Bulk binding approach for PMIPv6 protocol to reduce handoff latency in IoT

International Journal of Electrical and Computer Engineering (IJECE)
Vol. 9, No. 3, June 2019, pp. 1894~1901
ISSN: 2088-8708, DOI: 10.11591/ijece.v9i3.pp1894-1901  1894
Journal homepage: http://iaescore.com/journals/index.php/IJECE
Bulk binding approach for PMIPv6 protocol to reduce handoff
latency in IoT
Adnan J. Jabir
Departement of Computer Science, University of Baghdad, Iraq
Article Info ABSTRACT
Article history:
Received May 17, 2018
Revised Oct 7, 2018
Accepted Nov 2, 2018
Mobility management protocols are very essential in the new research area of
Internet of Things (IoT) as the static attributes of nodes are no longer
dominant in the current environment. Proxy MIPv6 (PMIPv6) protocol is a
network-based mobility management protocol, where the mobility process is
relied on the network entities, named, Mobile Access Gateways (MAGs) and
Local Mobility Anchor (LMA). PMIPv6 is considered as the most suitable
mobility protocol for WSN as it relieves the sensor nodes from participating
in the mobility signaling. However, in PMIPv6, a separate signaling is
required for each mobile node (MN) registration, which may increase the
network signaling overhead and lead to increase the total handoff latency.
The bulk binding approaches were used to enhance the mobility signaling for
MNs which are moving together from one MAG to another by exchanging a
single bulk binding update message. However, in some cases there might be
several MNs move at the same time but among different MAGs. In this
paper, a bulk registration scheme based on the clustered sensor PMIPv6
architecture is proposed to reduce the mobility signaling cost by creating a
single bulk message for all MNs attached to the cluster. Our mathematical
results show that the proposed bulk scheme enhances the PMIPv6
performance by reducing the total handoff latency.
Keywords:
Bulk registration
Clustered Sensor PMIPv6
Handoff latency
PMIPv6
Copyright © 2019 Institute of Advanced Engineering and Science.
All rights reserved.
Corresponding Author:
Adnan J. Jabir,
Departement of Computer Science,
University of Baghdad,
10071, Al-Jadriya, Baghdad, Iraq.
Email: adnanjjabir@gmail.com
1. INTRODUCTION
In recent years, the Internet of Things (IoT) is getting much attention by industry and academia due
to its new trend of connecting different types of devices to the Internet without human intervention [1].
Wireless Sensor Networks (WSN) are considered as the core of IoT which are characterized by low
bandwidth, power, cost, processing, and memory [2], [3]. With the advent of the IoT and ubiquitous
computing, the need has emerged to design protocols for connecting WSN to Internet. Since Internet is the
most widespread network, connecting the WSN to it in order to disseminate the sensed data is essential for
making the IoT into a reality [4]. Integrating the Internet Protocol (IP) with WSN would indeed facilitate
WSN to interconnect with other IP networks and to capitalize the existing Internet infrastructure and IP-
applications to sensor networks [5], [6].
In addition, mobility has also an important role in realizing the IoT by achieving “anywhere,
anytime” communication. Mobility management protocols are very essential in the new research area of IoT,
it aims to track and locate the Mobile Nodes (MNs) efficiently to provide the user with full access to
information irrespective of their locations.
Host based mobility management protocols including MIPv6 [7], FMIPv6 [8], and HMIPv6 [9],
involve the MNs in the mobility process and generally introduce a significant overhead on the MNs which

Int J Elec & Comp Eng ISSN: 2088-8708 
Bulk binding approach for PMIPv6 protocol to reduce handoff latency in IoT (Adnan J. Jabir)
1895
should install complex mobility protocols. In addition, in case where an MN has no capability to transmit the
mobility related signaling, host-centric mobility management protocols will be no longer functional [10].
Therefore, methods for relieving MNs from participating in mobility process and reducing handoff latency,
packet loss, and communication path are indeed essential for the IoT [11].
Proxy Mobile IPv6 (PMIPv6) [12] was standardized by the IETF NETLMM working group as a
network-based mobility management protocol to solve these problems associated to the host-based protocols
by relocating the mobility procedure from MNs to the network components. PMIPv6 adds two functional
entities which are the Mobile Access Gateway (MAG) and the Local Mobility Anchor (LMA). The main role
of LMA is to maintain reachability to the MN address while it moves in the local PMIPv6 domain. The MAG
is responsible for detecting the MN movements and initiating the required signaling to register the MN with
the LMA.
The salient feature of the PMIPv6 is its network-based attribute, in which, the network detects the
node mobility and initiates the required mobility signals, so that, it relieves the MN from participating in the
handoff process. This feature allows MNs which installed only the conventional IP protocol to roam freely
between wireless networks without any need for the installation of complicated mobility management
protocols [13]. Considering both the network-based feature of PMIPv6 and the low capabilities of WSN
which makes the use of host-based mobility protocols like MIPv6 insufficient; researchers are motivated to
use PMIPv6 for managing the WSN mobility. For example, the SPMIPv6 was proposed by Islam in [14] as
an adaptation of the PMIPv6 protocol.
Although PMIPv6 fits with the WSN needs as it reduces the signaling cost, mobility cost, and the
level of energy consumption. However, in PMIPv6, for each node a separate handoff signaling should be
exchanged between MAG and LMA to register the MN’s new location. This may cause an extensive network
signaling if there is a large number of moving MNs [15].
Several research works were done to reduce the signaling overhead by introducing the bulk
registration principles. By using the bulk registration, a single binding message is required to register a
number of MNs which are moving together from one MAG to another. However, MNs are not always
moving together between two MAGs, they may move at the same time among different MAGs. These MNs
do not share any common property that allows the conventional bulk schemes to register them by a single
bulk message. In this paper, we propose a new bulk registration scheme based on the Clustered Sensor
PMIPv6 (CSPMIPv6) [16] architecture, named BCSPMIPv6, to address the problem of MNs registrations
which are moving at the same time but between different MAGs. Although MNs move among different
MAGs, a group of these MNs belong to the same cluster. Thus, the clustering principle provides the required
common property for those MNs in order to create a single bulk registration message.
2. RELATED WORK
Reducing the handoff signaling cost for the MNs which are moving together was addressed by
several researches. The mobile node group identifier was first introduced in [17] to identify a set of mobile
nodes which have common characteristics. It provides performing the required binding update function for a
set of MNs using only a single binding message rather than using multiple atomic binding per mobile node.
The mobile node group identifier is exchanged between LMA and MAGs and can be used to optimize the re-
registration or revocation functions for a set of MNs. The bulk re-registration allows MAGs to extend the
binding life time for a set of MNs belong to the same group by requesting a bulk re-registration message. In
this case, the re-registration messages are independent on the number of MNs; instead it depends on the
number of MAGs [18].
To optimize the bulk message size, authors in [15] proposed a new optimized bulk message format
based on the bitmap structure. In the conventional bulk registration, the message is built by simply
concatenating multiple PBU messages tied for the MNs which are belonging to the same group. In this case,
the mobility options are repeated for each MN which enlarges the bulk message size. Instead, by using the
bitmap structure, only a single instance of the mobility options is added to the bitmap matrix. Hence, the bulk
message size is reduced and there will be no need for breaking the bulk message into multiple parts to avoid
the packet drop.
Reducing the number of binding update messages for a numerous number of MNs moving together
has attracted researchers to apply the bulk registration principles into the network mobility domains. Authors
in [19] introduced a bulk binding update to the intelligent transportation systems where a number of mobile
routers are moving between access networks. By assigning a mobile node group identifier to this set of
moving routers, the overall signaling overhead required during the periodic binding update is reduced.
When mobile router moves between MAGs in the LMA domain, a large number of binding update
and binding acknowledgement messages is required to be exchanged for MNs registration or update their

 ISSN: 2088-8708
Int J Elec & Comp Eng, Vol. 9, No. 3, June 2019 : 1894 - 1901
1896
new locations. Since these MNs are moving together as a single unit, authors in [20], [21] introduced the use
of bulk registration to combine individual binding messages of all nodes in mobile network and sends them
together as a single binding message. Heading this way, handoff time and number of signaling messages to
be exchanged during mobile network movement are reduced.
The conventional bulk scheme solves the group mobility scenarios, where a group of MNs move
between two MAGs, by introducing the bulk PBU and PBA messages to register a group of MNs using a
single bulk message. In many scenarios, as shown in Figure 1, MNs may move individually among different
MAGs, but they move at the same time. However, the current solutions are not able to use the bulk
registration for such mobility scenario since there is no common characteristic shared by these MNs. This is
due to the distribution of MNs among different MAGs which makes it difficult to create a single bulk
message for these MNs. Therefore, the case when a number of MNs belonging to different MAGs, move at
the same time to other MAGs need to be addressed to enhance the PMIPv6 performance.
Figure 1. Node mobility scenarios
3. PROPOSED SCHEME
In this section, the CSPMIPv6 architecture with it signaling and control messages are presented first,
and then we deliberate on the proposed bulk registration scheme.
3.1. CSPMIPv6 architecture
The main idea behind the CSPMIPv6 is to divide the PMIPv6 domain into a number of local sub-
domains. Each sub-domain is represented as a cluster which constitutes a group of MAGs with a one Head
MAG (HMAG). The HMAG is one of the LMA’s domain MAGs which is elected to play the role of a cluster
head [16].
The main roles of the HMAG are: providing a local mobility management for each cluster, reducing
the handoff latency for the MNs moving inside the cluster, and is used to provide a sufficient buffering
scheme for the MNs during the handoff process [16].
In CSPMIPv6, a number of messages should be exchanged among entities to perform the
registration and handoff processes. These messages include the Proxy Binding Update (PBU), Proxy Binding
Acknowledge (PBA), Local PBU (LPBU), Local PBA (LPBA). In addition, all the LMA, HMAGs, and the
MAGs maintain their own binding list to store the required information of the MN’s current location [16].
The PBU and PBA are used to register the MN with the LMA in the same manner of the [12], while the
LPBU and the LPBA are used between MAGs and their HMAGs for the local MN’s registration.
When MN enters the CSPMIPv6 domain for the first time, the registration of the MN’s information
in the MAG, HMAG, and the LMA is required, as follows: once a MAG detects an attachment of an MN, it
sends an LPBU request to its HMAG contains the MN identifier (MN-ID). Upon receiving the LPBU by
HMAG, it searches its BUL for an MN entry. Since it is a new MN registration process, the HMAG will not
find a matching MN entry. Therefore, the HMAG sends a PBU request to the LMA containing the MN’s

1897
identifier and the requesting HMAG. Upon receiving the PBU, the LMA creates a new BCE for the MN and
sends a PBA message back to the requesting HMAG. Then the LMA sets up a global bidirectional tunnel
with the corresponding HMAG to be used for routing the traffic to and from the MN. Upon receiving the
PBA message, the HMAG registers the MN in its BUL and sends an LPBA message back to the
corresponding MAG containing the MN prefix. Then the HMAG configures a local tunnel with the MAG to
be used for routing the packets destined to the MN. Upon receiving the LPBA by the requesting MAG, it
registers the MN in its BUL and sends a router advertisement message to the requesting node.
3.2. Bulk CSPMIPv6 scheme
Conventional bulk schemes assume that MNs move as a group from one or more MAGS to one
destination MAG. As it can be seen from Figure 2, when a MAG detects the attachment of a group of MNs, it
creates one bulk PBU message to register these MNs with LMA [15]. However, if a number of MNs move at
the same time among different MAGs such that a single MN moves to one MAG, the LMA receives a
number of PBU equals to the number of the attached MNs, as shown in Figure 3. Thus, we propose a bulk
scheme which can cope with this case by creating a single bulk message for MNs move among different
MAGs.
Figure 2. Bulk Registration Signaling
Figure 3. Single Registration per MN
The proposed scheme is designed based on the CSPMIPv6 architecture to reduce the extensive
network signaling by using a single bulk binding message for every group of MNs which have a common
property. By introducing the bulk registration to the CSPMIPv6 architecture, the case when MNs move
individually among different MAGs can be solved by utilizing the HMAG entity. As shown in Figure 4,
when MNs move among different MAGs, each MAG sends a single LPBU message to the HMAG which in
turn collects all these requests and creates a single bulk PBU message to the LMA.
In the proposed scheme, when a group of MNs move between two MAGs belonging to one or more
clusters, the new MAG sends a single bulk LPBU to its HMAG for the MNs group registration. If the source
and destination MAGs belong to the same cluster, the HMAG replies by a single bulk PBA message for this
group registration. Adversely, if the MAGs belong to different clusters, the HMAG sends a bulk PBU to the
LMA. On the other hand, when MNs move between MAGs belonging to different clusters, each MAG sends
a normal or bulk LPBU to the HMAG which in turn sends a bulk PBU to the LMA for these MNs

 ISSN: 2088-8708
1898
registrations. The HMAG collects the PBU requests from its attached MAGs and sends it as a single bulk
PBU message to the LMA.
Figure 4. Proposed Bcspmipv6 Bulk Scheme Signaling
4. PERFORMANCE EVALUATION
In this section, the performance of the proposed BCSPMIPv6 is evaluated and compared against the
basic PMIP and the conventional bulk schemes (BPMIPv6), in terms of total handoff latency.
4.1. Analytical model
In this section, the total Handoff Latency (HL) is derived to compare the performance of the
proposed BCSPMIPv6 with the basic PMIPv6 and BPMIPv6 schemes. It is necessary to mention here that
we focus on the intra-domain mobility only (i.e. the MN moves within the same LMA domain). In addition,
for the proposed BCSPMIPv6 scheme we consider both the intra- and inter-cluster mobility within the same
domain. The performance notations used in the analysis, are presented in Table 1.
Table 1: Parameters for Performance Analysis
Parameter Description
Cx-y Hop count between nodes x and y
Sctrl Size of a control packet (byte)
Size of bulk message for MAGi or HMAGi
IPhs IPv6 packet header size
MNci Number if MNs in MAGi or HMAGi
t Unit transmission cost of packet per a wired link
Bw Link bandwidth
HLx Handoff Latency for x protocol
4.2. Signaling cost analysis
In this section, the MN’s handoff latency is derived for all schemes under consideration. We assume
the distance between any two entities is symmetric. In addition to simplify the mathematical calculations, we
are not going to consider the AAA functions and the processing cost for LMA, MAGs, HMAGs, and the
intermediate routers.
4.2.1. Basic PMIPv6
In the basic PMIPv6 protocol, the location update is performed by exchanging the PBU and PBA
messages between the LMA and MAGs when the MN moves from one MAG to another, which can be
derived as follows:
(1)
4.2.2. BPMIPv6
In BPMIPv6 scheme, the bulk PBU and PBA messages are exchanged between MAGs and LMA to
register the MNs which are moving as a group between MAGs. Therefore, the handoff latency for BPMIPv6

1899
can be derived as follows:
∑ (2)
( ) (3)
4.2.3. BCSPMIPv6
In BCSPMIPv6, the intra-cluster mobility is controlled by HMAG through exchanging the
LPBU/LPBA messages with its attached MAGs. On the other hand, the inter-cluster mobility requires the
involvement of LMA to update the MN’s information using the PBU and PBA messages. Accordingly, the
total handoff latency, which is the sum of intra-cluster handoff and inter-cluster handoff signaling costs, can
be derived as follows:
∑ ∑ (4)
5. NUMERICAL RESULTS
The numerical results based on the derivations in the previous section are presented here. The
assumptions and parameter values used in this research are shown in Table 2. In addition, a simulator has
been designed to simulate the MN mobility based on fluid-flow mobility model, in which, MNs are rarely
changing their speed and directions. The number of MNs crossing MAGs at every simulation time is
obtained by the simulator.
Table 2: Parameters Values
Parameter Default Value
t 2
v 25 (m/s)
R 500 (m)
Sctrl 96 bytes
IPhs 40 bytes
CMAG-LMA 16
CMAG-MAG √
MNc 1000
Mobility model Fluid-flow
Figure 5 shows the effect of the wired link delay on the handoff latency. It can be clearly seen that
the handoff latency of all schemes is increased with the increment of the wired link delay. However,
BCSPMIPv6 performs better than other schemes because of the reduction of the signaling required for
performing the inter- and intra-cluster mobility.
Figure 5. Impact of Wired Link Delay on the Handoff Latency
The handoff latency is proportional to the number of hopes between the communicating entities.
Figure 6 shows the impact of LMA-MAG distance on the handoff latency. The basic PMIPv6 incurs the

 ISSN: 2088-8708
1900
highest handoff latency due to its highly dependent on the LMA. The low effect of the distance on the
BCSPMIPv6 is due to the use of bulk registration for the inter-cluster mobility and the exclusion of LMA
from participating in the intra-cluster operations.
The handoff latency is also affected by the number of MNs moving between MAGs, which is shown
in Figure 7. It can be seen that when the number of MNs increased, the handoff delay is increased for all
schemes. However, BCSPMIPv6 presents the lowest handoff delay due to rely the intra-cluster mobility on
the HMAGs and the use of bulk registration for the inter-cluster mobility.
Figure 6. Impact of LMA-MAG Distance on the Handoff Latency
Figure 7. Impact of the Number of MNs on the Handoff Latency
6. CONCLUSIONS
In this paper, the BCSPMIPv6 is proposed as a new bulk scheme based on the CSPMIPv6
architecture to reduce the total handoff latency. The main idea of the clustered PMIPv6, which is the base of
BCSPMIPv6, is to divide the proxy domain into sub-domains formed as clusters, such that the intra-cluster
handoff is performed locally by the HMAG which thereby leads to low handoff latency. In our proposed
BCSPMIPv6 scheme, the MNs which are moving among different MAGs at the same time are registered by
using a single bulk PBU message. Our analysis and numerical results prove that the proposed BCSPMIPv6
scheme outperforms the previous work including the basic PMIPv6 and the conventional bulk scheme
BPMIPv6 in terms of handoff latency. Future investigations could focus on the implementation of the
BCSPMIPv6 scheme to validate its scalability using a high and random mobility environment.
REFERENCES
[1] Jamali, M.S., P. Kumar, and U.A. Khan, "Internet of Things: Architecture & Integration with Other Networks,"
First International Conference on Modern Communication & Computing Technologies (MCCT'14), 2014.
[2] Akyildiz, I.F., et al., "Wireless sensor networks: a survey," Computer networks, vol. 38(4), pp. 393-422, 2002.
[3] Mahyastuty, Veronica Windha, and M. Sigit Arifianto, "Wireless Sensor Network over High Altitude Platform."
Telecommunication, Computing, Electronics and Control (TELKOMNIKA), vol. 16(1), pp. 129-133, 2018.
[4] Rodrigues J.J. and P.A. Neves, "A survey on IP‐based wireless sensor network solutions," International Journal of
Communication Systems, 23(8), pp. 963-981, 2010.

1901
[5] Zinonos Z. and V. Vassiliou, "Inter-mobility support in controlled 6LoWPAN networks," in GLOBECOM
Workshops (GC Wkshps). 2010: IEEE.
[6] Chen Y.-S., C.-S. Hsu, and H.-K. Lee, "An enhanced group mobility protocol for 6LoWPAN-based wireless body
area networks," In Wireless Communications and Networking Conference (WCNC). 2013: IEEE.
[7] Johnson D., C. Perkins, and J. Arkko, "Mobility support in IPv6," IETF RFC 3775, 2004.
[8] Soliman H., L. Bellier, and K.E. Malki, "Hierarchical mobile IPv6 mobility management (HMIPv6),", IETF RFC
4140, 2005.
[9] Koodli, R., "Fast handovers for mobile" IPv6, IETF RFC 4068, 2005.
[10] Jabir A.J., Shamala S., Zuriati Z. and Hamid N.," A comprehensive survey of the current trends and extensions for
the proxy mobile IPv6 protocol," IEEE Systems Journal, 2015.
[11] Bojkovic Z., B. Bakmaz, and M. Bakmaz, "Some Challenging Issues for Internet of Things Realization,"
Proceedings of the 12th International Conference on Data Networks, Communications, Computers (DNCOCO '13),
March 2013.
[12] Gundavelli S., et al., "Proxy mobile ipv6," IETF RFC 5213, 2008.
[13] Hashim, Aisha, Azana Hafizah Mohd Aman, and Huda Adibah Mohd Ramli. "Throughput and Handover Latency
Evaluation for Multicast Proxy Mobile IPV6," Bulletin of Electrical Engineering and Informatics (BEEI), vol. 6(4),
pp. 311-316, 2017.
[14] Islam M.M. and E.-N. Huh, "Sensor proxy mobile IPv6 (SPMIPv6)—A novel scheme for mobility supported IP-
WSNs," Sensors, vol 11(2), pp. 1865-1887, 2011.
[15] Jeyatharan M. and C.-W. Ng, "Bulk PBU using Bitmaps", 2009.
[16] Jabir, A.J., et al., "A cluster-based proxy mobile IPv6 for IP-WSNs," EURASIP Journal on Wireless
Communications and Networking, pp. 1-17, 2012.
[17] Gundavelli S., et al., "Mobile Node Group Identifier option," 2009.
[18] Krishnan S., et al., "Bulk Re-registration for Proxy Mobile IPv6," 2009.
[19] Vasu K., S. Mahapatra and C. Kumar, "Bulk Binding Update Procedure for Pmipv6 Based Intelligent Tansportation
Systems" 2012.
[20] Jebaseeli S.A. and J. Sundararajan, "Reducing Binding update in NEMO Supported PMIPV6," International
Journal of Computer Trends and Technology (IJCTT), vol. 6(3), pp.150-157, 2013.
[21] Guan, Jianfeng, Ilsun You, Changqiao Xu, and Hongke Zhang, "The PMIPv6-Based Group Binding Update for IoT
Devices," Mobile Information Systems, 2016.
BIOGRAPHY OF AUTHORS
Adnan J. Jabir received the B.Sc. and M.Sc. degrees in Computer Science from the University of
Baghdad, Iraq in 1995 and 1998, respectively. He received his Ph.D. degree at the Department of
Communication Technology and Networks, Faculty of Computer Science and Information
Technology, University Putra Malaysia (UPM), Malaysia in 2015. He is currently a senior lecturer
at University of Baghdad, college of Science, computer Science Department. His major research
interests are wireless sensor networks, ubiquitous computing, and network mobility management.

Bulk binding approach for PMIPv6 protocol to reduce handoff latency in IoT

More Related Content

What's hot (18)

Similar to Bulk binding approach for PMIPv6 protocol to reduce handoff latency in IoT (20)

More from IJECEIAES (20)

Recently uploaded (20)

Bulk binding approach for PMIPv6 protocol to reduce handoff latency in IoT