to accommodate the most recent research outcomes. The submissions should be formatted with single-spaced, double-column pages using at least 10 pt size fonts on A4 or letter pages in IEEE style format. The maximum number of pages is 6 for full papers and 3 for short papers. Detailed formatting and submission instructions will be available on the conference web site (http://www.icufn.org).Call for Workshop ProposalsProposals are invited for half- or full-day workshops in communication and networking topics. Please contactworkshop co-chairs, Sungrae Cho and Kyomin Jung at srcho@uclab.re.kr and kjung@snu.ac.kr.Selected Journal PublicationAccepted and presented papers will be published in the ICUFN 2015 Conference Proceedings and submitted to IEEE Xplore. The ICUFN 2015 Conference Proceedings will also be submitted for indexing through EI Compendex, ScienceDirect, Scopus, Web of Knowledge, Thomson ISI, and IET Inspect. Also extended versions of selected papers will be published in the following international journals as a special section for the ICUFN 2015. - Wireless Personal Communications (SCIE) - International Journal of Distributed Sensor Networks (SCIE) - Journal on Internet Technology (SCIE) - INFORMATION Journal (Scopus)- International Journal of E-Health and Medical Communications (Scopus)- ICT Express- Recent Advances on Communications and Networking TechnologyBest Paper AwardsBest papers will be selected by the technical program committee members among regular papers presented at ICUFN 2015.Important Dates- Submission Deadline (6-page Full Paper or 3-page Short Paper): March. 15, 2015- Acceptance Notification: April 15, 2015- Camera-Ready Paper (6-page Full Paper or 3-page Short Paper): May 1, 2015

The topics include, but are not limited to:- Wireless and Communication Network- Cognitive Radio Network- 5G, 4G, LTE, LTE-Advanced, WLAN, WPAN- Wireless Ad-hoc and Mesh Network- Small cell Networks- Heterogeneous Networks- Self-Organizing Networks- Interference Management- Radio Resource Management- QoE, QoS, SLA, and GoS- Future Internet and Network- SDN and Network Virtualization- Web-services and SOA- Network Security and Network Management- Network Platform, Software and Services- Network Theory and Network Science- Information Networks- Social Network- Cloud Computing and Networks- CCN/CDN/ICN/Delay-tolerant networks

- Big Data Networks and Data Center Architecture- Storage Networking Protocols- Body Area Network- Cooperative and Collaborative Communications- Cross-Layer Design and Optimization- Wireless Sensor Networks, RFID and QR Code- Cyber-physical system (CPS)- Ubiquitous Computing and Sensor Networks- Optical Wireless, VLC and LED Communication- Underwater Sensor Networks- Internet of things (IoTs)- D2D and M2M- Vehicular Ad-hoc Networks and Connected Car- Energy Internet, Smart Grid and Green Internets- Location-based Services- Location and mobility management- e-Health- Multi Screen and Digital Signage- Smartphone Applications

CALL FOR PAPERSDuring the last decade, we have witnessed fast developments of various networking technologies, and many forms of networking are becoming core parts of our daily lives. With the proliferation of future wireless technologies and electronic devices, there is a fast growing interest in ubiquitous and future networks. In the days to come, we expect that the ubiquitous communication and networking technologies will become ubiquitous along with the emergence of many future networking technologies. The ubiquitous and future network will offer multiservice, multimedia services convergence, mobility, service ubiquity and context awareness, fixed-mobile convergence, quality of service, variable connectivity, spontaneous networking, autonomic networking and other capabilities as the norm. Building on the success of the last six years, the seventh International Conference on Ubiquitous and Future Networks (ICUFN 2015) aims at addressing advances in research on ubiquitous and future networks, covering topics ranging from technology issues to emerging applications and test-bed developments. ICUFN 2015 solicits original, unpublished contributions in all aspects of ubiquitous and future networking. Submitted articles must not be concurrently considered elsewhere for publication. The conference is organized by KICS (The Korean Institute of Communications and Information Sciences) with the technical co-sponsorship of IEEE Communications Society and IEICE-CS. Accepted and presented papers will be published in the ICUFN 2015 Conference Proceedings and submitted to IEEE Xplore. The ICUFN 2015 Conference Proceedings will also be submitted for indexing through EI Compendex, ScienceDirect, Scopus, Web of Knowledge, Thomson ISI, and IET Inspect.

http://www.icufn.orgJuly 7(Tue.) ~ July 10(Fri.), 2015, Sapporo, Japan

The Seventh International Conference on Ubiquitous and Future NetworksICUFN 2015

TopS

Highlight

1

Mobility Support for Proxy Mobile IPv6 in

TRILL-based Mobile Networks

Moneeb Gohar Department of Information and Communication Engineering,

Yeungnam University moneebgohar@yu.ac.kr

Sang-Il Choi School of Computer Science and

Engineering, Kyungpook National University

overcycos@gmail.com

Seok-Joo Koh School of Computer Science and

Engineering, Kyungpook National University

sjkoh@knu.ac.kr

Abstract—In mobile networks, the Proxy Mobile IPv6 (PMIP)

scheme tends to induce large tunneling overhead and non-

optimal routes for data packet delivery. To overcome such

limitations, we propose the use of PMIP for mobility control in

mobile networks based on the Transparent Interconnection of

Lots of Links (TRILL). In the proposed scheme, each edge

Routing Bridge (RB) in the TRILL network will function as

Mobile Access Gateway (MAG) of PMIP, and an enhanced Local

Mobility Anchor (eLMA) is used at the gateway of TRILL mobile

network. With such eLMA, the control and data plane are

separated, and the data path is optimized in PMIP. Moreover,

each data packet in PMIP is encapsulated with the TRILL

header, instead of using the IP-in-IP header. By numerical

analysis, we show that the proposed scheme can give better

performance than the existing schemes in terms of traffic

processing overhead and total transmission delay.

Keywords—PMIPv6; Mobile networks; TRILL; Mobility

support; Analysis; Routing Bridges.

I. INTRODUCTION

As the Internet services become popular, the number of mobile Internet users has been rapidly increasing with wide popularity of smart phones. It is reported that the number of mobile Internet users will exceed the number of desktop users in near future [1, 2, 3].

The Transparent Interconnection of Lots of Links (TRILL) protocol has recently been discussed by IETF [4, 5]. In TRILL, an Ingress Routing Bridge (IRB) prepends a TRILL header to the data packet of a source host, and then an Egress Routing Bridge (ERB) strips the TRILL header prior to final delivery to the destination host.

The TRILL can be used as a key technology for mobile networks. In [6], it is defined how TRILL can be run over IP. In the meantime, we note that the Proxy Mobile IPv6 (PMIP) [7] can also be used for IP mobility support in the TRILL-based mobile networks.

In this paper, we discuss how to use the PMIPv6 protocol for mobility control in the TRILL-based mobile networks. To use PMIP in TRILL networks, each edge Routing Bridge (RB)

will function as the Mobile Access Gateway (MAG) of PMIP, and an enhanced Local Mobility Anchor (eLMA) is used at the gateway of TRILL mobile network. With such eLMA, the control and data plane are separated, and the data path is optimized in PMIP. Moreover, each data packet in PMIP is encapsulated with the TRILL header, instead of using the IP-in-IP header.

The rest of this paper is organized as follows. In Section II, we describe the existing PMIP schemes. In Section III, we propose how to extend the PMIP in TRILL-based networks. Section IV compares the proposed and existing schemes by performance analysis. Section V concludes this paper.

II. EXISITNG PMIP SCHEMES

In PMIP [7], the binding of Home of Address (HoA) and Care of Address (CoA) is maintained at a central mobility anchor such as LMA and process all of the control and data packets. When a mobile node (MN) is connected to a new MAG, the MAG of MN sends a Proxy Binding Update (PBU) message to LMA so as to bind the HoA and CoA of MN. Then, LMA responds with a Proxy Binding ACK (PBA) message to MAG of MN. Now, a corresponding node (CN) can send the data packets to MN via LMA.

PMIP-LR [8] is an extensional scheme of PMIP for localized routing, in which the PBU operation is performed, as done in PMIP. However, the data path between MN and CN is optimized after the PMIP-LR control operation, in which Localized Routing Initiation (LRI) and Localized Routing ACK (LRA) messages are used between MAGs and LMA.

III. EXTENSION OF PMIP OVER TRILL NETWORK

In this section, we discuss how to use and extend the PMIP in the TRILL-based mobile networks, named as PMIP-TRILL.

A. Packet Encapsulation

In the proposed scheme, each edge RB in TRILL network will function as MAG. The LMA is enhanced, named eLMA, which is employed at the gateway of the TRILL domain. The eLMA will process only the PMIP control messages, and the PMIP control operation is separated from the data delivery

2

function, which will be described in the subsequent section. In PMIP-TRILL, it is noted that each data packet is encapsulated with a TRILL header, not using the IP-in-IP tunneling scheme.

In PMIP, the data packet is encapsulated with 20 bytes of IP header and 20 bytes of IP header, which is total 40 bytes. However, in the propose PMIP-TRILL scheme, the data packet will be encapsulated with 12-byte TRILL header and 14-byte MAC header [4]. For example, in PMIP and PMIP-LR, let us assume that an IP packet has a user payload of 200 bytes. Then, we can get data tunneling overhead (DTO) as follows.

In the meantime, in the PMIP-TRILL scheme, we can get the data tunneling overhead as follows.

B. Protocol Operation of PMIP-TRILL

For description of the protocol operations, we will consider only the mobile CN case, because the operations for the fixed CN are almost the same with the original PMIP operations. Fig. 1 illustrates the protocol operation of PMIP-TRILL. In the figure, it is shown that each edge RB, Ingress RB (IRB) or Egress EB (ERB), will function as MAG. Transit RBs (TRBs) are used for packet delivery in the network. It is noted that the enhanced LMA (eLMA) is used at the gateway of TRILL network, which may work as an IRB/ERB for the outside CNs. When MN is connected to MAG (Step 1), MN-MAG sends a PBU message to eLMA so as to bind the HoA and CoA of MN. Then, eLMA will create the associated binding entry and respond with a PBA message to MAG of MN (Step 2). Now, CN sends a data packet to MN (Step 3). Then, CN-MAG will send a Proxy Binding Query (PBQ) message to eLMA so as to find the location of MN. Then, eLMA responds with a Proxy Query ACK (PQA) message to CN-MAG, after looking up its database (Step 4). Now, CN-MAG sends the data packets directly to MN-MAG by encapsulation with the TRILL header, between CN-MAG and MN-MAG (Step 5). Finally, MN-MAG will forward the data packets to MN (Step 6).

C. Extension of PMIP messages for TRILL

To use the proposed PMIP-TRILL scheme, the binding query operation should be performed between MAG and eLMA. For this purpose, the following PMIP control messages are newly defined: Proxy Binding Query (PBQ) and Proxy Query ACK (PQA). These two messages can be defined by adding the ‘Q’ flag bit into the existing PBU and PBA messages of PMIP, respectively, as shown in Fig. 2.

Figure 1. Protocol operations of PMIP-TRILL

Figure 2. Newly defined control packets: PBQ and PQA

IV. PERFORMANCE ANALYSIS

For performance analysis, we compare the traffic processing overhead and total transmission delay for the three candidate schemes: PMIP, PMIP-LR and PMIP-TRILL.

A. Analysis Model

For analysis, we use the notations given in Table 1.

TABLE1. PARAMETERS USED FOR ANALYSIS

Parameters Description

Sc Size of control packets (bytes) Sd Size of data packets (bytes)

NHost Number of hosts in the domain

NMAG Number of MAGs in the domain

N Number of data packets Bw Wired link bandwidth (Mbps)

Bwl Wireless bandwidth(Mbps)

Lw Wired link delay (ms)

Lwl Wireless link delay (ms)

Ha-b Hop count between node a and b in the

network q Wireless link failure probability

Tq Average queuing delay at each node

3

We also consider a network model, as illustrated in Fig. 3.

Figure 3. Network model for performance analysis

In the figure, we denote Tx-y(S) by the transmission delay of

a message with size S sent from x to y via the ‘wireless’ link. Then, Tx-y(S) can be expressed as Tx-y(S) = [(1+q)/(1-q)] × [(S/Bwl) +Lwl]. In the meantime, we denote Tx-y(S,Hx-y) by the transmission delay of a message with size S sent from x to y via ‘wired’ link, where Hx-y represents the number of wired hops between node x and node y. Then, Tx-y(S,Hx-y) is expressed as Tx-y(S, Hx-y) = Hx-y× [(S/Bw) + Lw+Tq].

B. Traffic Processing Overhead (TPO) at LMA

To analyze the performance of the candidate schemes, we evaluate the TPO required for binding update, binding query and data delivery at LMA (or eLMA). In PMIP, it is assumed that the hosts are equally distributed in mobile network. For binding update, all hosts in network will exchange PBU and PBA messages with LMA. Thus, the PBU/PBA messages of 2×Sc×NHost ×NMAG shall be processed by LMA. For data transmission, each host sends data directly to LMA. Thus, the data packets of Sd×NHost× NMAG shall be processed by LMA. Accordingly, we get the TPO of PMIP as follows.

In PMIP-LR, all hosts in network will exchange the PBU and PBA messages with LMA. Thus, the PBU and PBA messages of 2×Sc×NHost ×NMAG shall be processed by LMA. For data transmission, each host first sends data packet to LMA. Thus, the data packets of Sd×NHost×NMAG shall be processed by LMA. During data transmission, LMA decides to optimize the routing path between MN and CN, and it performs the localized routing operations by exchanging the LRI and LRA messages with both CN-MAG and MN-MAG. Thus, the LRI and LRA messages of 4×Sc×NHost ×NMAG shall be processed by LMA. Accordingly, we get the TPO of PMIP-LR as follows.

In PMIP-TRILL, for binding update, all hosts in network will perform the PBU operations with LMA. Thus, the PBU and PBA messages of 2×Sc×NHost ×NMAG shall be processed by eLMA. For data transmission, each host will perform the PBQ operations with eLMA. Thus, the PBQ and PQA messages of 2×Sc×NHost × NMAG shall be processed by eLMA. Accordingly, we get the TPO of PMIP-TRILL as follows.

C. Total Transmission Delay (TTD)

Let us denote binding query delay and data delivery delay by BQD and DDD, respectively. Then, the total transmission delay (TTD) can be represented as TTD = BQD + DDD. In PMIP, the binding query delay is 0. Thus, we get

In data delivery, the N data packets are directly sent to LMA. Then, LMA will forward these data packets to MN. Thus, the data delivery delay of PMIP can be represented as follows.

So, we get the total transmission delay of PMIP as follows,

The binding query delay of PMIP-LR is the same with that of PMIP. So, we get the BQD of PMIP-LR as follows.

In the PMIP-LR data delivery, CN sends first data packets directly to LMA. Then, LMA will forward those data packets to MN. During data transmission, LMA decides to optimize the routing path and performs the localized routing operations by exchanging the LRI and LRA messages with CN-MAG and MN-MAG. Then, the subsequent N-1 data packets will be delivered through the optimized route. Thus, the data delivery delay of PMIP-LR can be represented as follows.

So, we get the TTD of PMIP-LR as follows,

In PMIP-TRILL, the binding query delay from CN to MN can be calculated as follows. First, CN sends data packets to MAG. Then, MAG will perform the binding query operation with eLMA by exchanging PBQ and PQA messages with eLMA. So, we get the BQD of PMIP-TRILL as follows.

In the data delivery delay, N data packets are directly sent to MAG. Then, CN-MAG performs the query operation. After that, the N data packets will be delivered to MN through the optimized route. Thus, the data delivery delay of PMIP-TRILL can be represented as follows.

So, we get the TTD of PMIP-TRILL as follows,

4

D. Numerical Results

Based on the analysis given so far, we now compare the performance of the candidate schemes.

For numerical analysis, by referring to [9], the default value of each parameter is configured as follows: HMAG-LMA=10, HMAG-MAG=√NMAG, and Lwl=10 ms, Lw=2 ms, q=0.2, N=5, Tq=5 ms, NHost=500, NMAG=30, Sc=50 bytes, Sd=200 bytes, Bwl=11 Mbps and Bw=100 Mbps. Among these parameters, we note that Lwl, N, NHost, and NMAG may depend on the network conditions. Thus, we will compare the performance of candidate schemes by varying those parameter values. Fig. 4 and Fig. 5 compare the number of control/data messages to be processed by the central mobility anchor for different NHost and NMAG. We can see that the proposed PMIP-TRILL scheme provides smaller traffic processing overhead than the existing PMIP and PMIP-LR schemes. This is because all binding control and data messages shall be processed by LMA in the existing centralized schemes, whereas in the proposed PMIP-TRILL scheme only the binding control traffics are processed at eLMA. The PMIP-LR scheme gives the worst performance. This is because all the binding control packets and the first data packets shall be processed by LMA, together with the additional control messages for route optimization. The gaps of performance between the existing and proposed schemes get larger, as the number of hosts or mobile access gateways in the network increases. It is shown in the figure that the PMIP-TRILL gives the best performance among the candidate schemes.

0

200

400

600

800

1000

1200

1400

1600

100 200 300 400 500 600 700 800 900 1000

Traffic Processing Overhead (bytes)

x 103

NHost

PMIP

PMIP-LR

PMIP-TRILL

Figure 4. Impact of NHost on traffic processing overhead

0

500

1000

1500

2000

2500

3000

10 20 30 40 50 60 70 80 90 100

Traffic Processing Overhead (bytes)

x 103

NMAG

PMIP

PMIP-LR

PMIP-TRILL

Figure 5. Impact of NMAG on traffic processing overhead

Fig. 6 shows the impact of wireless link delay (Lwl) on total transmission delay. From the figure, we can see that the total transmission delay linearly increases, as Lwl gets larger, for all candidate schemes. It is shown that the proposed scheme gives better performance than the existing centralized schemes. It is also noted that PMIP-LR gives better performance than PMIP. This is because PMIP-LR delivers the first data packet to MN by way of LMA and the subsequent N-1 data packets is delivered through optimized route, while in PMIP all N data packets is delivered by way of LMA.

0

500

1000

1500

2000

2500

1 3 6 10 15 21 28 36 45 55

Total Transmission Delay (ms)

Lwl (ms)

PMIP

PMIP-LR

PMIP-TRILL

Figure 6. Impact of Lwl on total transmission delay

Fig. 7 shows the impact of number of data packets (N) on total transmission delay. From the figure, we can see that the total transmission delay linearly increases, as N gets larger, for all candidate schemes. It is also noted that PMIP-LR gives better performance than PMIP. This is because PMIP-LR deliver the first data packet to MN by way of LMA and the subsequent N-1 data packets is delivered through optimized route, while in PMIP all the N data packets is delivered by way of LMA. It is shown that the proposed scheme gives better performance than the existing schemes. The gaps of performance between the PMIP and PMIP-LR and PMIP-TRILL get larger, as the number of data packets increases.

0

2000

4000

6000

8000

10000

12000

14000

1 5 10 15 20 25 30 35 40 45

Total Transmission Delay (ms)

Number of data packets (N)

PMIP

PMIP-LR

PMIP-TRILL

Figure 7. Impact of N on total transmission delay

V. CONCLUSION

In this paper, we have proposed how to use and extend the PMIP protocol for mobility control in the TRILL-based mobile networks. For this purpose, each edge Routing Bridge (RB) will function as the Mobile Access Gateway (MAG) of PMIP, and an enhanced Local Mobility Anchor (eLMA) is used at the gateway of TRILL mobile network. With eLMA, the control

5

and data plane are separated, and the data path is optimized. Moreover, each data packet is encapsulated with the TRILL header, instead of using the IP-in-IP header.

From the numerical results, we can see that the PMIP-TRILL scheme provides better performance than the existing PMIP and PMIP-LR schemes in terms of traffic processing overhead and total transmission delay.

ACKNOLEDGEMENT

This research was supported by the Basic Science Research Program of NRF (2010-0020926).

REFERENCES [1] Morgan Stanley Report, Internet trends. http://www.morganstanley.com/,

2010.

[2] Erik Zamora and Wen Yu, Recent Advances on Simultaneous Localization and Mapping for Mobile Robots, IETE Technical Review, December 2013.

[3] Yung-Wey Chong and Tat-Chee Wan, Comparative Study on Hybrid Header Compression over Satellite-Wireless Networks, IETE Technical Review, December 2013.

[4] IETF RFC 6325, Routing Bridges (RBridges): Base Protocol Specification, July 2011.

[5] IETF RFC 5556, Transparent Interconnection of Lots of Links (TRILL): Problem and Applicability Statement, May 2009.

[6] M. Wasserman, et al., Transparent Interconnection of Lots of Links (TRILL) over IP, IETF Internet draft, draft-mrw-trill-over-ip-03.txt, October 2013.

[7] IETF RFC 5213, Proxy Mobile IPv6, August 2008. [8] S. Krishnan, et al., Localized Routing for Proxy Mobile IPv6, IETF

Internet Draft, draft-ietf-netext-pmip-lr-10, May 2012. [9] Makaya, C. and Pierre, S., An Analytical Framework for Performance

Evaluation of IPv6-based Mobility Management Protocols, IEEE

Transactions on Wireless Communication, Vol. 7. No. 3, pp. 972–983, 2008.