Future Generation Computer Systems 88 (2018) 309–318

Contents lists available at ScienceDirect

Future Generation Computer Systems

journal homepage: www.elsevier.com/locate/fgcs

CoAP-based group mobility management protocol for the

Internet-of-Things in WBAN environment
Moneeb Gohar a, *, Jin-Ghoo Choi b , Seok-Joo Koh c
a
Department of Computer Science, Bahria University, Islamabad, Pakistan
b
Department of Information and Communication Engineering, Yeungnam University, Republic of Korea
c
School of Computer Science and Engineering, Kyungpook National University, Republic of Korea

highlights

• To support mobility management in web based Internet-of-Things environment is critical issue.

• A CoAP-based mobility management protocol, named CoMP has been proposed, but this protocol was designed for a single sensor node mobility.
• To overcome this limitation, this paper propose a CoAP-based group mobility management protocol, named CoMP-G in which one of the body sensor
will function as a coordinator and it will exchange all the control messages with web-of-things mobility management system (WMMS) on behalf of
other body sensors.

article info a b s t r a c t
Article history: The Constrained Application Protocol (CoAP) has been widely used, as the number of embedded sensors
Received 31 October 2017 or devices increases. To support mobility management in web based Internet-of-Things environment
Received in revised form 4 May 2018 is critical issue. For this purpose, a CoAP-based mobility management protocol, named CoMP has been
Accepted 2 June 2018
proposed, but this protocol was designed for a single sensor node mobility. However, it does not perform
Available online 6 June 2018
well in group-based mobility. To overcome this limitation, we propose a CoAP-based group mobility
management protocol, named CoMP-G. In the proposed scheme, one of the body sensor will function
Keywords:
CoAP as a coordinator and it will exchange all the control messages with web-of-things mobility management
Mobility management system (WMMS) on behalf of other body sensors. Besides, each WMMS maintains the information of the
Web-based group of body sensors. From the numerical analysis, we proved that the proposed scheme gives the best
IoT mobility performance in terms of total signaling and handover delay from the existing CoMP protocol.
WBAN © 2018 Published by Elsevier B.V.
Group mobility

1. Introduction major role in the field of health care [6–8]. Intelligent medical small
sensors can be worn on or implanted in the human body. These
Wireless Networks are the biggest components of network sensors measured the data and send it to external medical doctor.
industry where computer networking is done using wireless con- With these sensors the patient can move from place to another.
nections by connecting network nodes [1]. With passing days’ Many people die from different fatal diseases, when it is diagnosed
different devices like mobile phones are now become the basic lately. For this purpose, the Wireless Body Area Network is used to
needs. So there is a need of Internet Protocol which are especially detect early and prevent from the fatal diseases [8–10]. A wireless
designed for the mobile devices when they move from one net- body area network are the specific types of sensor network which
work to another. Nowadays, we are not limited to one network are designed to handle multiple medical sensors placed inside or
devices, such as the communication becomes Machine to Machine, outside of the body. The generalized model of WBAN is shown in
Human to Machine and Machine to Human and Human to Human, Fig. 1.
this major penetration introduced the idea of Internet of Things These days, each person has multiple devices that can commu-
(IoT) [2–5]. IoT serves in many fields like Medical sciences, elec- nicate each other. The communication pattern can be machine-to-
tronic engineering as well as mechanical engineering. IoT play a machine, human-to-machine, machine-to-human, and human-to-
human. Anything we think of can exchange information to serve
human beings. It is the concept of IoT [11]. IoT has found many
* Corresponding author. applications in various fields such as medical science, electronic
E-mail addresses: [email protected] (M. Gohar), [email protected]
(S. Koh). engineering as well as mechanical engineering [12]. Suppose we

https://doi.org/10.1016/j.future.2018.06.003
0167-739X/© 2018 Published by Elsevier B.V.
310 M. Gohar et al. / Future Generation Computer Systems 88 (2018) 309–318

protocol, named CoMP has been proposed, but this protocol was
designed for a single sensor node mobility. However, it does not
perform well in group-based mobility [18]. To overcome this lim-
itation, we propose a CoAP-based group mobility management
protocol, named CoMP-G, in which one of the sensor will function
as a coordinator and it will exchange all the control messages
with web-of-things mobility management system (WMMS) on
behalf of other body sensors. Besides, each WMMS maintains the
information of the group of mobile sensors.
The rest of this paper is organized as follows. Section 2 de-
scribes the existing candidate mobility schemes for comparison.
In Section 3 describes the proposed group-based mobility scheme.
Section 4 demonstrates the performance analysis by comparing the
proposed and existing scheme by numerical analysis and results.
Section 5 concludes this paper.

2. Existing candidate scheme

The mobility management protocols in [19–26] may not be fit

for supporting the mobility Management of CoAP sensor nodes.
This is because the CoAP sensor nodes generally have constrained
CPU processing power and memory capacities. The IETF have not
Fig. 1. Generalized network model of WBAN. addressed these limitations on the design of mobility protocols.
Therefore in [18], the author proposed CoAP-based Mobility
Management Protocol (CoMP), which provide mobility manage-
have electric meters in our home that measures the amount of ment for CoAP sensor nodes. But this protocol was designed for a
electrical energy consumption. If the meter also has the network single sensor node mobility. In this paper, we will use the group of
ability and sends meter readings to the server, the bills are gener- sensors attached to human body.
ated automatically with less time. In this scenario, a machine-to-
machine communication is involved. 2.1. Network model of existing group mobility management using
IoT is very useful in the field of health care. Each patient can CoAP
carry many sensor devices that evaluate and analyze his or her
health state. If the devices can share the measured data with each Fig. 2 shows the network model of existing mobility manage-
other or report to the server in the Internet, doctors can know the
ment using CoAP. The existing scheme consider a group of sensors
exact health condition of the patient easily and send an alert in
attached to human body which monitors the measured data. All
the emergency situation. The Wireless Body Area Network (WBAN)
the sensors exchange the control signaling messages with Access
is designed for this application, which is a specialized wireless
Router (AR). The AR domain contains Full-Function Device (FFD).
network handling multiple medical and environmental sensors
The web-of-things mobility management system (WMMS) main-
inside or outside of the human body [13]. WBAN can be used with
ECG electrodes, LM35 temperature sensor, blood glucose sensor, tains the information of each of sensors. The IP addresses of sensors
and much more [14]. are permanent, while the IP addresses of AR are temporary. Firstly,
In most wireless sensor network applications, sensor nodes are the body sensors attached with the previous AR and then it moves
small, lightweight, inexpensive, and requires to operate perma- to the new AR domain.
nently. Current technology is not able to meet most of the require-
ments for the permanent operation of the sensor node due to the 2.2. Initial registration and data delivery
slow development of battery technology is realized. Due to the
limitation of sensor battery, the researcher designed a new Con- The existing CoMP scheme is based on CoAP protocol [18]. As
strained Application Protocol (CoAP). CoAP is a RESTful application shown in Fig. 3, when a set of body sensors are attached with
protocol for low power and constrained devices in Wireless Sensor
access router (AR), then each body sensor sends POST Request for
Networks (WSNs) or Low-power and Lossy Networks (LLNs). The
Registration message to WMMS via AR (Step 1,2). Upon reception
basic communication model for CoAP is the client–server model
of this POST Request for Registration message from each of the body
that exchanges messages between clients and servers. CoAP also
sensors, WMMS will update the database and responds with a ACK
provides group communication for effective communication with
Response for Registration to each of the body sensors via AR (Step
numerous sensors [15–17].
The Internet Engineering Task Force (IETF) has designed 3,4).
the CoAP group communication using IP multicast. However, On the other hand, the CoAP web client also registered with
multicast-based group communication scheme may not be reliable WMMS by exchanging POST Request for Registration and ACK Re-
in WSNs. It is difficult to receive a response message for CoAP client, sponse for Registration messages (Step 5,6,7,8). As the CoAP web
since connectivity between sensor and client may not be stable. client want to communicate with body sensors, it sends GET Re-
To solve these problems, the unicast-based group communication quest for Discovery message for each body sensor to WMMS via
scheme was developed. However, such unicast-based scheme has AR (Step 9,10). After finding each body sensor in the database,
low performance for transmission delay and large network over- the WMMS replies with ACK Response for Discovery message to
head. These group communication is for static nodes. CoAP web client via AR on behalf of each body sensor (Step 11,12).
In wireless domain, it is necessary to provide mobility to sensor After discovery, the CoAP web client and each body sensors can
nodes. So for this purpose, a CoAP-based mobility management exchange the data via AR’s.
 M. Gohar et al. / Future Generation Computer Systems 88 (2018) 309–318 311

Fig. 2. Network model of the existing scheme.

Fig. 3. CoMP initial registration and data delivery.

2.3. Handover operations WMMS via AR for each request on the basis of body sensor (Step
7,8).
In this section, we consider the case in which the group of We now assume that the body sensors changes its point of
sensors moves from previous AR (p-AR) to new AR (n-AR) as shown attachment in the same network domain. When the body sensors
in Fig. 4. In order to perform the handover operation, each of the is detached from p-AR and attached to n-AR and received new
body sensors first detect the radio signal strength (RSS) from the temporary address from new AR (n-AR), each body sensor notifies
previous AR. When RSS drops below by a certain threshold value,
the WMMS by sending PUT Request for Binding Update via n-AR
each of the body sensor starts handover operation by sending PUT
(Step 9,10). The WMMS update its database for each body sensor
Request for Holding to WMMS via p-AR (Step 1, 2). The WMMS
updates the H_Flag status to 1 for all the body sensors. After that, and respond with ACK Response for Binding Update to each body
the WMMS replies with ACK Response for Holding to each body sensor via n-AR (step 11,12). The body sensors also updates the
sensor via p-AR (Step 3,4). The WMMS also forwards the PUT CoAP web client by sending PUT Request for Binding Update via n-
Request for Holding on basis of each body sensor to CoAP web client AR to CoAP web client (Step 13,14,15). The CoAP web client update
via AR (Step 5,6). The CoAP client updates its cache for each sensor. its cache for each body sensor and replies with ACK Response
The CoAP web client respond with ACK Response for Holding to for Binding Update to each body sensor via n-AR (Step 16,17,18).
312 M. Gohar et al. / Future Generation Computer Systems 88 (2018) 309–318

Fig. 4. CoMP handover operation.

Now, each body sensor and CoAP web client communication is 3.2. Initial registration and data delivery
established through new AR.
The main purpose of the registration phase is to reduce the
3. Proposed scheme amount of control messages. Fig. 6 shows the detail initial regis-
tration and data delivery of the proposed scheme. When a group
This section first describes the network model of the proposed of body sensors enter a CoMP AR domain and the coordinator is
attached to AR, it sends aggregated POST Request for Registration
scheme and then describing the registration and handover opera-
message, containing the information on group, to WMMS via AR
tions.
(Step 1, 2). Upon reception of this POST Request for Registration mes-
sage from the coordinator, WMMS will make group of body sensors
3.1. Network model of group mobility management using CoAP and responds with a ACK Response for Registration to coordinator
via AR (Step 3, 4).
Fig. 5 shows the network model of group mobility management On the other hand, the CoAP web client also registered with
using CoAP. We consider a group of sensors attached to human WMMS by exchanging POST Request for Registration and ACK Re-
body which monitors the measured data. One of sensor function sponse for Registration messages (Step 5,6,7,8). As the CoAP web
as a coordinator which exchange the control signaling messages client want to communicate with body sensors, it sends GET Re-
with Access Router (AR) on behalf of other sensors. The AR domain quest for Discovery message for finding group of body sensors to
contains Full-Function Device (FFD). The web-of-things mobility WMMS via AR (Step 9,10). After finding group of body sensors in
management system (WMMS) maintains the information of the the database, the WMMS replies with ACK Response for Discovery
group of sensors which is required to perform mobility as shown in message to CoAP web client via AR (Step 11,12). After discovery, the
Table 1. The IP addresses of body sensors are permanent, while the CoAP web client and coordinator can exchange the data via AR’s.
IP addresses of AR’s are temporary. H_Flag indicate the handover
status of the group of sensors. If the value of H_Flag is 1, then the 3.3. Handover operations
corresponding group of body sensors become in handover status.
If the value is 0, then the corresponding group of body sensors are In this section, we consider the case in which the group of
not in handover status. A lifetime is the time in which the binding sensors moves from previous AR (p-AR) to new AR (n-AR) as
of permanent addresses of the group of sensors and temporary shown in Fig. 7. In order to perform the handover operation, the
address of the AR are effective. coordinator first detect the radio signal strength (RSS) from the
Initially, the coordinator communicates with CoAP web client previous AR. When RSS drops below by a certain threshold value,
in the previous access router (p-AR) domain, and then it moves to the coordinator starts handover operation by sending PUT Request
a new access router (n-AR) by handover. for Holding to WMMS via p-AR (Step 1,2). The WMMS updates the
 M. Gohar et al. / Future Generation Computer Systems 88 (2018) 309–318 313

Table 1
Web-of-things mobility management system (WMMS).
Group P_Addr (Sensor IP address) T_Addr (AR IP address) H_Flag Lifetime
1 P_Addr_1
1 P_Addr_2 T_Addr
1 P_Addr_3

Fig. 5. Network model of the proposed scheme.

Fig. 6. Initial registration and data delivery.

H_Flag status to 1. After that, the WMMS replies with ACK Response (Step 9,10). The WMMS update its database and respond with ACK
for Holding to coordinator via p-AR (Step 3,4). The WMMS also
Response for Binding Update to coordinator via n-AR (step 11,12).
forwards the PUT Request for Holding for the group to CoAP web
client via AR (Step 5,6). The CoAP client updates its cache. The CoAP The coordinator also updates the CoAP web client by sending
web client respond with ACK Response for Holding to WMMS via AR PUT Request for Binding Update via n-AR to CoAP web client (Step
(Step 7,8).
We now assume that the coordinator changes its point of at- 13,14,15). The CoAP web client update its cache and replies with
tachment in the same network domain. When the coordinator ACK Response for Binding Update to coordinator via n-AR (Step
is detached from p-AR and attached to n-AR and received new
16,17,18). Now, CoAP coordinator and CoAP web client communi-
temporary address from new AR (n-AR), the coordinator notifies
the WMMS by sending PUT Request for Binding Update via n-AR cation is through new AR.
314 M. Gohar et al. / Future Generation Computer Systems 88 (2018) 309–318

Fig. 7. Handover operations.

Table 2 by the transmission delay of a message with size S sent from x

Parameters used for analysis. to y via ‘wired’ link, where Hx-y represents the number of wired
Parameters Description hops between node x and node y. Then, Tx-y (S , Hx-y ) is expressed
Sc Size of control packets (bytes) as Tx-y (S , Hx-y ) = Hx-y × [(S /Bw ) + Lw + Tq ].
NS Number of sensors in the domain
Bw Wired link bandwidth (Mbps)
Bw l Wireless bandwidth (Mbps)
4.2. Total signaling delay (TSD)
Lw Wired link delay (ms)
Lwl Wireless link delay (ms) As shown in Fig. 1, when the body sensors are attached to an AR,
Ha-b Hop count between node a and b in the network each body sensor exchange POST Request for Registration and ACK
q Wireless link failure probability
Tq Average queuing delay at each node
Response for Registration messages with WMMS. After that WMMS
updates its database. Accordingly, we get the registration delay
(RD) of CoMP as follows.
4. Performance analysis RDCoMP = NS × {2TC -FFD (Sc ) + 2TFFD-AR (Sc ) + 2TAR-WMMS (Sc )}. (1)

For performance analysis, we compare the registration and In CoMP, the data delivery delay (DDD) from CoAP web client
handover delays for the two candidate mobility schemes: CoMP, to body sensors can be calculated as follows. First, the CoAP web
and CoMP-G. client registered with WMMS. After registration, the CoAP web
client want to communicate with body sensors. For finding the
4.1. Analysis model body sensors, CoAP web client exchanges GET Request for Discovery
and ACK Response for Discovery messages for each body sensor with
We consider a network illustrated in Fig. 8, in which each WMMS via AR. After discovery, the CoAP web client and each body
wired/wireless link is represented by bandwidth, latency, and av- sensors can exchange the data via AR’s.
erage queuing delay. We adopt a generic model for Multiple Access Thus, the data delivery delay (DDD) of CoMP can be represented
Control (MAC) scheme to focus on the analysis of registration as follows,
delay and handover delay associated with the proposed mobility
scheme. DDDCoMP = NS × {2TC -R (Sc ) + 2TAR-WMMS (Sc )}. (2)
We summarize the notations used in our analysis in Table 2.
So, we obtain the TSD of CoMP as
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, where TSDCoMP = RDCoMP + DDDCoMP . (3)
each message can experience the failure at the probability of q by
using ‘i.i.d’ error model. Then, Tx-y (S) can be expressed as Tx-y (S) = As shown in Fig. 4, the CoMP-G uses the aggregated POST Re-
[1/(1−q)]×[(S /Bwl )+Lwl ]. In the meantime, we denote Tx-y (S , Hx-y ) quest for Registration and ACK Response for Registration messages
 M. Gohar et al. / Future Generation Computer Systems 88 (2018) 309–318 315

Fig. 8. Network model for performance analysis.

between coordinator and WMMS. Thus, we get the RD of COMP-G the CoAP web client by exchanging PUT Request for Binding Update
as and ACK Response for Binding Update messages. Now, each body
sensor and CoAP web client communication is established through
RDCoMP-G = 2TC -FFD (Sc ) + 2TFFD-AR (Sc ) + 2TAR-WMMS (Sc ) . (4) new AR. Accordingly, we get the HD of CoMP as follows.
In CoMP-G, the data delivery delay (DDD) from CoAP web client
HDCOMP = NS × {2TC -FFD (Sc ) + 2TFFD-AR (Sc ) + 2TAR-WMMS (Sc )
to body sensors can be calculated as follows. First, the CoAP web
client registered with WMMS. After registration, the CoAP web + 2TC -R (Sc ) + 2TAR-WMMS (Sc ) + 2TC -FFD (Sc )
client want to find the body sensors to communicate with them. + 2TFFD-AR (Sc ) + 2TAR-WMMS (Sc ) + 2TC -FFD (Sc )
For this purpose, the CoAP web client exchanges GET Request for + 2TFFD-AR (Sc ) + 2TAR-AR (Sc ) + 2TC -R (Sc )}. (7)
Discovery and ACK Response for Discovery messages for body sensor
with WMMS via AR. After discovery, the CoAP web client and each In the proposed CoMP-G scheme, when the group of sensors
moves from previous AR (p-AR) to new AR (n-AR) as shown in
body sensors can exchange the data via AR’s.
Fig. 5. Firstly, each of the body sensors detect the RSS signal, if it
Thus, the data delivery delay (DDD) of CoMP-G can be repre-
drops below by a certain threshold value, the coordinator starts
sented as follows,
handover operation by exchanging PUT Request for Holding and
DDDCoMP-G = 2TC -R (Sc ) + 2TAR-WMMS (Sc ) . (5) ACK Response for Holding with WMMS via p-AR. Then, WMMS also
exchange the PUT Request for Holding and ACK Response for Holding
So, we obtain the TSD of CoMP-G as messages with CoAP web client via AR. The CoAP client updates its
cache for the group.
TSDCoMP-G = RDCoMP-G + DDDCoMP-G . (6) When body sensors change its point of attachment and de-
tached from p-AR and attached to n-AR, coordinator exchanges
PUT Request for Binding Update and ACK Response for Binding Update
4.3. Handover delay (HD)
messages with WMMS. Coordinator updates the CoAP web client
by exchanging PUT Request for Binding Update and ACK Response for
In this section, we consider handover delay in which the group Binding Update messages. Now, each body sensor and CoAP web
of sensors moves from previous AR (p-AR) to new AR (n-AR) as client communication is established through new AR. Accordingly,
shown in Fig. 2. Each of the body sensors first detect the RSS we get the HD of CoMP-G as follows.
signal from the previous AR. When RSS drops below by a certain
threshold value, each of the body sensor starts handover operation HDCOMP-G = 2TC -FFD (Sc ) + 2TFFD-AR (Sc ) + 2TAR-WMMS (Sc )
by exchanging PUT Request for Holding and ACK Response for Holding + 2TC -R (Sc ) + 2TAR-WMMS (Sc ) + 2TC -FFD (Sc )
with WMMS via p-AR. Then, WMMS also exchange the PUT Request
for Holding and ACK Response for Holding messages for each of body
+ 2TFFD-AR (Sc ) + 2TAR-WMMS (Sc ) + 2TC -FFD (Sc )
sensor with CoAP web client via AR. The CoAP client updates its + 2TFFD-AR (Sc ) + 2TAR-AR (Sc ) + 2TC -R (Sc ) . (8)
cache for each of body sensor.
When body sensors change its point of attachment and de- 4.4. Numerical results
tached from p-AR and attached to n-AR, each body sensor ex-
changes PUT Request for Binding Update and ACK Response for Bind- Based on the equations, we compare the performance of the ex-
ing Update messages with WMMS. Each body sensor also updates isting and proposed schemes. For numerical analysis, we configure
316 M. Gohar et al. / Future Generation Computer Systems 88 (2018) 309–318

Table 3
Default parameter values.
Parameter Default Minimum Maximum
Lwl 15 1 55
Ns 10 1 50
q 0.5 0.1 0.9
HAR-WMMS 5 1 55
HAR-AR 3
HC -AR , HC -FFD , HFFD-AR 1
Tq 5
Lw 2
Sc 50 bytes
Bw l 0.25 Mbps
Bw 10 Mbps

the default parameter values, as described in Table 3, by referring

to [18,27].

4.4.1. Total signaling delay (TSD)

Fig. 9. Impact of Lwl on total signaling delay.
Fig. 9 illustrates the impact of wireless link delay (Lwl ) on total
signaling delay. We can see in the figure that the total signaling
delay linearly increases, as Lwl gets larger for both the candidate
schemes. It is shown in the figure that the proposed CoMP-G
scheme perform better than the existing CoMP scheme. This is be-
cause coordinator in CoMP-G scheme perform signaling operation
with WMMS on behalf of body sensors.
Fig. 10 shows the impact of number of sensors (NS ) on to-
tal signaling delay. We observe from the figure that CoMP gives
worse performance than the CoMP-G. This is because each of the
body sensor exchanges signaling messages with WMMS and also
the CoAP web client sends discovery messages to find each body
sensors addresses to WMMS. In contrast, the proposed CoMP-G
scheme is not effected by the number of sensors. This is because in
the proposed scheme the coordinator can exchange the signaling
messages with WMMS on behalf of body sensors.
Fig. 11 shows the impact of wireless link failure probability
on total signaling delay. We observe that the total signaling delay
linearly increases, as q gets larger for both the candidate schemes.
This is because both the schemes use wireless links for the body
Fig. 10. Impact of NS on total signaling delay.
sensors. We can see in the figure that the proposed scheme gives
better performance than the existing scheme.
Fig. 12 illustrates the impact of hop count between AR and
WMMS on total signaling delay. From the figure, we can see that
hop count between AR and WMMS gives significant impact on both
the candidate schemes. In particular, CoMP is more sensitive. This
is because each of the body sensor exchanges signaling messages
with WMMS. While CoMP-G is less sensitive, this is because the
coordinator can exchange the signaling messages with WMMS on
behalf of body sensors. We can see in the figure that the proposed
scheme performs better than the existing scheme.

4.4.2. Handover delay (HD)

Fig. 13 shows the impact of wireless link delay (Lwl ) on total
signaling delay. In the figure, we can see, that the handover delay
linearly increases, as Lwl gets larger for both the candidate schemes.
The proposed CoMP-G scheme gives better performance than the
existing CoMP scheme. This is because in the proposed scheme the
coordinator can exchange the signaling messages with WMMS on
behalf of body sensors.
Fig. 11. Impact of q on total signaling delay.
Fig. 14 illustrates the impact of number of sensors (NS ) on
handover delay. We notice from the figure that CoMP perform
worse than the proposed CoMP-G scheme. This is because each of
the body sensor exchanges signaling messages with WMMS. On the Fig. 15 shows the impact of wireless link failure probability on
other hand, the proposed CoMP-G scheme is not effected by the handover delay. We can see in the figure that the total signaling
number of sensors. This is because the coordinator can exchange delay linearly increases, as q gets larger for both the candidate
the signaling messages with WMMS on behalf of body sensors. schemes. This is because both the schemes use wireless links for
 M. Gohar et al. / Future Generation Computer Systems 88 (2018) 309–318 317

Fig. 12. Impact of HAR-WMMS on total signaling delay.

Fig. 14. Impact of NS on handover delay.

Fig. 13. Impact of Lwl on handover delay.

Fig. 15. Impact of q on handover delay.

the body sensors. As shown in the figure, the proposed scheme

gives better performance than the existing scheme.
Fig. 16 illustrates the impact of hop count between AR and
WMMS on handover delay. From the figure, we can see that CoMP
is more sensitive as hop count between AR and WMMS increases.
This is because each of the body sensor exchanges signaling mes-
sages with WMMS. While CoMP-G is less sensitive, this is because
the coordinator can exchange the signaling messages with WMMS
on behalf of body sensors.

5. Conclusion

In this paper, we propose a CoAP-based group mobility man-

agement protocol, named CoAP-G. In the proposed scheme, one
of the sensor will function as a coordinator and it will exchange
Fig. 16. Impact of HAR-WMMS on handover delay.
all the control messages with web-of-things mobility management
system (WMMS) on behalf of other sensors. Besides, each WMMS
maintains the information of the group of mobile sensors. From the
numerical analysis, we proved that the proposed scheme gives the
best performance in terms of total signaling and handover delay
from the existing CoMP protocol.
318 M. Gohar et al. / Future Generation Computer Systems 88 (2018) 309–318

References [24] A. Berguiga, H. Youssef, Efficient mobility management in 6LoWPAN wireless

sensor networks, in: Proceedings of The Ninth International Conference on
[1] A. Al-Fuqaha, M. Guizani, M. Mohammadi, M. Aledhari, M. Ayyash, Internet of Wireless and Mobile Communications (ICWMC 2013), Nice, France, 21–26 July
things: a survey on enabling technolo-gies, protocols, and applications, IEEE 2013, pp. 244–250.
Commun. Surv. Tutor. 17 (4) (2015) 2347–2376. [25] S. Gligoric. Krco, D. Drajic, S. Jokić, B. Jakovljevic, M2M device management in
[2] Awais Ahmad, Anand Paul, M. MazharRathore, Hangbae Chang, Smart cyber LTE networks, in: Proceedings of 19th Telecommunications Forum (TELFOR),
society: Integration of capillary devices with high usability based on Cyber– Belgrade, 22–24 November 2011, pp. 414–417.
Physical system, Future Gener. Comput. Syst. 56 (2016) 493–503. [26] F. Ganz, R. Li, P. Barnaghi, H. Harai, A Resource mobility scheme for service-
[3] Guinard. Dominique, et al., Towards physical mashups in the web of things, continuity in theinternet of things, in: Proceedings of 2012 IEEE International
in: Networked Sensing Systems (INSS), Sixth International Conference on, Conference on Green Computing and Communications, Besançon, France, 20–
IEEE, 2009. 23 November 2012, pp. 261–264.
[4] M.A. Razzaque, M. Milojevic-Jevric, A. Palade, S. Clarke, Middleware for inter- [27] C. Makaya, S. Pierre, An analytical framework for performance evaluation of
net of things: a survey, IEEE Internet Things J. 3 (1) (2016) 70–95. IPv6-based mobility management protocols, IEEE Trans. Wirel. Commun. 7 (3)
[5] J. Jin, J. Gubbi, S. Marusic, M. Palaniswami, An informa-tion framework for (2008) 972–983.
creating a smart city through internet of things, IEEE Internet Things J. 1 (2)
(2014) 112–121.
[6] S.M.R. Islam, D. Kwak, M.H. Kabir, M. Hossain, K.-S. Kwak, The internet of things
for health care: a comprehensive survey, IEEE Access 3 (2015) 678–708. Moneeb Gohar received B.S. degree in Computer Science
[7] M. Khan, B.N. Silva, K. Han, Internet of things based energy aware smart home from University of Peshawar, Pakistan, and M.S. degree in
control system, IEEE Access 4 (2016) 7556–7566. Technology Management from Institute of Management
[8] Mohammad Ghamari, et al., A survey on wireless body area networks for Sciences, Pakistan, in 2006 and 2009, respectively. He
ehealthcare systems in residential environments, MDPI Sens. (2016). also received Ph.D. degree from the School of Computer
[9] Riccardo Cavallari, et al., Survey on wireless body area networks: Technologies Science and Engineering in the Kyungpook National Uni-
and design challenges, IEEE Commun. Surv. Tutor. 16 (3) (2014). versity, Korea, in 2012. From September 2012 to Septem-
[10] Raju Sharma, et al., Wireless body area nework–a review, Int. J. Eng. Sci. (2016). ber 2014, he worked as a Post-Doctoral researcher for
[11] Chen Shanzhi, Hui Xu, Dake Liu, Bo Hu, Hucheng Wang, A vision of IoT: Ap- Software Technology Research Center (STRC) in Kyung-
plications, challenges, and opportunities with china perspective, IEEE Internet pook National University, Korea. From September 2014 to
Things J. 1 (4) (2014) 349–359 52. September 2016, he worked as a Foreign Assistant Pro-
[12] Kumar Vinoth, Ontology based public healthcare system in Internet of Things fessor with the Department of Information and Communication Engineering in
(IoT), Procedia Comput. Sci. 50 (2015) 99–102. the Yeungnam University. He has been as a Senior Assistant Professor with the
[13] Purna B. Samal, Ping Jack Soh, Guy AE Vandenbosch, UWB all-textile antenna Department of Computer Science in the Bahria University Islamabad, Pakistan since
with full ground plane for off-body WBAN communications, IEEE Trans. An- September 2016. His current research interests include Network Layer Protocols,
tennas Propag. 62 (1) (2014) 102–108. Wireless Communication, Mobile Multicasting, Wireless Sensors Networks, TRILL,
[14] Devita Gabriele, Alan Chi Wai Wong, Mark Dawkins, K. Glaros, U. Kiani, Franco Big Data and Internet Mobility.
Lauria, V. Madaka, et al., A 5mW multi-standard Bluetooth LE/IEEE 802.15.
6 SoC for WBAN applications, in: European Solid State Circuits Conference
(ESSCIRC), ESSCIRC 2014-40th, IEEE, 2014, pp. 283–286. Jin-Ghoo Choi received his PhD degree from the School
[15] Bormann Carsten, Angelo P. Castellani, Zach Shelby, CoAP: An application of Electrical Engineering and Computer Science, Seoul Na-
protocol for billions of tiny internet nodes, IEEE Internet Comput. 16 (2) (2012) tional University in 2005. From 2006 to 2007, he worked
62–67. for Samsung Electronics as a senior engineer. In 2009, he
[16] Kovatsch Matthias, Simon Duquennoy, Adam Dunkels, A low-power CoAP for was with the Department of Electrical and Computer En-
Contiki, in: Mobile Adhoc and Sensor Systems (MASS), IEEE 8th International gineering in The Ohio State University as a visiting scholar.
Conference on, IEEE, 2011. He joined the Department of Information and Commu-
[17] Z. Shelby, K. Hartke, C. Bormann, The Constrained Application Protocol (CoAP). nication Engineering in Yeungnam University as a fac-
The Internet Engineering Task Force, Fremont, CA, RFC 7252, 2014. ulty member in 2010. His research interests include per-
[18] S.M. Chun, H.S. Kim, J.T. Park, CoAP-based mobility management for the Inter- formance analysis of communication networks, resource
net of Things, Sensors 15 (2015) 16060–16082. management in wireless networks, and wireless sensor
[19] A.J. Jara, D. Fernandez, P. Lopez, M.A. Zamora, A.F. Skarmeta, Lightweight MIPv6 network.
with ipsec support, Mob. Inf. Syst. 10 (2014) 37–77.
[20] A.J. Jara, D. Fernandez, P. Lopez, M.A. Zamora, A.F. Skarmeta, Lightweight
mobile IPv6: A mobility protocol for enabling transparent IPv6 mobility in the Seok-Joo Koh received the B.S. and M.S. degrees in Man-
internet of things, IEEE Glob. Commun. Conf. 12 (2013) 2791–2797. agement Science from KAIST in 1992 and 1994, respec-
[21] H. Sungmin, D. Kim, M. Ha, S. Bae, S. Park, W. Jung, J. Kim, SNAIL: An IP- tively. He also received Ph.D. degree in Industrial Engi-
based wireless sensor network approach to the internet of things, IEEE Wirel. neering from KAIST in 1998. From August 1998 to February
Commun. 17 (2010) 34–42. 2004, he worked for Protocol Engineering Center in ETRI.
[22] A.J. Jara, M.A. Zamora, A.F.G. Skarmeta, An architecture based on internet of He has been as a professor with the school of Computer
things to support mobility and security in medical environments, in: Pro- Science and Engineering in the Kyungpook National Uni-
ceedings of 7th IEEE Consumer Communications and Networking Conference versity since March 2004. His current research interests
(CCNC), Las Veagas, NV, USA, 9–12 January 2010, pp. 1–5. include mobility management in the future Internet, IP
[23] C. Kai, Y. Zhimin, C. Rongyi, L. Chenghao, A handoff algorithm based on care-of mobility, multicasting, and SCTP. He has so far participated
address pool for hierarchical mobile IPv6, in: Proceedings of 3rd International in the international standardization as an editor in ITU-T
Conference on Pervasive Computing and Applications, Alexandria, 6–8 Octo- SG13 and ISO/IEC JTC1/SC6.
ber 2008, pp. 302–306.