Content uploaded by Smita Sharma
Author content
All content in this area was uploaded by Smita Sharma on Feb 28, 2022
Content may be subject to copyright.
RESEARCH ARTICLE
Intelligent-Routing Algorithm for wireless body area
networks
Smita Sharma
1
| V M Mishra
2
| M M Tripathi
3
1
Department of Electronics and
Communication Engineering,
Uttarakhand Technical University,
Dehradun, India
2
Department of Electrical Engineering,
G. B. Pant Engineering College, Pauri
Garhwal, India
3
Department of Electrical Engineering,
Delhi Technological University, Delhi,
India
Correspondence
Smita Sharma, Department of Electronics
and Communication Engineering,
Uttrakhand Technical University,
Dehradun, India.
Email: smitapandey86@gmail.com
Summary
The progressive strides in the exploration of sensing technology and the poten-
tial use of electrical devices have rendered a promising technology, termed as
Wireless Body Area Networks (WBANs). This technology has rendered compe-
tency to reform healthcare and making available ubiquitous health care sup-
port to remotely located patients. However, the sensor nodes which are
embedded on or underneath the body of the patients to measure multifaceted
physiological parameters are suffered from the limited battery constraints. In
this paper, to address this issue, we propose Intelligent-Routing Algorithm for
WBANs (I-RAW) to elongate life period of these sensor nodes. The two sinks
are located on the front and back side of the patient's body that collects data
from the sensor nodes which form clusters and forward the data through the
cluster head (CH). We apply Tunicate Swarm Algorithm (TSA) for selection of
CH which considers the essential parameters namely, residual energy, net-
work's average energy, distance of node from the sink, path loss model, and
energy consumption rate. The use of two sinks in WBAN mitigate hot-spot
problem in the network by avoiding the multi-hop communication. The
simulation results show that I-RAW improves stability period and network
operational period by 37.7% and 42.7%, respectively, as compared to Dual Sink
approach using Clustering in Body area network (DSCB) protocol, respectively.
Further, I-RAW also shows supreme performance for various performance
metrics as compared to other state-of-the-art protocols.
KEYWORDS
cluster head (CH), Intelligent-Routing Algorithm (I-RAW), Network Lifetime, Tunicate
Swarm Optimization (TSA), Wireless Body Area Networks (WBANs)
1|INTRODUCTION
The proliferation in ubiquitous communications and the ever growing low-power-based wireless technologies have
given the promising scope for the development of wireless network operating around humans. The design of intelligent
bio-medical sensors operating at radio-frequency has made it possible to monitor remotely located patients. A Wireless
Body Area Network (WBAN) is a network of such bio-medical sensors leveraging the emerging IEEE 802.15.6 and IEEE
802.15.4j standards which are particularly meant for its operation.
1
The transmission range of these sensor nodes is
Received: 18 December 2020 Revised: 12 July 2021 Accepted: 17 August 2021
DOI: 10.1002/dac.4984
Int J Commun Syst. 2021;e4984. wileyonlinelibrary.com/journal/dac © 2021 John Wiley & Sons Ltd. 1of17
https://doi.org/10.1002/dac.4984
observed to be about 2 m.
1
These nodes are installed on or beneath the human body to sense various physiological attri-
butes of the patient's body.
One of the peculiar concerns of the sensor nodes in WBANs, is the battery limitation of these nodes.
2
Once these
nodes are implanted in the human body or worn by them, the task of removing the batteries or recharging them
becomes a daunting task. Sometimes, while dealing with the patients who are COVID-positive
3
or other communicable
diseases, there is a high need to utilize the employed sensor nodes to elongate their survival period to the maximum
possible duration.
There is the other challenge while dealing with WBANs which includes the data transmission in the fixed slots so
that it does not interfere with the signals which are getting transmitted at the same radio frequency. However, in this
paper, we assume the other factors to be prevailing under ideal conditions. Hence, we predominantly target the energy
limitation of these sensor nodes.
A profusion of research is reported working towards the energy efficiency of WBANs.
4,5
Various researchers adopted
the Time Division Multiple Access (TDMA)-based method to schedule the duty cycle of the sensor nodes with an aim
to reduce the energy consumption of the sensor nodes. However, the routing of data packets is still an important factor
that helps in efficiently utilizing the energy of the sensor nodes. Since the development of WBANs, several routing pro-
tocols have been proposed. One of crucial fact that differentiates the WBANs from traditional WSNs or Ad-hoc network
is the replacement of the planted node on the patient's body will require the surgical operations and prove to be quite
uncomfortable to the patient. Further, due to the continuous or the frequent movement of the patient's body the proto-
cols designed for WSNs are not pertinent to the WBANs.
To avoid the multiple transmissions from the sensors to the sink, the cluster-based topology is adopted. In every
cluster, the cluster head (CH) plays a significant role in gathering data from the cluster members.
6
The selection of CH
is done based on various essential parameters, namely, residual energy, distance, and so on. It is said to be a Non-
Polynomial (NP)-Hard problem.
7
The role of meta-heuristic approach in the CH selection has been appealing to render
optimal performance of the network.
8
When there is a inter-cluster communication, the nodes which are located
nearest to the sink, experience a great amount of energy consumption. Momentarily, these relaying nodes exhaust their
energy completely and this problem is termed as hot-spot problem. To curb this problem, we employ dual sink for the
purpose of data collection.
Various nature-inspired optimization methods, namely, Particle Swarm Optimization (PSO),
9,10
Ant Colony Optimi-
zation (ACO),
11
Genetic Algorithm (GA),
12
and so on. However, it is observed that most of these methods either suffer
from the late convergence, or with less exploration capabilities. Therefore, there is a need for the CH selection with the
technique which is tested on the benchmark functions for the improved convergence and exploration capabilities.
Therefore, we consider recently developed Tunicate Swarm Algorithm (TSA)
13
to optimize the selection of CH as it
fulfills the basic requirements as discussed above.
1.1 |Research contributions
It is imperative to address the concern of limited energy of sensor nodes in the WBAN because it is infeasible to charge
these nodes continuously. The major contributions of our work are listed as follows.
1. We propose Intelligent-Routing Algorithm for WBANs (I-RAW) that works on a cluster-based topology pertaining
to achieve enhanced network lifetime. The selection of CH is done based on the TSA
13
method.
2. We employ two sinks, on each side of the patient's body, that is, on the chest and on the back. While doing so, we
mitigate the hot-spot problem by proposing single hop communication on both sides. As shown in Figure 1, one side
of patient's body is shown. Similar structure of WBAN is installed on the back side of the patient's body.
3. The selection of CH is done based on the residual energy, distance of a node from the sink, and average energy of
the network, energy consumption rate (ECR) and path loss factor. To the best of our knowledge, it is the first time
when TSA algorithm is applied for routing in WBANs.
4. The performance comparison of I-RAW is done against the state-of-the-art algorithms that have targeted the cluster-
based routing techniques.
Novelty analysis: It is learned from the retrospective survey, various evolutionary algorithms are employed to
address the routing issues in WBAN. However, the proposed work is novel in the following aspect. First, we employ
2of17 SHARMA ET AL.
most recently proposed TSA optimization method that has been evaluated based on different test functions as studied
from.
13
It is the first ever work that uses TSA optimization technique in WBAN for addressing the energy efficiency.
Secondly, the aspect of CH selection uses five essential novel parameters which are optimized through TSA. Hence, the
proposed work is claimed to be novel in the perspective of utilizing the TSA for CH selection pertaining to WBAN
application.
The remaining part of the paper is structured as follows. Section 2 highlights the related work that addresses the
sensor's energy related issues in WBANs. Section 3 presents the background of TSA algorithm then I-RAW is explained
in detail. Section 4 presents the results and discussion, wherein simulation analysis of proposed technique is discussed.
Lastly, the conclusion is given in Section 5.
2|RELATED WORK
The crucial role of WBANs is discussed comprehensively in the survey papers mentioned as follows:
14-20
These survey
papers targeted various attributes related to WBANs that include security issues, architecture, energy efficiency issues,
and many more.
In this paper, we focus the energy issues related to the sensor nodes, therefore we have targeted papers related to
routing in WBANs. The researchers have focused on the cluster-based routing mechanism in which the selection of CH
has been of crucial concern.
21-23
Ullah et al
4
made use of the dual sink in clustering for body area network (DSCB). For critical data transfer the pro-
posed protocol used single hop communication, whereas for normal data transfer adapted the multi-hop communica-
tion leading to hot-spot problem. Hence, the forwarder nodes consume their energy gigantically. Khan et al
24
employed
forwarder node for the transmission of normal data in the network whereas the critical data is forwarded directly to the
sink. The authors selected forwarder node based on the distance and the residual energy of the sensor nodes. However,
this technique suffers from heavy shortcomings in terms of the selection of forwarder node, as there is no provision to
select the node which is closer to the other nodes in the cluster. Furthermore, the direct transmission from the nodes
which are sending critical data will not last long due to heavy energy transmission. Ullah et al
5
in other work proposed
the concept of harvested-aware cluster-based routing in WBAN. The authors used various parameters for the selection
of forwarder node by considering the Signal-to-noise ratio (SNR), energy (sum of harvester and residual energy), dis-
tance, required transmission power. However, it is observed that the proposed work still suffers from the hot-spot prob-
lem due to multi-hop communication. Roopali and Kumar
25
proposed a routing strategy that uses GA to determine the
cost function. The selection of CH is done based on the energy and sensing radius. The routing path selection is done
FIGURE 1 Proposed demonstration of Intelligent-Routing Algorithm for wireless body area networks (I-RAW)
SHARMA ET AL.3of17
with the help of cost function. The routing technique requires the Cartesian coordinates of the CH node which adds
cost to the network. Raj and Chinnadurai
26
proposed a load balancing approach to mitigate the delay involved in data
aggregation at the smart wearable patches. The proposed work helps in energy conservation by selecting the neighbor
based on the data handling and residual energy. The authors used three transmission modes: (1) The energy efficient
mode; (2) switch-over mode; (3) data dissemination mode in a balanced way. Although the proposed work by the
authors give optimal performance, but it suffers heavily from the overhead and hot-spot problem due to multi-hop com-
munication. The proposed algorithm is quite complex in nature which limits its pertinence in real time.
The above discussed protocols suffered from the hot-spot problem due to the multi-hop communication among the
sensors nodes which are located on the body of the patients. Furthermore, it is observed that routing mechanism is still
to be applied efficiently in case of WBANs that could enhance the network lifetime without causing any added delay
and any financial burden on the network. Other than routing mechanism targeted by the researchers targeted interfer-
ence and congestion avoidance in WBANs. Xie et al
27
focused on the interference as they propose Nest-based WBAN
Scheduling (NBWS). The authors employ graph coding theory to fix the scheduling of the sensor nodes with the help of
Time Division for Multimodal Sensor (TDMS) group scheduling model. There is still a great scope for reducing the
energy consumption of the sensor nodes and hence, enhancing the network longevity. Amjad et al
28
proposed a strategy
that maximizes the energy efficiency for the critical nodes. The authors used numerical methods for solving the linear
fractional problem which is devised from the linear fractional problem. The authors further devised the upper and
lower bounds with least computational complexity. Ahhmed et al
29
emphasized on the role of Software-Defined Net-
work (SDN) in controlling the management processes of WBANs. The authors proposed congestion control-based rou-
ting algorithm which is temperature aware. The striking feature of the proposed work was to limit the adverse effect of
temperature on the routing of data packets. The authors used Enhanced Multi-objective Spider Monkey Optimization
(EMSMO) for SDN-based WBAN. The routing mechanism proposed by the authors lacks the presence of some crucial
parameters, namely, ECR, node density, and so on. The takeaways from the above discussed literature survey are as
follows:
1. The research gap lies in the energy efficient CH selection for WBAN that could mitigate the hot-spot problem in the
network effectively.
2. Most of the discussed algorithms have used conventional methods of forwarder node selection, or relay selection.
The use of efficient meta-heuristic approaches will help in the optimized CH selection. The TSA
13
is still to be used
for WBANs. The various steps of the TSA algorithm are explained in detailed in Algorithm 1.
3|OPERATION OF INTELLIGENT-ROUTING ALGORITHM FOR
WIRELESS BODY AREA NETWORKS
In this section, we highlight the network assumptions and then we propose the operational structure of proposed work.
3.1 |Network assumptions for Intelligent-Routing Algorithm for wireless body area
networks
It is important to understand that while simulating the proposed work, there are some network assumptions pertinent
to WBANs which must be highlighted.
1. We assume the ideal conditions in the context of repercussion caused due to use of sensor nodes on both sides of
body and we neglect it for our proposed work. We add description regarding this as follows:
The sensor nodes are located on the each side of the human body, that is, chest side and back side. Two sinks are
installed one on each side of the patient's body. The use of various sensor devices on both sides of body, may faces
various challenges. It may cause several problems in data collection as well. Since, there are different ways to use
the sensors in WBAN applications, for example, implant node that is planted underneath the sink or within the
body tissue; body surface node which is either placed on the surface or 2 cm away from the human body.
Although, our proposed scheme uses body surface node which is placed on the surface of the human body, it will
still face various challenges. On the other side, it is required for these nodes to be in contact with the human body
4of17 SHARMA ET AL.
to acquire the precise reading of the various physiological factors, for example, blood pressure, temperature, and
pulse rate.
2. Since IEEE 802.15.6 standard was established to help meet the requirements of WBANs in the context of Quality of
Service (QoS), low power, low data rate, and also non-interference.
30
We assume our proposed work to consider
aforesaid standard for communication. However, the primary concern of our proposed work is in saving the energy
of the sensor nodes that use IEEE 802.15.6 standard. Hence, during the realization of proposed technique, this stan-
dard will be used with the routing technique proposed in this paper.
3. The nodes are assumed to be static, though patient may be moving, but the whole WBAN is static as it is fixed on
the patient's body.
4. The sensor nodes are energy limited and it is assumed that there is not charging source for them. However the sink
is having no energy constraint and is given a continuous source of power.
5. The other factors namely, radio interference, signal attenuation, signal splitting, and damage of sensor nodes due to
various reasons, are not considered.
6. The nodes are assumed to be homogeneous i.e., they are having similar configuration in terms of energy, computa-
tion, sensing range, etc.
7. The whole routing mechanism in this work, is assumed to be secured one. Further, we do not take the congestion of
radio signals into account.
SHARMA ET AL.5of17
3.2 |Operational steps of Intelligent-Routing Algorithm for wireless body area
networks
It is noted that I-RAW works by following the clustering fundamentals. The network formation and CH selection are
two phases in which I-RAW operates. Further, we highlight the reason for using dual sink and also the significance of
CH selection in WBAN.
3.2.1 | Why dual sink?
We follow dual sink approach by installing each on the front and back of the patient's body. The installation of dual
sink is motivated from the work,
410
as it helps in avoiding the sole dependency on the single sink. Further, to avoid con-
gestion around single sink, the availability of the second sink helps in the data collection in a single hop communica-
tion. One of the most important reasons is the avoidance of multi-hop or chain based communication that gives rise to
hot-spot problem in the WBANs. Hence, the use of dual sink tend to serve the purpose of dealing with hot-spot
problem.
3.2.2 | Why clustering? And what is the significance of cluster head selection in wireless body
area networks?
We apply clustering approach to reduce the individual data transmission from the sensor nodes to the sink. Further,
the selection of CH is of prime importance which acts as a forwarder node for the sensed data to the sink. We apply
TSA optimization method to select CH and the conventional steps followed are discussed as below.
I-RAW operates in two phases: set-up phase and steady state phase. The first phase deals with the nodes and sink
installation on the body and second phase is linked with the way data is forwarded from the sensor nodes to the CH
and from the CH to the sink.
3.2.3 | Set-up phase of Intelligent-Routing Algorithm for wireless body area networks
The deployment of homogeneous nodes is done on the body of the patient along with the two sinks which are installed
on the abdomen and on the back side of the patient. Once, the installation of nodes is done, the following steps are
followed.
1. Sink located on the each side of the patient, sends Hello message to the entire network of nodes in its corresponding
side.
2. The nodes acknowledge the Hello message by sending the unique id in which the energy information, and location
information is encapsulated in that unique id.
3. When each sink receives the unique ids of all the nodes on its corresponding side, it broadcasts collected ids to the
network. As a result every node gets aware of the energy status of its neighboring node.
These steps are followed to adapt distributed mechanism of selecting the CH node. We apply TSA algorithm and
specifically, it is the fitness function that decides which node is suitable to become CH. The fitness function decides the
fitness of each individual which is the node here, that gives the optimum solution in terms of preserving the energy of
the nodes. The crucial point to note here is, the involvement of fitness parameters. It is important to include those
essential fitness parameters that eventually decide the selection of CH. Further, we explain in detail the role of these
parameters and how do they leave a major impact on the overall performance of the network. The whole operation of
I-RAW is demonstrated in Algorithm 2. The various symbols used in the proposed I-RAW is given in Table 1.
6of17 SHARMA ET AL.
1. Fitness parameters for TSA: We employ four essential fitness parameters that helps in the selection of CH.
(1) Residual energy/ Current energy value: The very basic and fundamental parameter, that is, current energy
value of the node. The CH consumes the energy at the higher level in comparison to the other nodes. Therefore,
a node must be selected with the higher value of energy. Furthermore, though all nodes have similar energy in
the start of the network, but gradually these nodes deplete energy differently with respect to the distance of
these nodes from the sink. Hence, the residual energy is considered for CH selection. Equation (1) determines
the residual energy of the sensor node and it is summed up to compute the residual energy of all nodes.
F1¼Pn
i¼1
EresðiÞ
EinðiÞð1Þ
We compute the ratio to see how much the energy of the sensor node has decreased by far. Hence, it is observed that
more value of fitness parameter F
1
, more is the chance for a node to be selected as CH.
(2) Distance between the sink and sensor node: According to Rappaport et al,
31
the wireless communication is sub-
jected to the distance factor between two communicating entities. Therefore, while selecting the CH, the closest
node to the sink is given preference. Hence, it is imperative to include this parameter to conserve the energy of
the CH node. Equations (2) and (3) compute the distance ratio and average distance of the nodes from the sink,
respectively. It is imperative to note that higher value of F
2
supports the node's selection as CH.
SHARMA ET AL.7of17
F2¼Pn
i¼1
DavgnsðiÞ
DnsðiÞð2Þ
DavgnsðiÞ¼1
nXn
i¼1DnsðiÞð3Þ
(3) Network's average energy: It is essential to understand the current energy scenario of the WBANs. The net-
work's energy is monitored continuously, if the total energy of the network goes below the threshold value, the
clustering process is changed to the direct communication to the sink. This is done to render continuity in the
data transmission in the network. The third fitness parameter, that is, F
3
, describes the computation of net-
work's average energy given in Equation (4). The higher value of network's energy favors the selection of CH in
the WBANs.
F3¼1
nXn
i¼1EresðiÞð4Þ
(4) ECR: (FP
4th
) (ECR)) While selecting the node as a CH, the ECR is examined for each sensor node. The ECR for
every node is different which is subjected to the varying distance of the node from the sink. Therefore, those
TABLE 1 Symbols definitions
Symbols Meaning
F
!Gravity force
W
!Water flow advection
K
!New search agent position
S
!Social forces between search agents
a
1
,a
2
,a
3
,r
ndm
Random numbers
Total
Itrn
Iteration count
T
min
Initial speed
T
max
Subordinate speed
BN
!Distance between search agent and food source
xCurrent iteration
Tp
!ðxÞTunicate position
E
in
Initial energy of node
E
res
Residual energy of node
E
THD
(i) Threshold energy of i_th node
nTotal number of nodes in the network
D
nsink(i)
Distance of i_th node from sink
Davg
nsink
Average distance of all nodes from sink
Dist
Ratio
Ratio of distance factor
Davg
ij
Distance of i_th node from j_th node
D
nodes
Number of dead nodes in the network
δ1, δ2, δ3, δ4 and δ5 Weighted variables with equal values
E
nodes
Number of expired nodes
O
nodes
Number of operating nodes
Near
DCS
Nearest DCS
tx transmission
R
C
Current round
8of17 SHARMA ET AL.
nodes should be preferred to be CH whose ECR is least among all the nodes. Therefore, fourth fitness parameter
is given by Equation 5
F4¼EinðiÞðEtxðiÞErxðiÞEaggðiÞÞ
EinðiÞ
ð5Þ
In Equation (5), F
4th
denotes the ECR value and it should be maximized to increase the energy preservation or
decreasing the energy consumption of the nodes.
(5) Path loss factor: (F
5th
)) While dealing with bio-medical sensors implanted in the patient's body, the impairments
are arisen from the surrounding or due to the mixing of signals operating at same frequency. The other factors
such as sensors being in No Line of Sight (NLOS), there is decrease in the power density of electromagnetic
wave (em) which helps in the communication of these sensor nodes. When such decrease in the power density
of em occurs due to reflection, refraction, absorption, diffraction and propagation through a wireless medium, it
is referred as path loss.
It primarily depends on the distance of the path and the frequency of the travelling of em.
PathlossðdÞ¼Plossðdist:Þþ10 nlogðD
dist:
Þð6Þ
The calculation of reference distance is done through Equation 7
Plossðdist:Þ¼10 log ð4πdist:fqÞcð7Þ
It is noted that when the mean value of pathloss is different from its value, then the shadowing effect comes into
existence. The cumulative value of pathloss having shadowing factor is given by equation (8).
TpðlossÞ¼Plossðdist:Þþγð8Þ
Hence, the final fitness parameter, that is, F
5th
is given as follows:
F5th ¼1
TpðlossÞ
ð9Þ
It is evident from Equation (9) that the F
5th
should be maximized for the selection of that node as CH which has
least path loss component in WBANs.
2. Fitness function: After computing the fitness parameters, the next task for the TSA algorithm is to compute the linear
combination of the fitness parameters. It is due to the fact that I-RAW is single objective function which consider
various fitness parameters. One of the important point to note here is that the in the fitness function, all parameter's
value are normalized and confined between [0 1]. The fitness function is given by equation
F¼1
δ1F1þδ2F2þδ3F3þδ4F4þδ5F5
ð10Þ
δ1þδ2þδ3þδ4þδ5¼1ð11Þ
Equation (11) denotes the weighted sum of weight factors multiplied with various fitness parameters. These factors
are assigned equal weights to bring load balancing in the WBAN. However, it is up to the user to assign the different
weights to these parameters according to their requirements.
SHARMA ET AL.9of17
3.2.4 | Steady-state phase
Once the CH is selected through the computation of fitness function, the next task is forwarding the data to the CH
from the cluster member nodes. CH forwards the data to the sink from where it is sent to associated mobile or access
point. Lastly, from the access point, the collected data is sent to the healthcare venues.
One important point to note here is the direct communication is involved when the energy of the whole network is
below than the threshold energy which is pre-fixed in I-RAW. It is done to avoid the unnecessary overheads created
while selection of CH.
3.3 |Computational complexity analysis of Intelligent-Routing Algorithm for Wireless
body area networks
It is imperative to compute the computational complexity of I-RAW as it helps in determining the convergence of the
proposed algorithm and also explains the feasibility to implement I-RAW in the real time applications. After combining
the time complexities of both algorithms; algorithm 1 and algorithm 2, the overall complexity becomes O
(Iteration
max
n)+O(Iteration
max
nTotal
Itrn
sdP)=O(Iteration
max
nTotal
Itrn
sdP). It is
noted that the distance computation among the sensor nodes also cause computational complexity, however it is quite
negligible. Therefore, only those computational complexities are considered which has been covered in the various for
and other loops.
3.4 |Radio energy model
As the communication among nodes is processed, the nodes starts consuming their energy according to the radio
energy model as used by.
25,31
This model explains about the energy consumption on the transmission and reception of
data packets at a particular distance. Furthermore, it also accounts to the energy consumed in the data aggregation
process.
4|RESULTS AND DISCUSSIONS
The simulations of I-RAW is discussed in MATLAB Software. The investigation of I-RAW is done based on various met-
rics namely, stability period, network lifetime, network energy consumption analysis, and packet delivery to the sink.
The simulation parameters that are used while implementing I-RAW is given in Table 2. The performance comparison
of I-RAW is done against the DSCB,
4
OE2-LB,
26
and MS-GAOC.
10
TABLE 2 Simulation parameters
Parameters Values
Position of data collecting sink (sink) (50,50),
Total bio-medical sensor 50 (C1), 25 (C2) on each side
Initial energy of bio-medical sensor 0.5 (C1) and
WBAN's total energy 50 J (C1) and 5 J (C2)
Size of data packet 2000 bits
TSA parameters Values
Parameter T
min
1
Search agents 80
Parameter T
max
4
Number of generations 1000
10 of 17 SHARMA ET AL.
4.1 |Performance metrics
WBANs have various metrics which decide about its performance. Here in this work, the following metrics are consid-
ered that examine the performance of I-RAW at various metrics. It is crucial to define the round, as it is being used in
the performance metrics. A round is said to be one, when the data from the nodes is forwarded to the sink for one time.
Stability period: While dealing with WBANs, it is important that the reliability is maintained. This could be possible
only when all the sensor nodes are operational, as it ensures the transfer of complete information about the physiologi-
cal measurements of patient's body. The moment at which the first sensor node gets expired, or the number of rounds
covered at that point, is termed as stability period. It is evident from the Figure 2 that for C1, I-RAW reports the stability
period equal to 2731 rounds, whereas the protocols DSCB, OE2-LB and MS-GAOC accounts to the 1983, 1403, and 1130
rounds, respectively. Similarly for C2, the I-RAW reports enhanced performance for stability period as given in
Figure 3. The reason for the improvement of the stability period in I-RAW is the TSA-based CH selection. The crucial
parameters considered enhance the load balancing in the network.
FIGURE 2 Performance analysis of Intelligent-Routing Algorithm for wireless body area networks (I-RAW); C1
FIGURE 3 Performance analysis of Intelligent-Routing Algorithm for wireless body area networks (I-RAW); C2
SHARMA ET AL.11 of 17
Network operational period: Once the nodes are located over the patient's body, the data transmission is processed
and the nodes start consuming their energies. When all nodes in WBAN becomes non-operation, then the number of
rounds covered at that time is termed as network operational period or network lifetime. It is evident from the Figure 4
that for C1; I-RAW delivers network survival period to be equal to 5435 rounds whereas the protocols DSCB, OE2-LB,
and MS-GAOC accounts to the 3808, 2728, 2200 rounds, respectively. Similar enhancement is observed for C2 by I-
RAW as shown in Figure 5. The dead nodes versus rounds is also observed for I-RAW for both cases; C1 and C2 as
shown in Figures 6 and 7. The reason for this improvement is the use of parameters like network's average energy
which brings energy balancing in the whole network.
Network's remaining energy: To observe the status of the nodes which are consuming the energy gradually, this
parameter helps to the great level. The total energy of I-RAW is equal to 13 Joules and as the other protocols are also
simulated in the same simulation environment, therefore the other protocols have same stock of energy. It is observed
in Figures 8 and 9 for C1 and C2; the protocol I-RAW cover more number of rounds for the particular value of energy.
FIGURE 4 Alive nodes versus rounds comparison of Intelligent-Routing Algorithm for wireless body area networks (I-RAW); C1
FIGURE 5 Alive nodes versus rounds comparison of Intelligent-Routing Algorithm for wireless body area networks (I-RAW); C2
12 of 17 SHARMA ET AL.
Hence, the protocol I-RAW performs better than the other protocols. The reason for this improvement is the energy effi-
cient CH selection and removal of hot-spot problem.
Number of packets delivered: While dealing with WBANs, it is important to deliver the critical information to the
healthcare venue without suffering from any loss of information. Hence, it becomes important to deliver data packets
to the maximum level. As given in Figure 10 for C1, the I-RAW delivers 112951 packets, whereas the packets delivered
by DSCB, OE2-LB and MS-GAOC accounts to the 72726, 56865, and 47395 packets, respectively. Similar proliferation is
observed for I-RAW for C2 as given in Figure 11. The reason for the improvement in the throughput is the elongation
in the life period of sensor nodes which delivers more packets in I-RAW as compared to other protocols mentioned
above.
FIGURE 6 Dead nodes versus rounds comparison of Intelligent-Routing Algorithm for wireless body area networks (I-RAW); C1
FIGURE 7 Dead nodes versus rounds comparison of Intelligent-Routing Algorithm for wireless body area networks (I-RAW); C2
SHARMA ET AL.13 of 17
4.2 |Cost effective analysis of Intelligent-Routing Algorithm for wireless body area
networks
The installation of sensor nodes on body of patients and hence, the communication of data carrying information about
the physiological attributes, faces various challenges. The overall expenditure gets very high when it comes to realiza-
tion of the WBAN. Therefore, to address this concern, we have given cost-effective analysis of proposed I-RAW by con-
sidering the two cases; C1 and C2. In first case, where we evaluate I-RAW for 100 nodes in total, in second case C2, we
only consider 50 nodes. More importantly, where in C1 we use nodes of 0.5 J of energy, in C2, we only use nodes with
0.1 J of energy. The crucial finding while performing this experiment for two cases is, the cost-effective characteristic of
proposed work while still maintaining the optimized network performance.
Overall empirical analysis of the proposed work for two cases: C1 and C2 are given in Table 3.
FIGURE 8 Network's remaining energy of Intelligent-Routing Algorithm for wireless body area networks (I-RAW) against others; C1
FIGURE 9 Network's remaining energy of Intelligent-Routing Algorithm for wireless body area networks (I-RAW) against others; C2
14 of 17 SHARMA ET AL.
FIGURE 10 Throughput analysis of Intelligent-Routing Algorithm for wireless body area networks (I-RAW); C1
FIGURE 11 Throughput analysis of Intelligent-Routing Algorithm for wireless body area networks (I-RAW); C2
TABLE 3 Empirical evaluation of I-RAW
Name of performance metrics
Protocols
MS-GAOC OE2-LB DSCB I-RAW
Stability period (C1) 1130 1403 1983 2731
C2 294 336 387 539
Half network dead (C1) 1822 2288 3100 4468
C2 434 550 587 749
Survival period (C1) 2200 2728 3808 5435
C2 572 688 743 992
Number of packets sent (C1) 47,395 56,865 72,726 112,951
C2 2380 5126 3669 5630
SHARMA ET AL.15 of 17
5|CONCLUSION
WBANs have brought the revolution in making use of growing technology in handling the patient's healthcare who are
located remotely. However, the use of energy limited sensor nodes impose a huge challenge to utilize the potential of
WBANs. In this paper, we have proposed intelligent routing mechanism that employs two sinks in the cluster-based
topology, where the CH is selected based on TSA algorithm. The use of essential parameters for the CH selection has
led to the improved performance of the proposed work in terms of network lifetime, stability period, and packets
delivery. We have also examined I-RAW for cost-effective analysis and we discern from the simulation outcomes that
I-RAW outperforms the other protocols.
However, there are some shortcomings of proposed work that we will address in our future work. The proposed
work is simulation-based; it will be interesting to observe the performance of proposed technique for the real time anal-
ysis. Further, the use of real data set will give more scope to the researchers while using this proposed technique.
DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were created or analyzed in this study.
ORCID
Smita Sharma https://orcid.org/0000-0003-0067-9853
M M Tripathi https://orcid.org/0000-0003-1675-1039
REFERENCES
1. Cao H, Leung V, Chow C, Chan H. Enabling technologies for wireless body area networks: a survey and outlook. IEEE Commun Mag.
2009;47(12):84-93.
2. Verma S, Kaur S, Khan MA, Sehdev PS. Towards green communication in 6g-enabled massive internet of things. IEEE Internet Things J.
2020:8(7):5408-5415. https://ieeexplore.ieee.org/abstract/document/9261453
3. Severe acute respiratory syndrome coronavirus 2 (SARS-cov-2) and corona virus disease-2019 (covid-19): the epidemic and the
challenges.
4. Ullah Z, Ahmed I, Razzaq K, Naseer MK, Ahmed N. Dscb: dual sink approach using clustering in body area network. Peer-to-Peer Netw
Appl. 2019;12(2):357-370.
5. Ullah Z, Ahmed I, Ali T, Ahmad N, Niaz F, Cao Y. Robust and efficient energy harvested-aware routing protocol with clustering
approach in body area networks. IEEE Access. 2019;7:33,906-33,921.
6. Verma S, Sood N, Sharma AK. Cost-effective cluster-based energy efficient routing for green wireless sensor network. Recent Adv
Comput Sci Commun. 2020;12:1-00.
7. Pokhrel SR, Verma S, Garg S, Sharma AK, Choi J. An efficient clustering framework for massive sensor networking in industrial IoT.
IEEE Trans Ind Inform. 2020:1-1.
8. Verma S, Kaur S, Sharma AK, Kathuria A, Piran MJ. Dual sink-based optimized sensing for intelligent transportation systems. IEEE
Sensors Journal. 2020:21(14):15867-15874. https://ieeexplore.ieee.org/abstract/document/9151230
9. Kennedy J, Eberhart R. Particle swarm optimization. In: Proceedings of icnn'95-international conference on neural networks, Vol. 4.
Perth, WA, Australia: IEEE; 1995:1942-1948. https://doi.org/10.1109/ICNN.1995.488968
10. Verma S, Sood N, Sharma AK. Genetic algorithm-based optimized cluster head selection for single and multiple data sinks in
heterogeneous wireless sensor network. Soft Comput. 2019;85:105788.
11. Dorigo M, Birattari M, Stutzle T. Ant colony optimization. IEEE Comput Intell Mag. 2006;1(4):28-39.
12. Harik GR, Lobo FG, Goldberg DE. The compact genetic algorithm. IEEE Trans Evol Comput. 1999;3(4):287-297.
13. Kaur S, Awasthi LK, Sangal AL, Dhiman G. Tunicate swarm algorithm: a new bio-inspired based metaheuristic paradigm for global opti-
mization. Eng Appl Artif Intel. 2020;90:103541.
14. Chen M, Gonzalez S, Vasilakos A, Cao H, Leung VictorCM. Body area networks: a survey. Mobile networks and applications. 2011;16(2):
171-193.
15. Tob
on DP, Falk TH, Maier M. Context awareness in WBANs: a survey on medical and non-medical applications. IEEE Wire Commun.
2013;20(4):30-37.
16. Hayajneh T, Almashaqbeh G, Ullah S, Vasilakos AV. A survey of wireless technologies coexistence in WBAN: analysis and open
research issues. Wirel Netw. 2014;20(8):2165-2199.
17. Movassaghi S, Abolhasan M, Lipman J, Smith D, Jamalipour A. Wireless body area networks: a survey. IEEE Commun Surveys Tutorials.
2014;16(3):1658-1686.
18. Ghamari M, Janko B, Sherratt RS, Harwin W, Piechockic R, Soltanpur C. A survey on wireless body area networks for ehealthcare
systems in residential environments. Sensors. 2016;16(6):831.
16 of 17 SHARMA ET AL.
19. Al-Janabi S, Al-Shourbaji I, Shojafar M, Shamshirband S. Survey of main challenges (security and privacy) in wireless body area
networks for healthcare applications. Egyptian Informatics J. 2017;18(2):113-122.
20. Khan RA, Pathan A-SK. The state-of-the-art wireless body area sensor networks: a survey. Int J Distrib Sensor Netw. 2018;14(4):
1550147718768994.
21. Verma S, Sood N, Sharma AK. Qos provisioning-based routing protocols using multiple data sink in IoT-based WSN. Modern Phys
Letters A. 2019;34(29):1950235.
22. Menon VG, Verma S, Kaur S, Sehdev PS. Internet of things-based optimized routing and big data gathering system for landslide
detection. Big data; 2021.
23. Verma S, Kaur S, Rawat DB, Xi C, Alex LT, Zaman Jhanjhi N. Intelligent framework using IoT-based WSNs for wildfire detection. IEEE
Access. 2021;9:48185-48196.
24. Khan RA, Xin Q, Roshan N. Rk-energy efficient routing protocol for wireless body area sensor networks. Wirel Pers Commun. 2020:116:
709-721. https://link.springer.com/article/10.1007/s11277-020-07734-z
25. Kumar R, et al. Energy efficient dynamic cluster head and routing path selection strategy for WBANs. Wirel Pers Commu. 2020:113:
33-58. https://link.springer.com/article/10.1007/s11277-020-07177-6
26. Raj AS, Chinnadurai M. Energy efficient routing algorithm in wireless body area networks for smart wearable patches. Comput
Commun. 2020;153:85-94.
27. Xie Z, Huang G, Zarei R, Ji Z, Ye H, He J. A novel nest-based scheduling method for mobile wireless body area networks. Digital
Communications and Networks; 2020.
28. Amjad O, Bedeer E, Ikki S. Energy-efficiency maximization of self-sustained wireless body area sensor networks. IEEE Sensors Letters.
2019;3(12):1-4.
29. Ahmed O, Ren F, Hawbani A, Al-Sharabi Y. Energy optimized congestion control-based temperature aware routing algorithm for
software defined wireless body area networks. IEEE Access. 2020;8:41,085-41,099.
30. Saboor A, Ahmad R, Ahmed W, et al. Dynamic slot allocation using non overlapping backoff algorithm in ieee 802.15. 6 WBAN. IEEE
Sensors J. 2020;20(18):10862-10875.
31. Rappaport TS. Wireless communications: principles and practice, Vol. 2. New Jersey: Prentice Hall PTR; 1996.
How to cite this article: Sharma S, Mishra VM, Tripathi MM. Intelligent-Routing Algorithm for wireless body
area networks. Int J Commun Syst. 2021;e4984. doi:10.1002/dac.4984
SHARMA ET AL.17 of 17
A preview of this full-text is provided by Wiley.
Content available from International Journal of Communication Systems
This content is subject to copyright. Terms and conditions apply.
