SPECIAL SECTION ON EMERGING TRENDS OF

ENERGY AND SPECTRUM HARVESTING TECHNOLOGIES

Received January 21, 2021, accepted February 1, 2021, date of publication February 19, 2021, date of current version April 2, 2021.
Digital Object Identifier 10.1109/ACCESS.2021.3060549

Intelligent Framework Using IoT-Based WSNs for

Wildfire Detection
SANDEEP VERMA 1 , (Senior Member, IEEE), SATNAM KAUR 1 , (Student Member, IEEE),
DANDA B. RAWAT 2 , (Senior Member, IEEE), CHEN XI3 ,
LINSS T. ALEX 4 , (Senior Member, IEEE), AND NOOR ZAMAN JHANJHI 5
1 Dr.B.R. Ambedkar National Institute of Technology, Jalandhar 144011, India
2 Department of Electrical Engineering and Computer Science, Howard University, Washington, DC 20059, USA
3 School of Business, Nanjing University, Nanjing 210000, China
4 Department of Electrical and Electronics Engineering, Albertian Institute of Science and Technology (AISAT), Kochi 682022, India
5 School of Computer Science and Engineering, SCE, Taylor’s University, Subang Jaya 47500, Malaysia

Corresponding author: Chen Xi ([email protected])

This work was supported in part by the National Natural Science Foundation of China under Grant 71771118, Grant 72071104,
and Grant 71471083, in part by the Ministry of Education Humanities and Social Sciences Foundation of China under
Grant 18YJCZH146, and in part by the Key Project of Jiangsu Social Science Foundation under Grant 20GLA007.

ABSTRACT IoT-based WSNs have proved their significance in delivering critical information pertaining
to hostile applications such as Wildfire Detection (WD) with the least possible delay. However, the sensor
nodes deployed in such networks suffer from the perturbing concern of limited energy resources, restricting
their potential in the successful detection of wildfire. To extenuate this concern, we propose an intelligent
framework, Sleep scheduling-based Energy Optimized Framework (SEOF), that works in two folds. Firstly,
we propose an energy-efficient Cluster Head (CH) selection employing a recently developed meta-heuristic
method, Tunicate Swarm Algorithm (TSA), that optimizes the five novel fitness parameters by integrating
them into its weighted fitness function. Secondly, we perform a sleep scheduling of closely-located sensor
nodes based on the distance threshold calculated through a set of experiments. Sleep scheduling methodology
plays a pivotal role in abating the number of data transmissions in SEOF. Finally, we simulate SEOF
in MATLAB under different scenarios to examine its efficacy for the various performance metrics and
scalability features. Our empirical results prove that SEOF has ameliorated the network stability period
for two different scenarios of network parameters by 35.3% and 216% vis-à-vis CIRP.

INDEX TERMS Cluster head (CH), energy efficiency, IoT-based WSN, sleep scheduling, tunicate swarm
algorithm (TSA), wildfire detection (WD).

I. INTRODUCTION reports fire statistics of any country, and it also presents the
With the evolution of sensing technology, IoT-based WSNs status of fire alarms for different regions [9]. The Mendocino
have been proliferating in handling multifaceted applications Complex Fire loomed up in Northern California on July 27th,
[1]. IoT-based sensor nodes play a pivotal role in disseminat- 2018 was registered as the largest fire in the history that
ing information from hostile areas where human’s interven- burnt 459,123 acres of land. According to a report by Forest
tion is quixotic [2]–[4]. The sensed information is gathered Survey of India (FSI), occasional fires occur to the 54.40%
from these hostile areas and then forwarded to the sink from of forests in India and 7.49% of forest experience moderately
where it is sent to the user for performing the required course frequent fires [10]. Basically, the 90% forest fires are caused
of action [5]. There is an enormous number of applications by the anthropogenic activities namely, unattended camp-
of IoT-based Wireless Sensor Network (WSN), and among fire, smoking, burning debris, firework, poachers and timber
them, one predominantly prevalent is the Wildfire Detection mafia, etc. The remaining 10% are caused due to the natural
(WD) [6], [7]. Wildfires are a recurrent phenomena around phenomenon namely, lightening, volcanic eruption, and due
the world. Millions of forest hectares are burnt in flames from to meteorites, etc. The forest fires are mostly dependent upon
wildfires every year [8]. The Global Forest Watch (GFW) the type of vegetation or trees grown in the forest areas.
Pine tree forests are highly fire prone as the leaves/needles
The associate editor coordinating the review of this manuscript and of pine act as a fuel for forest fires due to its resin content
approving it for publication was Guangjie Han . [11]. Therefore, many significant attempts are reported to

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME 9, 2021 48185
 S. Verma et al.: Intelligent Framework Using IoT-Based WSNs for WD

replace the pine trees with other suitable trees that grows
broad-leaves [12].
According to the National Fire Danger Rating System
(NFDRS), firefighters must be made aware at a maximum
of six minutes after the fire is started, to curb it, before it
spreads to a large scale [13]. Therefore, it can be concluded
that there is a great significance of early detection of wildfires
to avoid the high magnitude of loss to the property, lives,
flora and fauna. Since the advancements in Micro-Electro-
Mechanical-System (MEMS), the various researchers have
worked to pact with WD through multiple imaging patterns
or the use of sensors. It is evident that the former method
fails in adverse environmental conditions. Therefore, we use
the latter approach for WD while considering the energy
efficiency of the IoT-based sensor nodes. In our proposed
work, four Data Collecting Sinks (DCS) are placed outside
the network to deliver the fire related information with the
least delay. The placement of four DCS avoids the hot-spot
problem and assists in ameliorating the network lifetime [14].
Further, to resolve the issues of WD, energy-efficient
routing is required that must immunizes the network from
the hot-spot problem and also delivers critical information FIGURE 1. The proposed scenario of SEOF.

with least delay. Multitudinous attempts are reported so

far that address the above said concern by optimizing the a meta-heuristic optimization method named TSA [16].
cluster-based routing which incorporates significant param- We consider five essential parameters for selecting CH,
eters for Cluster Head (CH) selection [15]. However, routing which are optimized in the integrated fitness function
strategies that optimize clustering by employing optimization of TSA. These parameters include distance between the
technique with high exploration and fast convergence are DCS and the node, node proximity, network’s average
still left unnoticed. To resolve this concern, we use Tunicate energy, residual energy and time delay.
Swarm Algorithm (TSA) that possesses high exploration and 2) We apply sleep scheduling mechanism in each cluster,
exploitation capabilities, and due to such remarkable features, just before the start of intra-cluster data transmission.
it has high convergence [16]. TSA is tested on different As nodes are randomly deployed, we consider distance
benchmarks, and it is revealed through its statistical outcomes and energy level of adjacent nodes to put them into
that it is more efficient than the competitive algorithms. sleep or active state. If the distance among the adjacent
A detailed study about TSA can be done from study [16]. nodes is lower than the threshold distance, then only
The other crucial approach in preserving energy of one node is made active which transfers the data to CH
IoT-based sensor nodes is sleep scheduling of these nodes. until it sustains.
These nodes when put into sleep state, turn off either 3) We employ four DCS around the network periphery to
the radio (disrupting communication capability) or sensory extenuate the concern of hot-spot problem and early
device (halting the sensing/detection of events) [17]. Sleep data delivery to a sink in large area networks. Con-
scheduling must be examined with prudence and random sequently, it eradicates the aforementioned problem
sleep scheduling is not encouraged due to its repercussion on by introducing single-hop communication between the
network connectivity and topological efficiency. A profusion sensor nodes and the DCS. Figure 1 shows the proposed
of sleep scheduling mechanisms are discussed in detail by the scenario for WD.
various researchers [18], [19]. However, none of the study 4) We perform extensive simulation analysis for two dif-
considered the joint consideration of two crucial aspects; ferent scenarios (Case I and Case II) to examine the
sleep scheduling and energy efficient CH selection. performance and scalalability. These cases have differ-
ent network dimensions including the count of nodes
A. RESEARCH CONTRIBUTIONS deployed and area of network. We use several crucial
To address the concern of WD with energy optimized performance metrics that include stability period, net-
clustering and sleep scheduling, we state our contributions work survival period, network’s remaining energy and
as follow. throughput to evaluate the efficacy of SEOF.
1) We propose an intelligent framework named as This is the first ever work for WD that incorporates mul-
Sleep scheduling-based Energy Optimized Frame- tiple DCS to immune the network from the hot-spot prob-
work (SEOF) pertinent to WD that uses IoT-based sen- lem, and expedite the data delivery through optimized CH
sor nodes. In SEOF, we optimize the CH selection using selection. In addition to this, it is the first instance when the

48186 VOLUME 9, 2021

S. Verma et al.: Intelligent Framework Using IoT-Based WSNs for WD

joint consideration of sleep scheduling and energy efficient presented CH selection by considering the residual energy
clustering pertinent to WD is taken into account. solely. Hence, the proposed approach is energy inefficient
The rest of manuscript is structured as follows. Section II as the ignorance of the distance factor for CH selection
addresses the related work for energy-efficient routing tech- ameliorates the energy consumption. Pokhrel et al. [35] pro-
niques, WD and sleep scheduling mechanisms. Further, jected the Cluster-based Intelligent Routing Protocol (CIRP)
in Section III, we discuss the operation of TSA and SEOF. that used crucial factors for CH selection to eradicate the
Next, in Section IV, we present outcomes and discussions. hot-spot problem. However, in CIRP, the CH selection did
Finally, Section V concludes and highlights directions for not consider the delay factor involved in delivering the data
future work. to the sink. Therefore, it is not suitable for WD applica-
tion. Verma et al. [5] proposed Genetic Algorithm-based
II. RELATED WORK Optimized Clustering (GAOC) by considering three factors
The role of IoT in various real-time applications has namely, node density, energy and distance for choosing
been unparallel to any other technology [20]–[26]. Many CH. Furthermore, the Genetic Algorithm (GA) employed
researchers have targeted Industrial IoT for its efficacy in for CH selection suffered from slow convergence. Further,
network performance [27]–[29] however, a comprehensive Behera et al. [36] proposed i-SEP technique in which the
attention is still required for some hostile applications for choice of CH was based on the threshold value calculated
an example WD. WD has been one of the challenging tasks for every type of heterogeneous node. However, the proposed
to curb the wildfires, which have been the prominent cause technique did not consider the various other essential param-
behind damaging several hectar [8]. Since the advancements eters, namely node density, network average energy, etc.
in Micro-Electro-Mechanical-System (MEMS), the various Sharma et al. [37] proposed Energy-efficient Trusted Moth
researchers have worked to pact with WD through multiple Flame Optimization and Genetic Algorithm (eeTMFOGA)
imaging patterns or sensors. It is evident that the former based clustering algorithm, which selects CH based on
method fails in adverse environmental conditions. There- node density, remaining energy, packet forwarding progress,
fore, we use the latter approach for WD while considering distance, and delay in transmission. However, there is a
the energy efficiency of the 5G-Integrated IoT-based sensor significant problem with eeTMFOGA which suppress its
nodes. Further, we briefly discuss the related work targeting pertinence to hostile applications. The sensor nodes and Base
WD, CH selection and sleep scheduling in sensor network. Station (BS) are moving, which will bring the high challenges
for WD in the context of its real time implementation. Some
A. WD TECHNIQUES of the researchers [38], [39] also exploited the fuzzy sys-
Cao et al. [30] proposed ABi-LSTM for WD which involves tem for CH selection. The authors in [38] proposed Fuzzy
a large amount of image capturing and complexities in the Logic-based Effective Clustering (FLEC) technique but it did
installation over the forest cover. Leal et al. [31] utilized the not consider the delay factor. In [39], CH selection is done
onboard fuzzy logic approach for the detection of fire spots solely based on the RSSI value. Therefore, the scope of these
in Amazon forest of Brazil. However, the system gets com- fuzzy-based algorithms is not pertinent to WD. We discern
promised when the environmental conditions get worsened. from the literature study that the meta-heuristic methods are
Aslan et al. [32] proposed a general WD framework that more promising as they optimize each factor opted for CH
works for developing a technique for strategically deploying selection to acquire the best possible solution [3], [4], [24],
the sensor nodes. The authors used deterministic deployment [27]–[29], [40].
of sensor nodes that adds complexity in real time imple-
mentation. The authors proposed an architecture for WD and 1) WHY TSA FOR CH SELECTION?
designed a clustering protocol, but it is observed that regular TSA possesses high exploration and exploitation capabilities,
or deterministic sensor nodes deployment adds complexity and due to such remarkable features, it has high convergence.
to the network. Jan et al. [33] proposed a Sybil detection TSA is tested on different benchmarks, and it is revealed
approach which employs residual energy and RSSI-based through its statistical outcomes that it is more efficient than
Sybil attack detection strategies. But, it is noted that the the competitive algorithms. A detailed study about TSA can
choice of CH is computationally expensive, and it further be done from study [16].
introduces overheads in the proposed algorithm. As a result,
this problem ultimately leads to gigantic energy consumption. B. SLEEP SCHEDULING METHODS
The wildfires have been so frequent across the globe, but It is learnt from the various studies related to cluster-based
the promising solution has not come yet into the limelight. routing mechanism, if the number of participating nodes can
It is anticipated that the appropriate CH selection exploited be reduced, it can elongate the network lifetime comprehen-
for WD will help in handling this cause efficiently. sively. Therefore to address this concern, many researchers
It is learnt from the retrospective study that the have considered sleep scheduling of sensor nodes [41], [42].
cluster-based routing mechanism plays a pivotal role in WD The paramount reason behind it is the fact that the adjacent
[34]. A plethora of attempts for the optimized CH selection nodes have this propensity to sense similar value of targeted
are reported in the existing literature. Behera et al. [15] attribute. Hence, keeping only one node in active state till

VOLUME 9, 2021 48187

 S. Verma et al.: Intelligent Framework Using IoT-Based WSNs for WD

it has energy, does not harm the network [43], [44]. The B. WORKING STRUCTURE OF SEOF
authors in [45] proposed an Intelligent Sleep scheduled algo- It functions in two stages namely set-up and steady-state
rithm (iSleep) that considered the temperature sensitivity of phase [5]. In the former stage, network formation and CH
the nodes for taking the decision to put it in active or sleep selection are done for the network, whereas the latter involves
state. Due to the multi-hop communication, iSleep suffers the process of data transmission during inter-cluster and
from hot-spot problem. The sleep scheduling has also played intra-cluster communication in the network. It is worth noting
a pivotal role in handling the energy conservation for Wireless that we apply sleep scheduling mechanism in each cluster
Body Area Network (WBAN) [18], [19]. The authors in [44] before data transmission is commenced. In Algorithm 1,
proposed a hybrid approach considering energy harvesting we mention various steps involved in the working process of
and non-harvesting nodes to elongate the network lifetime. SEOF.
Due to the added harvesting resources, the proposed approach
is costly. The authors in [46] proposed Energy aware Schedul- 1) SET-UP PHASE
ing with Quality Guarantee method (ESQG) that aimed to This phase pacts with the network formation which involves
decrease the count of awakening nodes with the process of ’the node deployment and sink placement’ and CH selec-
information fusion. The authors considered the importance tion. We employ four DCS outside the network as illustrated
degree for different surveillance locations. The proposed in Figure 1. Then, we deploy heterogeneous IoT-based sensor
method suffers from high computational overheads. Hence, nodes randomly in the network which is assumed to be a
there is a need of a sleep scheduling mechanism that avoids forest cover of pre-defined dimensions. We use TSA to select
hot-spot problem and also maintain energy balancing in the CH through the mathematical modelling of TSA’s fitness
network. We addresses this concern by a novel method of function which we discuss in detail as follows.
sleep scheduling mechanism in SEOF. Further, we explain its (a) Fitness parameters employed in SEOF for CH selec-
operational framework. tion: We use five essential fitness parameters for the
III. PROPOSED WORK: SEOF selection of CH. We define symbols with their mean-
In this section, the network presumptions and working ings in Table 1 that are used in proceeding sections.
operation of SEOF are discussed. These parameters are discussed as follow:
1) Energy (Residual, initial and threshold value):
A. NETWORK ASSUMPTIONS
The first parameter, i.e., F1 considers residual,
We consider some network assumptions while implementing
initial, and threshold energy of a node for CH
the proposed work.
selection. It is imperative to consider the updated
1) The nature of the network is stationary, i.e., nodes and
energy value of the candidate nodes as nodes con-
DCS remain stationary throughout the network run.
sume energy in commensurate with a gradual pro-
Three heterogeneous nodes namely, normal, advanced
gression of rounds. Equation (1) determines the
and super nodes are used which have least, intermediate
summation ratio of the residual energy of the ith
and maximum level of energy, respectively.
node to its initial value of energy, multiplied with
2) We assume that DCS has no constraints of energy, com-
its threshold energy value. The threshold energy
putation, and network coverage as long as the operation
level of a node defines the required energy value
of SEOF is concerned.
of a candidate node for its selection as CH. In this
3) Further, we assume our network to be a square-shaped
work, the threshold energy is taken as 0.3 Joule
area to use it as a paradigm to perform our simulation.
after performing a number of experiments. It is
4) The deployment of IoT-based sensor nodes is done
noted that higher the value of F1 , the greater
randomly but in a uniform pattern. Each node is
the probability of a node for its election
location-unaware and is given a unique id once it
as CH.
is deployed. These nodes compute distance between Xn Eres (i)
different entities based on Received Signal Strength F1 = × ETHD (i) (1)
i=1 Ein (i)
Indicator (RSSI) value [35].
5) We do not consider the factors of radio interference and 2) Distance among node and DCS: This fitness
any hindrance or signal attenuation due to the presence parameter called the distance factor is the
of physical objects. fundamental factor for any successful wire-
6) We consider the proposed protocol to be ideally less communication between two entities [47].
secured. The security considerations for this work, are The second fitness parameter i.e., F2 computes
out of scope. ratio of distance factor. It is the summation ratio
7) We have not considered homogeneous nodes due to of the distance among ’a candidate node and the
the fact that existence of homogeneity is not feasible nearest DCS’ and an average distance among all
due to different factors that include manufacturing dif- nodes from their nearest DCS. It is noted that a
ferences, and different period of activation once these node should have a low value of distance ratio
nodes are deployed. for its selection as CH to preserve its energy.

48188 VOLUME 9, 2021

S. Verma et al.: Intelligent Framework Using IoT-Based WSNs for WD

Algorithm 1 SEOF Algorithm TABLE 1. Symbol definition.

Input : n = 100, Rmax (Rounds maximum value),

DCS1 = (110, 50), DCS2 = (50, 110),
DCS3 = (−10, 50), DCS4 = (50, −10)
Output: A = CH _N , deadnodes , alivenodes and r
1 Deployment of heterogeneous sensor nodes CH _N = 0
2 Initializing CH nodes
3 for r (round) = 1 to Rmax Then
4 alivenodes = n
5 deadnodes = 0
6 for i = 1 to n Then
7 if Eres (i) == 0 then
8 deadnodes = deadnodes + 1
9 if deadnodes == 0.9 × n then
10 all_dead = RC (current value of
round)
11 end if
12 alivenodes = alivenodes − deadnodes
13 end if
14 for i = 1 to n Then
15 if Eres (i) > 0 then
16 Applying TSA for choosing CH
17 CH _N = CH _N + 1
18 for k = 1 to (CL − 1) Then
19 Compute distance of k th to jth node
20 if Dk−j < Dthd then
21 if Eres (k) > Eres (j) then
22 jth node ← Sleepstate
23 else
24 k th node ← Sleepstate
25 end if CH election, a candidate node should have higher
26 else value of F2 as depicted from equation (4).
27 break Xn Dnd−DCS (i)
DistRatio = (2)
28 end if i=1 Davgnd−DCS
29 end for 1 Xn
30 ith active node ← Assigning TDMA Davgnd−DCS = × Dnd−DCS (i) (3)
n i=1
slot 1
31 CH _N ←− ith active node F2 = (4)
DistRatio
32 NearDCS ←− CH _N
3) Node proximity: The third fitness parameter
33 Update Eres (i) using [47]
i.e, F3 ) assists in choosing a node having greater
34 else
node count around its vicinity as a CH. Conse-
35 break
quently, the distance among the cluster member
36 end if
nodes and a candidate node (to be selected as
37 end for
CH), decreases. Equation (5) determines the aver-
38 if deadnodes == 0.9 × n then
age distance between the adjacent nodes i.e., ith
39 break
and jth node in a cluster. Equation (6) illustrates
40 end if
that a node having lowest average distance from
41 end for
the other nodes in the cluster is selected by F3 .
42 end for
In the other words, a node with higher number of
43 return A
neighboring nodes is selected.
1 XCL−1 XCL−1

Ndpxm = × Davgi−j (i)
CL i=1 j=1
Equation (2) and equation (3) show the compu- (5)
tation of ’distance ratio’ and ’average distance of 1
F3 = (6)
the node from the DCS’. For an energy efficient Ndpxm
VOLUME 9, 2021 48189
 S. Verma et al.: Intelligent Framework Using IoT-Based WSNs for WD

4) Network’s average energy: This parameter, 2) STEADY-STATE PHASE

i.e., F4 considers the network’s average energy, Once, the CHs are selected in the network, the process of
which must be considered while selecting a node sleep scheduling mechanism in each cluster, is applied in the
as a CH. As the data transmission continues, following steps.
the network’s average energy abates which in turn 1) In the first step, we compute the distance of a cluster
abates the count of a number of CHs to be selected member node with the other nodes in that cluster. If the
for a particular round. Equation (7) depicts the distance computed among any two nodes is lower than
network’s average energy. It is evident that higher the threshold distance (we define it through multiple
the value of F4 for a candidate node, more chances simulations as given in Figure 5 (c)), then those two
it will have, to become a CH. nodes are considered further for energy checking.
2) In the second step, we check the current energy value
1 Xn
F4 = × Eres(i) (7) of those two nodes selected from the first step, and the
n i=1
node with lesser energy from the other, enters into sleep
5) Time Delay: The fifth fitness parameter, i.e., F5 , state whereas other stays active until it sustains.
is one of the significant contributions of pre- 3) In the third step, as soon as the active node drains its
sented work due to its direct dependence on energy, the node with sleep state is triggered to active
curbing the wildfire at the earliest. When the state.
network area is huge and the placement of the The whole process of operation of SEOF including the sleep
sink inside the network is not possible due to scheduling mechanism, is explained in the Algorithm 1. The
the hostile environment, the sensors inevitably CH gathers, aggregates and then forwards the information to
follow multi-hop communication. Consequently, the nearest DCS.
the critical fire-alarming information is delivered
C. DESCRIPTION FOR ALGORITHM 1
very late leading to massive damage to the for-
We present the operational framework of SEOF in
est covers. Equation (8) computes the time delay
Algorithm 1. The detail description of this algorithm is given
from the ratio of distance and speed of the data
as follows. We consider the network dimensions as an input
packet which is assumed to be traveling with
to the algorithm. It include number of nodes, maximum
the speed of light i.e., c under ideal conditions.
number of rounds i.e., Rmax for which the algorithm has to
Equation (9) normalizes the time delay to limit
run, and the Cartesian coordinates which define the location
its value in the range [0 1].
of the four DCS placed outside the network. In output of
Dnd−DCS (i) this algorithm, we get the selected CH by applying TSA
Tdelay (i) = (8) optimization method, the dead and alive nodes status with the
c
Tdelay (i) − Tdelay(min)(i) respect to round, the final value of round when the algorithm
TNdelay = (9) is halted. We delineate the whole process covered by SEOF
Tdelay(max)(i) − Tdelay(min)(i)
in Algorithm 1 as follows.
1
F5 = (10) In Line 1, we deploy heterogeneous sensor nodes and
TNdelay placement of DCS is done. In Line 2, we initialize the
CHN variable to store the count of CHs. In Line 3-Line
In equation (10), the value of F5 should be max-
42, we explain the whole process of SEOF with the steady
imized to select the node with the least delay
progression of each round. It covers the for loop that
involved in transmission of data to the DCS.
stops its operation only when the satisfactory condition is
(b) Fitness Function: The fitness function is the linear met. The status of alive and dead nodes is initialized in
combination of the fitness parameters computed above. Line 4-Line 5.The count of alive and dead nodes with
The fitness function, i.e., F, should be minimized to respect to the remaining energy of a node is updated in
attain the optimized selection of the node as CH. Equa- Line 6-Line 14. In Line 9-Line 11, we compare the count
tion (11) gives the value of the fitness function, which of all dead nodes to the 90% of total nodes. In Line 15-Line
is processed further in TSA optimization method. 38, we present the CH selection, sleep scheduling and steady
state phase. In Line 16, we check the remaining energy of a
1
F= node if it is not equal to zero then only it is considered for
λ × F1 + δ × F2 + γ × F3 + β × F4 + σ × F5 CH selection and steady state phase. In Line 17, we apply
(11) TSA [16] optimization method for CH selection. In Line 18,
δ+λ+β +γ +σ =1 (12) the count to CHs is updated. In Line 19-Line 30, we intro-
duce the sleep scheduling concept in a cluster. In Line 20,
Equation (12) represents a weighted sum of we compute the distance between k th and jth node to find if
above-mentioned weight factors associated with the computed distance is lower than the threshold distance
different fitness parameters. (5 meter). In Line 22, we compare the remaining energy of

48190 VOLUME 9, 2021

S. Verma et al.: Intelligent Framework Using IoT-Based WSNs for WD

k th with jth node, and whosoever possesses lower energy than TABLE 2. Simulation parameters.
the other, goes to sleep state. In Line 31, the active node in a
cluster is assigned TDMA slot for data transmission to CH.
In Line 33, CH transmits data to nearest DCS in a single-hop
communication. In Line 34, the energy of a node is updated
using radio energy model [47]. In Line 39-Line 41, the count
to the dead nodes is checked whether it is equal to 90% of total
nodes, then the algorithm stops. Finally, in Line 43, the output
A is obtained.
D. COMPUTATIONAL COMPLEXITY ANALYSIS
We present the computational complexity of SEOF through
the Lemma 1 that we define as follows.
Lemma 1 (:) SEOF terminates in a finite number of iter-
ations i.e., Rmax = O(1) and possesses overall computation
complexity equals to O(Rmax ×N ×CL ×TotalItrn ×s×d ×P).
Proof (:) SEOF has a fixed count of nodes deployed
in the network. Once the steady state phase commences,
the nodes gradually starts decreasing their energies. This
process occurs till the point when 90% of the total nodes
completely exhaust their energies. The energy consumption
of each node occurs according to the radio energy model
[47]. The whole algorithm is made to run for fixed maximum
number of iterations i.e., Rmax which is commensurate with
the 90% dead nodes in the network. Hence, SEOF algorithm
terminates with finite iterations.
We discuss the computational complexity of SEOF as fol-
lows. The first for loop iterates for Rmax number of rounds scalability characteristics, we consider two cases of different
and the dead nodes are checked for every value of n. Hence, network dimensions as given in Table 2. First case considers
the complexity of this loop becomes O(Rmax × N ). Every nth 100 × 100 meter 2 forest cover area deployed with 100 nodes
node is considered for CH selection for every round of Rmax . whereas, second case considers 200 nodes that we deploy in
Also, Line 17 computes the CH selection by applying TSA 500 × 500 meter 2 forest cover area. We empirically evaluate
algorithm which takes O(TotalItrn × s × d × P) time, as deter- the threshold distance for sleep scheduling and found the
mined from Algorithm 1. At the fine grain level, TSA takes optimum values; 5 meter for Case I as depicted from Figure 5
O(s × d) time to initialize the population, O(TotalItr × s × d) (c), and 10 meter for Case II. The performance evaluation of
time to compute the fitness of each agent and O(P) time to SEOF is done against the CIRP [35], FLEC [38], GAOC [5],
model tunicate’s behavior. In (Line 19-Line 30), the steps for and eeTMFOGA [37].
sleep scheduling mechanism are represented. Thus, the com-
mutative complexity of this loop (Line 15-Line 38) becomes B. PERFORMANCE METRICS
O(Rmax × N × CL × TotalItrn × s × d × P). The efficacy evaluation of SEOF is performed based on the
performance measures that are discussed as below.
IV. RESULTS AND DISCUSSIONS
In this section, we firstly discuss the simulation setting 1) Stability Period: It is the count of the rounds before
employed for the proposed work later, we discuss the per- the first node in the network completely exhausts its
formance metrics used in this paper for evaluation of the energy and hence, is said to be dead. We examine
proposed protocol. SEOF for two cases; for Case I, we find through
the simulation analysis that SEOF acquires a stability
A. SIMULATION SETTING period of 2747 rounds. However, the other protocols,
The simulation of SEOF is done in MATLAB Software ver- namely eeTMFOGA, GAOC, FLEC, and CIRP, posses
sion 2016a with the simulation parameters defined in Table 2. the stability period of 621, 859, 1748, and 2029 rounds,
The worth-noting point is that the DCS is placed around respectively as depicted in Figure 2 (a). The amelio-
the network at 10-meter distance from the periphery of each ration in the stability period by SEOF accounts for
side of the square-shaped network. The purpose of doing so, 35.3% and 57.1% vis-a-vis CIRP and FLEC protocols,
is to keep the sink out of the network/forest cover. To have respectively. For Case II, Figure 2 (b) and Figure 3
a comprehensive simulation investigation, the simulation of (c) illustrate that the stability period of SEOF is having
SEOF is performed ten times, and the best results are taken low value due to more number of transmissions in the
with a 95% confidence interval. To examine the SEOF for its network.

VOLUME 9, 2021 48191

 S. Verma et al.: Intelligent Framework Using IoT-Based WSNs for WD

FIGURE 2. Performance analysis of SEOF for Case I; (a) and Case II; (b, c).

FIGURE 3. Alive nodes versus rounds comparison of SEOF for Case I; (a) and Case II; (c) and network remaining energy comparison of
Case I; (b).

FIGURE 4. Dead nodes versus rounds comparison of SEOF for Case I; (a) and Case II; (b) and network remaining energy comparison of
Case II; (c).

The prominent cause behind such improvement in they contribute negligibly and do not transmit any vital
this performance measure is associated with the pro- information. Figure 2 (a), Figure 2 (c), Figure 4 (a) and
posed CH selection and the single-hop communication Figure 4 (b), represent the number of rounds covered
involved due to the placement of multiple numbers until 90% of the nodes are dead. For Case I, Figure 2
of DCS. Further, the sleep scheduling mechanism (a) elucidates that SEOF covers 10509 rounds, whereas
helps in energy preservation of all nodes in the net- CIRP and FLEC cover 9879 and 6923 rounds, respec-
work by switching their roles from active to sleep tively. SEOF accounts for 6% and 51% increase in
state. the network survival period vis-a-vis CIRP and FLEC
2) Network survival period: It is the count of the rounds protocols, respectively. For Case II, Figure 2 (c) shows
till 90% of nodes completely exhaust their energies. the network survival period of SEOF is 11365 which is
The reason for not considering 10% is the fact that 70% more than that of CIRP.

48192 VOLUME 9, 2021

S. Verma et al.: Intelligent Framework Using IoT-Based WSNs for WD

FIGURE 5. Throughput analysis of SEOF for Case I; (a) and Case II; (b) and Threshold distance computation for SEOF (c).

TABLE 3. Percentage amelioration by SEOF against other protocols.

The amelioration in this performance metric is due to participating in the data transmissions over the
the energy-efficient election of CH and the decrease particular count of rounds is more than the other
in the average distance of the CH nodes from the protocols.
sink.
3) Network’s remaining energy: This metric assists in C. SUMMARY
understanding the rate of energy consumption of the To summarize our findings, we determine the percentage
nodes while the network is operational. The network’s amelioration by SEOF against the CIRP [35], FLEC [38],
remaining energy corresponding to the total rounds GAOC [5], and eeTMFOGA [37] in Table 3. We specify
covered is depicted in Figure 3 (b) and 4 (c). The the performance of SEOF for two cases; Case I and Case II
total energy of the network is 70 Joules for Case I considered for investigation for different performance
and 100 Joules for Case II. Further, it is evident from metrics.
Figure 3 (b) and Figure 4 (c) that the value of the V. CONCLUSION AND FUTURE WORK
remaining energy of SEOF is more than the other We propose an intelligent framework namely, SEOF to
protocols during its entire run. address the concerns related to the limited energy resources
The reason behind this improvement is the single-hop of IoT-based sensor nodes that are deployed in the forest
data communication among the CHs and the DCS that covers for Wildfire Detection (WD). In our proposed frame-
eventually minimizes the energy consumption. work, we present a novel approach of joint consideration of
4) Throughput: The count of packets sent to the DCS energy efficient CH selection using a meta-heuristic method
successfully is referred to as throughput. The through- named TSA [16] and sleep scheduling methodology to pre-
put analysis of SEOF against the other protocols serve the energy of sensor nodes. We investigate the per-
for Case I and Case II is depicted in Figure 5 (a) formance of SEOF through a set of experiments based on
and Figure 5 (b), respectively. For Case I, DCS four performance metrics namely, stability period, network
receive 105519 packets in case of SEOF whereas, survival period, network’s remaining energy and throughput.
CIRP, FLEC, GAOC, and eeTMFOGA receive 85209, During empirical investigation of SEOF, we consider two
64032, 28536 and 18763 packets, respectively. For cases of different network dimensions in terms of count of
Case II, the throughput in case of SEOF and CIRP nodes and network area, to ensure the scalability of SEOF.
is 100667 and 40500 packets, respectively. However, The simulation results reveal that SEOF delivers superior
FLEC, GAOC and eeTMFOGA receive 22633, 28767, performance against four state-of-the-art algorithms, namely
and 14021 packets over the rounds. CIRP [35], FLEC [38], GAOC [5], and eeTMFOGA [37].
Due to the enlarged network survival period, the count More specifically, we find that SEOF ameliorates the sta-
of packets transmitted or the count of nodes bility period by 35.3% and 216% for Case I and Case II,

VOLUME 9, 2021 48193

 S. Verma et al.: Intelligent Framework Using IoT-Based WSNs for WD

respectively, vis-‘a-vis CIRP [35] and also outperforms the [17] A. Trotta, M. di Felice, L. Bononi, E. Natalizio, L. Perilli, E. F. Scarselli,
other techniques. The performance improvement of SEOF is T. S. Cinotti, and R. Canegallo, ‘‘BEE-DRONES: Energy-efficient data
collection on wake-up radio-based wireless sensor networks,’’ in Proc.
acquired at the cost of various assumptions for its real time IEEE Conf. Comput. Commun. Workshops (INFOCOM WKSHPS),
implementation. However, when the real time implementa- Apr. 2019, pp. 547–553.
tion of SEOF is performed, there are various factors which [18] Z. Xie, G. Huang, R. Zarei, Z. Ji, H. Ye, and J. He, ‘‘A novel
nest-based scheduling method for mobile wireless body area
may affect the operation of SEOF. The presence of physical networks,’’ Digit. Commun. Netw., vol. 6, no. 4, pp. 514–523,
obstacles between sensor node, the ability to withstand high Nov. 2020.
temperature when the wildfire is in intact, and many others, [19] R. Zhang, A. Nayak, and J. Yu, ‘‘Sleep scheduling in energy harvest-
ing wireless body area networks,’’ IEEE Commun. Mag., vol. 57, no. 2,
can not be ignored ideally. pp. 95–101, Feb. 2019.
In extension of this work, the sink placement can be opti- [20] H. Rahman, N. Ahmed, and M. I. Hussain, ‘‘A hybrid data aggregation
mized to achieve better network performance. Further, it will scheme for Internet of Things (IoT),’’ in Proc. IEEE Annu. India Conf.
(INDICON), Dec. 2016, pp. 1–6.
be interesting to observe the performance of SEOF under the
[21] H. Rahman, N. Ahmed, and M. I. Hussain, ‘‘A QoS-aware hybrid data
sink mobility scenario to reduce the cost added in the network aggregation scheme for Internet of Things,’’ Ann. Telecommun., vol. 73,
with the addition of four DCS. nos. 7–8, pp. 475–486, Aug. 2018.
[22] N. Ahmed, D. De, and I. Hussain, ‘‘Internet of Things
(IoT) for smart precision agriculture and farming in rural
REFERENCES areas,’’ IEEE Internet Things J., vol. 5, no. 6, pp. 4890–4899,
[1] P. Zhou, C. Wang, and Y. Yang, ‘‘Self-sustainable sensor networks Dec. 2018.
with multi-source energy harvesting and wireless charging,’’ in [23] H. Rahman, N. Ahmed, and M. I. Hussain, ‘‘A hybrid data aggrega-
Proc. IEEE Conf. Comput. Commun. (INFOCOM), Apr. 2019, tion scheme for provisioning quality of service (QoS) in Internet of
pp. 1828–1836. Things (IoT),’’ in Proc. Cloudification Internet Things (CIoT), Nov. 2016,
[2] G. Kunzel, L. S. Indrusiak, and C. E. Pereira, ‘‘Latency and life- pp. 1–5.
time enhancements in industrial wireless sensor networks: A Q-learning [24] G. Yang, M. A. Jan, V. G. Menon, P. G. Shynu, M. M. Aimal, and
approach for graph routing,’’ IEEE Trans. Ind. Informat., vol. 16, no. 8, M. D. Alshehri, ‘‘A centralized cluster-based hierarchical approach for
pp. 5617–5625, Aug. 2020. green communication in a smart healthcare system,’’ IEEE Access, vol. 8,
[3] S. Jacob, V. G. Menon, P. G. Shynu, S. K. S. Fathima, B. Mahapatra, pp. 101464–101475, 2020.
and S. Joseph, ‘‘Bidirectional multi-tier cognitive swarm drone 5G net- [25] V. G. Menon, S. Jacob, S. Joseph, P. Sehdev, M. R. Khosravi, and
work,’’ in Proc. IEEE Conf. Comput. Commun. Workshops (INFOCOM F. Al-Turjman, ‘‘An IoT-enabled intelligent automobile system for smart
WKSHPS), Jul. 2020, pp. 1219–1224. cities,’’ Internet Things, Jun. 2020, Art. no. 100213.
[4] G. Rathee, N. Jaglan, R. Iqbal, S. P. Lal, and V. G. Menon, ‘‘A trust analysis [26] H. Zhang, M. Babar, M. U. Tariq, M. A. Jan, V. G. Menon, and X. Li,
scheme for vehicular networks within IoT-oriented green city,’’ Environ. ‘‘SafeCity: Toward safe and secured data management design for IoT-
Technol. Innov., vol. 20, Nov. 2020, Art. no. 101144. enabled smart city planning,’’ IEEE Access, vol. 8, pp. 145256–145267,
[5] S. Verma, N. Sood, and A. K. Sharma, ‘‘Genetic algorithm-based opti- 2020.
mized cluster head selection for single and multiple data sinks in hetero-
[27] L. Liu, G. Han, Z. Xu, J. Jiang, L. Shu, and M. Martinez-Garcia,
geneous wireless sensor network,’’ Appl. Soft Comput., vol. 85, Dec. 2019,
‘‘Boundary tracking of continuous objects based on binary tree struc-
Art. no. 105788.
tured SVM for industrial wireless sensor networks,’’ IEEE Trans.
[6] M. Hefeeda and M. Bagheri, ‘‘Forest fire modeling and early detection
Mobile Comput., early access, Aug. 25, 2020, doi: 10.1109/TMC.
using wireless sensor networks,’’ Ad Hoc Sensor Wireless Netw., vol. 7,
2020.3019393.
nos. 3–4, pp. 169–224, Apr. 2009.
[28] L. Liu, G. Han, Z. Xu, L. Shu, M. Martinez-Garcia, and B. Peng,
[7] F. Cui, ‘‘Deployment and integration of smart sensors with IoT devices
‘‘Predictive boundary tracking based on motion behavior learning
detecting fire disasters in huge forest environment,’’ Comput. Commun.,
for continuous objects in industrial wireless sensor networks,’’ IEEE
vol. 150, pp. 818–827, Jan. 2020.
Trans. Mobile Comput., early access, Jan. 5, 2021, doi: 10.1109/TMC.
[8] Y. Meng, Y. Deng, and P. Shi, ‘‘Mapping forest wildfire risk of the
2021.3049220.
world,’’ in World Atlas of Natural Disaster Risk. Springer, 2015,
[29] Z. Xu, G. Han, H. Zhu, L. Liu, and M. Guizani, ‘‘Adaptive DE algorithm
pp. 261–275. [Online]. Available: https://link.springer.com/chapter/
for novel energy control framework based on edge computing in IIoT
10.1007/978-3-662-45430-5_14
applications,’’ IEEE Trans. Ind. Informat., early access, Jul. 7, 2020, doi:
[9] Global Forest Watch. Accessed: Jun. 7, 2020. [Online].
10.1109/TII.2020.3007644.
Available: https://fires.globalforestwatch.org/home/
[10] Welcome to Forest Survey of India. Accessed: Jun. 7, 2020. [Online]. Avail- [30] Y. Cao, F. Yang, Q. Tang, and X. Lu, ‘‘An attention enhanced bidirectional
able: https://fsi.nic.in/forest-fire-activities?pgID=forest-fire-activities LSTM for early forest fire smoke recognition,’’ IEEE Access, vol. 7,
[11] R. W. Mutch, ‘‘Wildland fires and ecosystems—A hypothesis,’’ Ecology, pp. 154732–154742, 2019.
vol. 51, no. 6, pp. 1046–1051, Nov. 1970. [31] B. E. Z. Leal, A. R. Hirakawa, and T. D. Pereira, ‘‘Onboard fuzzy
[12] B. I. Yoo, ‘‘Forest fire management and autonomy in the republic logic approach to active fire detection in Brazilian amazon forest,’’
of Korea,’’ in Lessons From Forest Decentralization: Money, Justice IEEE Trans. Aerosp. Electron. Syst., vol. 52, no. 2, pp. 883–890,
Quest for Good Governance Asia–Pacific. London, U.K.: Earthscan, Apr. 2016.
2008. [32] Y. E. Aslan, I. Korpeoglu, and Ö. Ulusoy, ‘‘A framework for use
[13] V. Chowdary, M. K. Gupta, and R. Singh, ‘‘A review on forest fire of wireless sensor networks in forest fire detection and monitor-
detection techniques: A decadal perspective,’’ Networks, vol. 4, p. 12, ing,’’ Comput., Environ. Urban Syst., vol. 36, no. 6, pp. 614–625,
2018. Nov. 2012.
[14] V. Valls, G. Iosifidis, and T. Salonidis, ‘‘Maximum lifetime analytics in [33] M. A. Jan, P. Nanda, X. He, and R. P. Liu, ‘‘A Sybil attack
IoT networks,’’ in Proc. IEEE Conf. Comput. Commun. (INFOCOM), detection scheme for a centralized clustering-based hierarchical net-
Apr. 2019, pp. 1369–1377. work,’’ in Proc. IEEE Trustcom/BigDataSE/ISPA, vol. 1, Aug. 2015,
[15] T. M. Behera, S. K. Mohapatra, U. C. Samal, M. S. Khan, M. Daneshmand, pp. 318–325.
and A. H. Gandomi, ‘‘Residual energy-based cluster-head selection in [34] S. Verma, N. Sood, and A. K. Sharma, ‘‘Cost-effective cluster-based energy
WSNs for IoT application,’’ IEEE Internet Things J., vol. 6, no. 3, efficient routing for green wireless sensor network,’’ Recent Adv. Comput.
pp. 5132–5139, Jun. 2019. Sci. Commun., vol. 13, p. 1, Mar. 2020.
[16] S. Kaur, L. K. Awasthi, A. L. Sangal, and G. Dhiman, ‘‘Tunicate [35] S. R. Pokhrel, S. Verma, S. Garg, A. K. Sharma, and J. Choi, ‘‘An effi-
swarm algorithm: A new bio-inspired based metaheuristic paradigm cient clustering framework for massive sensor networking in indus-
for global optimization,’’ Eng. Appl. Artif. Intell., vol. 90, Apr. 2020, trial IoT,’’ IEEE Trans. Ind. Informat., early access, Jul. 1, 2020, doi:
Art. no. 103541. 10.1109/TII.2020.3006276.

48194 VOLUME 9, 2021

S. Verma et al.: Intelligent Framework Using IoT-Based WSNs for WD

[36] T. M. Behera, S. K. Mohapatra, U. C. Samal, M. S. Khan, M. SATNAM KAUR (Student Member, IEEE)
Daneshmand, and A. H. Gandomi, ‘‘I-SEP: An improved routing received the B.Tech degree in computer science
protocol for heterogeneous WSN for IoT-based environmental and engineering from Punjab Technical University,
monitoring,’’ IEEE Internet Things J., vol. 7, no. 1, pp. 710–717, Jalandhar, India, in 2010, and the M.Tech. degree
Jan. 2020. in computer science and engineering from Guru
[37] R. Sharma, V. Vashisht, and U. Singh, ‘‘eeTMFO/GA: Nanak Dev University Amritsar, India, in 2012.
A secure and energy efficient cluster head selection in wireless She is currently pursuing the Ph.D. degree in com-
sensor networks,’’ Telecommun. Syst., vol. 74, pp. 253–268,
puter science and engineering with the National
Feb. 2020.
Institute of Technology Jalandhar, Punjab, India.
[38] A. Verma, S. Kumar, P. R. Gautam, T. Rashid, and A. Kumar, ‘‘Fuzzy
logic based effective clustering of homogeneous wireless sensor net- Her research interests include software design,
works for mobile sink,’’ IEEE Sensors J., vol. 20, no. 10, pp. 5615–5623, software refactoring, software quality, optimization algorithms, and wireless
May 2020. sensor networks. She is a Student Member of the IEEE Communications
[39] N. Shivappa and S. S. Manvi, ‘‘Fuzzy-based cluster head selection and Society and a Life Member of CSI.
cluster formation in wireless sensor networks,’’ IET Netw., vol. 8, no. 6,
pp. 390–397, Nov. 2019.
[40] S. Kaur, L. K. Awasthi, and A. L. Sangal, ‘‘HMOSHSSA: A hybrid meta-
heuristic approach for solving constrained optimization problems,’’ Eng.
Comput., pp. 1–37, Mar. 2020. DANDA B. RAWAT (Senior Member, IEEE)
[41] L. Wang and Y. Xiao, ‘‘A survey of energy-efficient scheduling mecha- received the Ph.D. degree from Old Dominion
nisms in sensor networks,’’ Mobile Netw. Appl., vol. 11, no. 5, pp. 723–740, University, Norfolk, VA, USA.
Oct. 2006. He is currently a Professor with the Department
[42] A. P. Renold and S. Chandrakala, ‘‘Survey on state scheduling-based of Electrical Engineering and Computer Science
topology control in unattended wireless sensor networks,’’ Comput. Electr. (EECS), the Founding Director of the Howard
Eng., vol. 56, pp. 334–349, Nov. 2016. University Data Science and Cybersecurity Cen-
[43] C.-C. Lin, D.-J. Deng, Z.-Y. Chen, and K.-C. Chen, ‘‘Key design of driving
ter, the Director of the Cyber-security and Wire-
industry 4.0: Joint energy-efficient deployment and scheduling in group-
less Networking Innovations (CWiNs) Research
based industrial wireless sensor networks,’’ IEEE Commun. Mag., vol. 54,
no. 10, pp. 46–52, Oct. 2016. Laboratory, and the Graduate Program Director
[44] M. Mukherjee, L. Shu, L. Hu, G. P. Hancke, and C. Zhu, ‘‘Sleep of the Howard-CS Graduate Programs and the Director of the Graduate
scheduling in industrial wireless sensor networks for toxic gas Cybersecurity Certificate Program, Howard University, Washington, DC,
monitoring,’’ IEEE Wireless Commun., vol. 24, no. 4, pp. 106–112, USA. He is involved in research and teaching in the areas of cybersecurity,
Aug. 2017. machine learning, and wireless networking for emerging networked sys-
[45] P. S. Banerjee, S. N. Mandal, D. De, and B. Maiti, ‘‘iSleep: Ther- tems, including cyber-physical systems, the Internet-of-Things, smart cities,
mal entropy aware intelligent sleep scheduling algorithm for wire- software-defined systems, and vehicular networks. He has served as the
less sensor network,’’ Microsyst. Technol., vol. 26, pp. 2305–2323, Vice-Chair of the Executive Committee of the IEEE Savannah Section from
Dec. 2019. 2013 to 2017. He has been serving as an Editor/Guest Editor for more
[46] K. Xiao, R. Wang, H. Deng, L. Zhang, and C. Yang, ‘‘Energy-aware than 30 international journals. He has been in Organizing Committees for
scheduling for information fusion in wireless sensor network surveil- several IEEE flagship conferences, such as the IEEE INFOCOM, IEEE CNS,
lance,’’ Inf. Fusion, vol. 48, pp. 95–106, Aug. 2019. IEEE ICC, IEEE GLOBECOM, IEEE CCNC, IEEE ICNC, and IEEE AINA.
[47] T. Rappaport, Wireless Communications: Principles and Practice. He served as a Technical Program Committee (TPC) Member for several
Upper Saddle River, NJ, USA: Prentice-Hall, 1996.
international conferences, including IEEE INFOCOM, IEEE GLOBECOM,
IEEE CCNC, IEEE GreenCom, IEEE AINA, IEEE ICC, IEEE WCNC, and
IEEE VTC conferences.

CHEN XI received the Ph.D. degree. He is cur-

rently a Full Professor, a Ph.D. Supervisor, and the
Chair of the Marketing and Electronic Business
Department, a Visiting Scholar of the Michael G.
Foster School of Business, University of Washing-
SANDEEP VERMA (Senior Member, IEEE) ton. He is also the Innovation Team Leader of Nan-
received the B.Tech. degree from the Department jing University. He has authored more than 60 ref-
of ECE, Ludhiana College of Engineering and ereed journal/international conference papers and
Technology, in 2010, the M.Eng. degree in ECE book chapters, more than 60 articles among them
from NITTTR Chandigarh, in 2013, and the Ph.D. have been indexed by SCI/SSCI and EI, including
degree from the Electronics and Communication lots of articles in JCR I and II sections. He has published two monographs
Engineering (ECE) Department, Dr. B. R. Ambed- and owned several national authorized patents. He got the 2015 Youth Award
kar National Institute of Technology, Jalandhar, of Management in China, the Best Paper Award of Chinese Academy of
in 2020. He is currently working as an Assistant System Simulation, the Outstanding Achievement Award of Philosophy and
Professor with the Department of ECE, Dr. B. R. Social Sciences of Jiangsu Province, and so on. He has been elected to
Ambedkar National Institute of Technology. His research interests include Ministry of Education for New Century Excellent Talents Scheme since
energy-efficient wireless sensor networks, the IoT-based architectures, and 2011, and three high-level talents plans in Jiangsu Province. He is the Chair
intelligent transportation system. He has published more than 20 articles and the Commissioner of some international/national academic associa-
in reputed journals namely, IEEE TRANSACTION ON INDUSTRIAL INFORMATICS, tions, the Chair of some technical committees, the Chair of organization
IEEE INTERNET OF THINGS (IoT) JOURNAL, Applied Soft Computing, Peer-to- committee of some refereed conferences, and the Associate Editor and an
Peer Networking and Applications, and many others. He has also reviewed Editorial Member of some refereed international journals and conferences.
more than 40 articles from the refereed journals which include International His research interests include business intelligence, services engineering and
Journal of Communication System, Online (Wiley), IEEE ACCESS, Wireless management, computational simulation experiments, Internet data analysis,
Network, and IET WSS. and so on.

VOLUME 9, 2021 48195

 S. Verma et al.: Intelligent Framework Using IoT-Based WSNs for WD

LINSS T. ALEX (Senior Member, IEEE) received NOOR ZAMAN JHANJHI is currently working
the degree in electrical and electronics engineering as Associate Professor with Taylor’s University,
from Mahatma Gandhi University, India, in 2007, Malaysia. He has great international exposure in
the master’s degree in power electronics from academia, research, administration, and academic
Amrita Vishwa Vidyapeetham University, India, quality accreditation. He worked with ILMA Uni-
in 2011, and the Ph.D. degree in electrical and versity, and King Faisal University (KFU) for a
electronics engineering from the SRM Institute of decade. He has 20 years of teaching and admin-
Science and Technology, Chennai, India, in 2020. istrative experience. He has an intensive academic
He is currently working as an Associate Professor quality accreditation background in higher educa-
with the Department of Electrical and Electronics tion besides scientific research activities. He had
Engineering, Met’s School of Engineering, A. P. J. Abdul Kalam Techno- worked a decade for academic accreditation and earned ABET accreditation
logical University, Kerala, India. His research interests include renewable twice for three programs at CCSIT, King Faisal University, Saudi Arabia.
energy systems, the Internet of Things, artificial intelligence, and electric He has awarded as top reviewer 1% globally by WoS/ISI (Publons) recently
vehicles. for 2019. He has supervised several postgraduate students, including the
master’s and Ph.D. degree. He is also an Associate Editor of IEEE ACCESS,
moderator of IEEE TechRxiv, a Keynote Speaker for several IEEE interna-
tional conferences globally, an External Examiner/Evaluator of Ph.D. and
master’s for several universities, a Guest editor of several reputed journals,
a member of the Editorial Board of several research journals, and an active
TPC Member of reputed conferences around the globe.

48196 VOLUME 9, 2021