This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2749606, IEEE Access

JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 1

An Efficient Centroid-based Routing Protocol for

Energy Management in WSN-Assisted IoT
Jian Shen, Member, IEEE, Anxi Wang, Chen Wang, Patrick C. K. Hung, Member, IEEE,
and Chin-Feng Lai, Senior Member, IEEE

Abstract—Wireless sensor networks (WSNs) distribute hun- intelligent transportation [5], civilian domains [6] and other
dreds to thousands of inexpensive micro-sensor nodes in their fields [7], [8], [9]. In addition, WSNs can be used for collecting
regions, and these nodes are important parts of Internet of data in IoT [10]. With the help of cloud computing, this
Things (IoT). In WSN-assisted IoT, the nodes are resource
constrained in many ways, such as storage resources, computing network can provide great convenience for our daily lives [11],
resources, energy resources and so on. Robust routing protocols [12].
are required to maintain a long network lifetime and achieve WSN-assisted IoT in a wired network has the advantages
higher energy utilization. In this paper, we propose a new energy- of low cost, convenient deployment and good scalability.
efficient centroid-based routing protocol (EECRP) for WSN- However, the defects cannot be ignored. One of the main
assisted IoT to improve the performance of the network. The
proposed EECRP includes three key parts: a new distributed drawbacks is limited energy resources. In fact, sensor nodes
cluster formation technique that enables the self-organization of are generally powered by batteries, and it is difficult to add
local nodes, a new series of algorithms for adapting clusters and energy to the sensor nodes due to the rugged environments
rotating the cluster head (CH) based on the centroid position where they operate, which makes energy management an
to evenly distribute the energy load among all sensor nodes, important task for WSN-assisted IoT.
and a new mechanism to reduce the energy consumption for
long-distance communications. In particular, the residual energy Due to the temporary validity of sensors and the high
of nodes is considered in EECRP for calculating the centroid0 s price of replacement, replacing the node components is not
position. Our simulation results indicate that EECRP performs worthwhile. Hence, prolonging the lifetime of the network and
better than LEACH, LEACH-C and GEEC. In addition, EECRP managing the energy resources throughout the entire network
is suitable for networks that require a long lifetime and whose are the most important aspects when improving the perfor-
base station (BS) is located in the network.
mance of WSN routing protocols. The low-energy adaptive
Index Terms—Internet of Things, wireless sensor networks, clustering hierarchy (LEACH) [13] protocol is one of the most
energy management, cluster. common protocols in this field. A considerable number of new
protocols have been designed to improve the performance of
I. I NTRODUCTION LEACH [14], [15], [16], [17], [18], [19], [20], [21] by reducing
the energy of the cluster head (CH) nodes or finding a multi-
I NTERNET of Things (IoT) is the inter-networking of
physical devices, vehicles, buildings, and other items that
contain embedded electronics, software, sensors, actuators and
hop path from CHs to the base station (BS) [22], [23], [24],
[25], [26], [27], [28].
so on. IoT enables these intelligent objects to collect and To simplify the management of the network, the concept
exchange data [1], [2], [3] for different purposes. For example, of cluster is proposed, and the CH nodes are the managers
wireless sensor networks (WSNs) are a typical type of IoT of local clusters. A CH node takes the responsibilities of
network, in which the sensors can detect and monitor the organizing the cluster, establishing the routing table, and
network area. WSNs are developed based on microelectrome- collecting, compressing, and transmitting data. Because of its
chanical systems (MEMS), system on chip (SoC), wireless high frequency of utilization, the energy consumption of a CH
communications and low-power embedded technology. At node is the fastest among the entire network. Searching for a
present, WSNs are widely used in military applications [4], multi-hop path from the CHs to the BS is fundamental for
reducing the energy consumption of CH nodes. The energy
J. Shen is with Jiangsu Engineering Center of Network Monitoring, School consumption of long-distance node communication is very
of Computer and Software, Nanjing University of Information Science &
Technology, Nanjing, China, 210044 and State Key Laboratory of Informa- large. If the network can find a node that does not considerably
tion Security (Institute of Information Engineering), Chinese Academy of contribute to the network and that is located at the edge of the
Sciences, Beijing 100093. E-mail: s [email protected] cluster to be the forwarder, then the energy dissipation rate
A. Wang and C. Wang are with the school of Computer and Soft-
ware, Nanjing University of Information Science & Technology, Nanjing, of CH nodes will be significantly reduced. This approach will
China, 210044. E-mail: Anxi [email protected], [email protected], not reduce the life cycle of the entire network. As mentioned
[email protected] in [29], at the end of the entire sensor network life cycle, the
P. Hung is with the Faculty of Business and IT, University of Ontario
Institute of Technology, 2000 Simcoe Street North, Oshawa, ON L1H 7K4, energy consumption of sensor nodes at the edge of the network
Canada. E-mail: [email protected] is only 10%. Thus, selecting an edge node as an intermediate
C. F. Lai (corresponding author) is with the Department of Engineering node for communication between a CH and BS will not reduce
Science, National Cheng Kung University, 62102 Chia-Yi, Taiwan, ROC. E-
mail: [email protected] the life cycle of the entire network. To summarize, controlling
Manuscript received April XX, XXXX; revised August XX, XXXX. the energy consumption of CH nodes plays an important role

2169-3536 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2749606, IEEE Access

JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 2

in prolonging the lifetime of the network and achieving energy their position and residual energy to the BS. After receiving
efficiency. this information, the BS calculates the average energy of
Aiming at a higher energy efficiency for the entire network, all nodes. Nodes whose energy is lower than the average
a new protocol named energy-efficient centroid-based routing energy will not be selected to be a candidate CH node. In
protocol (EECRP) is proposed to manage energy resources in this way, the energy consumption resulting from the CH node
WSN-assisted IoT. The main contributions of this paper are selection process and the overhead caused by data transmis-
as follows: sion are significantly reduced. However, the large amount of
• A clustering algorithm that operates according to the data transmission tasks at the beginning of each round still
position of the energy centroid and the residual energy cause considerable overhead. In addition, LEACH-C does not
of nodes is constructed. perform as well as LEACH when the BS is located inside of
• An optimization algorithm that is based on the number the sensor network.
of dead nodes and CH nodes is added to the protocol. Lindsey et al. [30] introduced a protocol called power-
• To reduce the number of long-distance communications, efficient gathering in sensor information systems (PEGASIS),
a protection mechanism for EECRP to save the energy in which a chain is formed for all nodes to transfer data packets
of CH nodes is established. to the BS. In [31], Loscri et al. proposed a two-level hierarchy
routing protocol (TL-LEACH), which uses random rotations of
Notably, the above contributions substantially broaden the
local cluster base stations. In this way, TL-LEACH can better
field of applications of the energy-efficient routing protocol
distribute the energy load among the sensors in the network,
by applying an energy centroid with local management and
particularly when the density of the network is higher. In
global control. Moreover, the average energy consumption of
[32], Wei et al. proposed a distributed clustering algorithm
each round is reduced without impacting the network lifetime.
(EC) that determines a suitable cluster size based on the hop
The remainder of this paper is organized as follows. In
distance to the BS. By applying EC, the network makes a
section II, the related works and the motivation for our paper
good balance between the cluster’s energy consumption and
are discussed. Our proposed EECRP is described in detail in
the lifetime of nodes. In 2016, Razaque et al. [33] combined
section III. In section IV, the energy consumption model of
the features of LEACH and PEGASIS to improve the energy
EECRP is presented. In section V, the performance evaluations
efficiency in routing. In the same year, Razaque et al. [34] de-
of our protocol and comparisons with other protocols are
signed H-LEACH, which is used to solve problems of energy
presented. Conclusions are drawn in section VI, and future
considerations while selecting a CH. H-LEACH considers the
works are also mentioned.
residual and maximum energies of nodes for every round while
selecting a CH using threshold conditions. Lin et al. [35] took
II. R ELATED W ORK advantage of a game model to select CH nodes. In addition,
Cluster routing protocols work on the network layer, which a routing protocol named game-theory-based energy-efficient
connects the data link layer and the transport layer. When de- clustering (GEEC) was proposed. GEEC, which is a type
ployed in the network, these protocols can avoid the ”hot spot” of clustering routing protocol, adopts an evolutionary game
problem and obtain better network performance. In previous theory mechanism to achieve energy balance and longevity.
studies, clustering algorithms play a very important role in From the above analysis, note that CH node selection algo-
designing cluster routing protocols. Here, we mainly discuss rithms should take four aspects into consideration: the local
the clustering and CH selection algorithms in hierarchical autonomy, network coverage, node location and the remaining
networks. energy of the node. Previous CH node selection algorithms
LEACH [13] is the first hierarchical routing protocol de- mostly meet one or two with respect to three aspects, and
signed for WSNs. The main characteristics of LEACH in- there is no comprehensive consideration of the influence of
clude local cluster generation and dynamic CH node rotation all four aspects.
combined with data fusion technology. In LEACH, the node
uses a stochastic algorithm, which is shown as Eq. (1), to III. E NERGY-E FFICIENT C ENTROID -BASED ROUTING
decide whether to be a CH node. The details of this stochastic P ROTOCOL (EECRP)
algorithm are discussed in [27]. In this section, the centroid-based routing protocol is dis-
(
P cussed in detail. Subsequently, the energy consumption model
1 , if n N is introduced. Moreover, some definitions, terminologies and
T (n) = 1−P ·(r mod P ) (1)
0 , otherwise assumptions are presented for a better understanding.
However, the process of CH node selection leads to a
significant amount of energy consumption. The large amount A. System Model
of transmitted data leads to a short lifetime of nodes and a 1) Some assumptions and notations: We assume that wire-
lack of monitoring of the area filled with dead nodes. Note less sensor nodes are randomly distributed in the network.
that the extendibility of LEACH is not suitable for large-scale Once the arrangement of the entire sensor network is com-
networks. pleted, the positions of the sensor nodes will not be changed.
LEACH-C is an improved version of LEACH. In LEACH- In addition, the location information of the node is already
C, at the beginning of each round, nodes send information of loaded into the node when the network is deployed. We also

2169-3536 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2749606, IEEE Access

JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 3

assume that each node knows the position of the BS and its of the CH node in each cluster. The Avg-energy indicates the
remaining energy at any time. The shape of the entire tested average energy of the network. The receiver will update all the
area that the nodes are distributed in is a rectangle. We create information of the FEEDBACK message in its routing table.
a Cartesian coordinate system with its origin point located at
the lower left corner of the entire region. In EECRP, CH nodes
have direct communication with the BS. We set the number
of CH nodes at the percentage of 5% during the simulation.
2) Sensor Energy Consumption Model: The sensor energy
model is an important metric used to measure the performance
of EECRP. The model that we choose is a popular one
mentioned in previous works [36]. The model is shown as Fig. 2: The format of the FEEDBACK message
Eq. (2).
( The main task of the initialization phase is to exchange
l · (er + et + f s d2 ) if d < dT h messages between the BS and the sensor nodes. The messages
E= (2)
l · (er + et + mp d4 ) if d > dT h mainly contain the location and energy information of the
In Eq. (2), E is the total energy consumption when de- nodes, the average energy of the entire network, the CH
livering a single l-bit packet from a sender to a receiver. nodes chosen by the BS in the first round and the longest
The baseline energy consumption levels at the sender and transmission distance. After implementing the entire phase,
receiver radios are indicated by er and et , respectively. The the exchanged information will be stored in the routing tables
value of d is the distance of the link between the sender of the BS and sensor nodes. In addition, the information of
and receiver. The transmission energy consumption is denoted the routing table will update in real time as the entire network
by either f s d2 or mp d4 depending on the distance d and operates.
the distance threshold dT h . For d ≤ dT h , f s is used to 2) First cluster head selection phase: After the LOCA-
reflect “free-space” conditions, while mp represents longer TION message and FEEDBACK message are received, the BS
links potentially affected by “multi-path” fading. determines the node whose energy is greater than the Avg-
energy. Specifically, the first round of CH node selection is
random because the energy levels of each node are identical.
B. Clustering Algorithm Scheme
The selection criteria include the most suitable percentage P
The clustering algorithm is used to find the most appropriate and the entire network, which should be monitored by being
CH node for the cluster. The algorithm includes three phases: covered by clusters. In fact, in the first CH node selection
the initialization phase, the first cluster head selection phase phase, each node in a cluster checks its own ID to determine
and the rotate phase. whether it is a CH node.
1) Initialization phase: A LOCATION message is first The first cluster head selection phase is mainly to complete
sent to the BS from every sensor node. The format of a the identification of the broadcast information of the BS. When
LOCATION message is shown in Fig. 1. The field Message the ID of the node is the same as that of the CH node, the
type indicates that it is a LOCATION message. The field node confirms that the node itself is the CH node and opens
Sender ID contains the node ID of the message sender. The the transmitting antenna to prepare for the next phase. When
X coordinate represents the abscissa of the node0 s location. the IDs are different, the node closes the transmitting antenna
The Y coordinate provides the ordinate of the node0 s location. to save energy and opens the receiving antenna to prepare for
The Energy level provides the energy status of the node. the arrival of information from the CH node.
3) Rotate phase: After the first cluster selection phase,
the first CH node broadcasts the schedule message to the
neighbor nodes. The CH node0 s ID and position are contained
in this message. All CH nodes in the network send this
message. When the neighbor nodes receive this message,
they will determine whether it belongs to this cluster based
Fig. 1: The format of the LOCATION message on the CH node0 s ID in the FEEDBACK message and the
schedule message. At this point, the clustering is completed.
At the end of this phase, the BS will calculate the distance The general nodes send the information about their locations
between each node and the BS itself in the network. The BS and energy to the associated CH node. The CH node calculates
clusters based on the distances. Subsequently, the BS will the location of the energy centroid of the cluster. The node
update the node table, including each node0 s position and that is nearest to the energy centroid will be chosen to be the
energy. Then, the BS broadcasts a FEEDBACK message to the candidate CH node.
nodes in one cluster specifically. The format of the message is The rotate phase is established to choose a candidate CH
shown in Fig. 2. Message type is used to specify the type of node. In this way, the network can uniformly distribute the
the message to inform the receiver. The MAX-dist field, which energy consumption to all nodes in the network. EECRP
is calculated by the BS, delivers the maximum broadcast range simultaneously meets the four aspects that were mentioned in
to each node in the network. The CH0 s ID represents the ID section II. In the first round, the CH nodes are chosen by the

2169-3536 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2749606, IEEE Access

JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 4

BS. Thus, the BS has an overall view of the network. When the RR
network is running, the CH node is selected in the local cluster, x · µdσ
Mx D
which means that the network is self-adapting. Moreover, the Ywc = = RR (4)
M µdσ
selected CH node is the closest node to the energy centroid D
that can maximize the network coverage. From the calculation n
X Ei rs
of the position of the energy centroid, which is described in · X
detail below, we find that the calculation is based on the node i=0
E0
Xec = (5)
location and the remaining energy of the node. In other words, N
the cluster algorithm in EECRP can make a improvement to n
the existing algorithms.
X Ei rs
· Y
Then, a general node is added to the candidate CH node i=0
E0
sequence. The added nodes must meet the following two Yec = (6)
N
conditions. If the weights of sensor nodes in the cluster are known
1) and evenly distributed, then we can use Eqs. (3) and (4) to
1) Its energy level is more than the average energy level of calculate the position of the weight centroid [37]. However,
the clusters. in the real case, the effect of the node weight of the nodes in
2) The distance from the energy centroid of the network to the network is not important for the network lifetime. Hence,
the node is less than the average distance of the entire with node location information and the residual energy level,
nodes to the energy centroid. we can use Eqs. (5) and (6) to calculate the position of the
Moreover, in the process of CH node rotation, the numbers energy centroid. The energy centroid can reflect the energy
of CH nodes and dead nodes are taken into consideration. In distribution during smooth operation of the network.
other words, the number of clusters decreases as the number
of dead nodes increases to keep the value of P unchanged. D. Protective Mechanism Scheme
Here, note that P represents the desired percentage, which is As mentioned above, the MAX-dist value is broadcast to
defined in section II. each node by the BS with the FEEDBACK message. This
value is the protective mechanism of EECRP. The MAX-dist,
C. Centroid Algorithm Scheme calculated by the BS, is the communication threshold. Note
In the field of mathematics, the centroid is the center of that the MAX-dist value is calculated by taking the Avg-energy
weight, which is the imaginary point of mass concentration. in Eq. (2). When CH nodes are ready to send packets to the
Centroid position is important in engineering fields. In our BS, the CH nodes compare their own distances to the BS with
study, we use the concept of the cluster energy centroid rather the MAX-dist sent by the BS in the FEEDBACK message.
than the traditional weight centroid. The reasons for using If the distance is smaller than the MAX-dist, then CH nodes
the cluster energy centroid are as follows. First, the weight will turn on their antennas and send packets. Conversely, if the
of sensor nodes in the network is meaningless. Second, the distance is larger than the MAX-dist, then CH nodes will stop
center of the weight centroid of nodes for the entire cluster is transmitting the data to the BS and store the packets, waiting
meaningless because node location and weight do not change for the next round. Although this mechanism will lead to a loss
in the operation of the network. Finally, in the entire network, of data packets in the short term, it can avoid the long-distance
the energy of the sensor nodes is the only factor that changes. communications of CH nodes, which can reduce the energy
The energy centroid can intuitively display the distribution of consumption of the network, as shown in Eq. (16) in section
residual energy in the network. IV. From the perspective of monitoring the entire network, it
Note that Eqs. (3) and (4) are used to calculate the position is more good than harm to utilize the protective mechanism.
of the weight centroid in the field of mathematics. Eqs. (5) and The details of EECRP are presented in Algorithm 1.
(6) are used to calculate the position of the energy centroid.
IV. E NERGY C ONSUMPTION M ODEL
Clearly, we take the residual energy and the position of node
i into consideration to calculate the energy centroid. In this section, we provide the value of energy consump-
Next, the meanings of the parameters used in the equations tion Eround of the network during a round period. This
are introduced. µ is the density of nodes’ weights in the value is needed by the EECRP algorithm to calculate the
cluster, S is the measure of cluster area, dσ is the differential equalized lifetime value and to determine the effect of the
of weight, and dMx and dMy are the static moment to the protective mechanism. The total initial deployment
P energy in
x-axis and y-axis, respectively. Ei rs represents the residual the region can be represented as E0 all = j  Ci E0 (j) =
2
P
energy level of node i, E0 is the initial value of energy, and j  Ci E0 (j) · π ri σ = E0 · abσ, where E0 is the average
X and Y are the X coordinate and the Y coordinate of node deployment energy and ri denotes the radius of cluster i. The
i, respectively. Xwc and Ywc are the results of the weight lifetime of the entire network is L ≈ EEround
0 all
. The round energy
centroid. Xec and Yec are the results of the energy centroid is the sum of energy consumption values for cluster formation,
RR route discovery and data communication events. The round
x · µdσ energy is given by Eround = Ecluster + Ecomm + Ein−comm .
My
Xwc = = D RR (3) In the following parts of this section, the individual phases of
M µdσ EECRP are separately calculated.
D

2169-3536 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2749606, IEEE Access

JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 5

Algorithm 1 The process of EECRP lower energy consumption.

BS ← LOCATION messages
Calculate: Avg-energy and MAX-dist Eout1 = lloc {(ab − π d2T h )σ et
N odes ← FEEDBACK message (ab−π d2T h )σ
X (8)
if (Candidate CHs exist) then + mp [(XB − Xi )2 + (YB − Yi )2 ]2 }
Clustering i=1
Recalculate centroid
Create schedule Eqs. (9) and (10) describe the energy consumed by nodes to
if (Alive node number > number of clusters) then receive the FEEDBACK message from the BS. The condition
if (Round time is not over then of receiving the LOCATION message is the same as that of
CH ← data (normal nodes) receiving the FEEDBACK message. Eq. (9) and Eq. (10) are
Fusion data related to the receiving energy. We can determine that in this
BS ← data (CH nodes) phase, when the nodes are located far from the BS, the number
return of long-distance communications is two times the number of
end if nodes. This result demonstrates that the conclusion drawn from
Recalculate: Avg-energy and MAX-dist Eq. (8) holds.
N odes ← FEEDBACK message
Select candidate CH nodes Ein2 = lf d {π d2T h σ er
return π d2T h σ
end if X (9)
+ f s [(XB − Xi )2 + (YB − Yi )2 ]}
return i=1
end if

Eout1 = lf d {(ab − π d2T h )σ er

A. Energy Consumption in Cluster Formation (ab−π d2T h )σ
X
+ mp [(XB − Xi )2 + (YB − Yi )2 ]2 }
The energy consumption of nodes sending a LOCATION
i=1
message to the BS is calculated in this section. The energy (10)
consumption can be described as Eq. (7) when the distance of After exchanging LOCATION and FEEDBACK messages,
the link between the sender and receiver is less than dT h . We normal nodes join the nearest cluster. The value of l0 is the
assume that the area is circular with the centroid as the center size of the control message. The energy consumption can
and dT h as its radius. be calculated as Eq. (11). In contrast to the formulas for
In Eq. (7), lloc represents the size of the LOCATION calculating information exchange, Eq. (11) is used to calculate
message. lf d represents the size of the FEEDBACK message. the energy consumptions of both the sender and the receiver.
Clearly, the relationship between lloc and lf d can be described This is because the BS is independent of the entire network
as lf d = 0.8· lloc . σ indicates the density distribution of sensor in our calculations. In addition, the energy of the BS is
nodes. XB and YB are the location information of the BS. Xi limitless, which indicates that the broadcast and receive energy
and Yi are the location information of ordinate nodes. XC consumptions of the BS do not need to be calculated.
and YC are the location information of CH nodes. ab shows
the size of the entire sensor network. P indicates the desired
percentage of CH nodes. ldata is the size of the packet that Enon ch = l0 {ab(1 − P )σ et + abP σ er
has been compressed by the CH node. ldata0 is the size of the ab(1−P )σ
packets that are sent to the CH node by common nodes. The
X (11)
+ f s [(XC − Xi )2 + (YC − Yi )2 ]}
symbols used in the following equations are the same as in i=1
Eq. (7).
The total energy consumption in the cluster formation phase
Ein1 = lloc {π d2T h σ et is shown in Eq. (12), which is the sum of Eqs. (7) ∼ (11).
π d2T h σ
X (7)
+ f s [(XB − Xi )2 + (YB − Yi )2 ]} Ecluster = Ein1 + Eout1 + Ein2 + Eout2 + Enon−ch (12)
i=1
After a simple mathematical reduction using l = 0.8· lloc =
The energy consumption can be described as Eq. (8) when lf d , Eq. (12) can be simplified as Eq. (13). The relationship
the distance of the link between the sender and receiver between LOCATION and FEEDBACK messages is shown in
is larger than dT h . The energy consumption of transmitting subsection III(B). Here, the problem of long-distance com-
packages by nodes with a larger d in Eq. (8) is the exponential munication must be mentioned again. From Eq. (13), we can
time of that in Eq. (7). A conclusion can be drawn that observe that the energy consumption of the communication
shorter long-distance transmission times can contribute to channel is an important component of the energy consumption

2169-3536 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2749606, IEEE Access

JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 6

of the entire network. Thus, it is essential to reduce the channel

energy consumption. Eround = Ecluster + Ecomm + Ein−comm =
l · {1.25abσ et + abσ er
Ecluster = l · {1.25abσ et + abσ er π d2T h σ
X
π d2T h σ + 2.25 f s [(XB − Xi )2 + (YB − Yi )2 ]}
X
+ 2.25 f s [(XB − Xi )2 + (YB − Yi )2 ]} i=1
(ab−π d2T h )σ
i=1 X
(ab−π d2T h )σ + 2.25 mp [(XB − Xi )2 + (YB − Yi )2 ]2 }
X
+ 2.25 mp [(XB − Xi )2 + (YB − Yi )2 ]2 } i=1

i=1 + l0 {ab(1 − P )σ et + abP σ er

+ l0 {ab(1 − P )σ et + abP σ er ab(1−P )σ
X
ab(1−P )σ + f s [(XC − Xi )2 + (YC − Yi )2 ]}
X
+ f s [(XC − Xi )2 + (YC − Yi )2 ]} i=1

i=1 + ldata {abP σ et

(13) abP
Xσ
+ f s [(XB − XCi )2 + (YB − YCi )2 ]}
i=1
B. Energy Consumption in Transmitting Data to the BS + ldata0 {ab(1 − P )σ et + abP σ er
ab(1−P )σ
In this part, the energy consumption model is very sim- X
+ f s [(XC − Xi )2 + (YC − Yi )2 ]}
ple because only cluster head nodes consume energy during
i=1
transmission. Moreover, the protective mechanism has a con- (16)
siderable influence on the performance. The result is shown As shown in Eq. (16), the energy consumption of the
in Eq. (14). Thanks to the protective mechanism, the long- network using EECRP is the information exchange in the
distance communication is greatly reduced in Eq. (14). The network initialization phase. This means that when the network
purpose of doing so is to prolong the survival time of CH enters the stable operation phase, the energy consumption
nodes, thereby prolonging the network life cycle. of the normal nodes can be ignored. First, a considerable
amount of energy consumption occurs in the initial phase of
Ecomm = ECH−BS = ldata {abP σ et the network. As mentioned in section III, the nodes need to
abP
Xσ send the LOCATION message to the BS at the very beginning.
(14)
+ f s [(XB − XCi )2 + (YB − YCi )2 ]} The distance between the node and the BS is relatively larger
i=1 than that between the node and the CH node in one cluster. In
addition, nodes also need to receive the FEEDBACK message
sent from the BS in the initialization phase. For most of
C. Energy Consumption in Transmitting Data to the CH nodes the nodes, these two communications belong to long-distance
in Clusters communication; thus, the energy consumption will be very
In this part, the normal nodes transmitting data packets significant. Second, the energy consumption of the nodes in
consume their own energy, and the cluster head nodes re- the stable operation phase is relatively low. Hierarchical sensor
ceiving data also consume energy. Because data packets are networks are known to achieve local clustering; thus, the local
transmitted in one cluster, the value of d is smaller than that of normal node only communicates with the CH node of the
dT h . The energy consumption model is shown in Eq. (15). The cluster. The average transmission distance in the cluster is
calculation in Eq. (15) is similar to that in Eq. (11). Sending highly related to the location of the CH node. In this paper, we
and receiving information in Eq. (15) is the data information propose a CH node selection algorithm based on the location
perceived by sensors rather than control information. of the energy centroid, where the CH node is located in
the energy center of the cluster and the most concentrated
Ein−comm = Ein−cluster = area of the nodes. In LEACH, the CH node is selected by
ldata0 {ab(1 − P )σ et + abP σ er Eq. (1). If the CH node is located at one edge of the network,
then the transmission distance of nodes on the other side of
ab(1−P )σ
X the network will be greatly increased. As time passes, the
+ f s [(XC − Xi )2 + (YC − Yi )2 ]}
network’s coverage rate will be greatly reduced, which will
i=1
(15) directly lead to a higher error rate of data. Regarding the stable
operation phase of the proposed EECRP, the CH node of the
next round is calculated by the CH node of the previous round.
D. Total Energy Consumption in a Round Note that the candidate CH node is located in the center of
the energy network; thus, it can greatly improve the ability of
The total energy consumption in a round is the sum of real-time monitoring.
Ecluster , Ecomm and Ein−comm , which is shown in Eq. (16). Moreover, when the distance between the CH node and

2169-3536 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2749606, IEEE Access

JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 7

TABLE I: Parameters used in simulation

Parameter Value
Network size 100×100
BS’s locations inside
Number of sensor nodes 100
Initial energy (J) 2
f s 10pJ/bit/m2
mp 0.0013pJ/bit/m4
Transmitting/Reception energy 50 nJ/bit
l (length of data) 500 bit
Mac 802.11
Bandwidth 200 Kbps
Beam forming energy (nJ/bit) 5

the local node is larger than the MAX-dist, the data will not
be sent and will be stored locally until a neighbor node is Fig. 3: The rounds at which the first and last nodes die
selected to be a new CH node. The number of long-distance versus different values of Ps
communications is reduced by the protective mechanism in
EECRP. In addition, the number of long-distance communica-
tions between CH nodes and the BS is also greatly reduced.
The stored data packets will be sent to new CH nodes in the
following rounds. The influence of the energy consumption of
long-distance communication on the survival time of cluster
heads is remarkable. The protective mechanism plays a very
important role, as indicated by the simulation results and
comparisons.

V. P ERFORMANCE E VALUATION
Taking the location of the BS into consideration, the
selection of the value of P in Eq. 1 and the comparison
between LEACH and LEACH-C are implemented using the
ns-2 simulator [38]. The version of ns-2 used in our simulation
is ns-2.35. The software platform that we run the protocols on Fig. 4: Average data transmission versus different values of
is Ubuntu 12.04. Ps

A. Simulation Parameters
In the simulation part, the network is implemented in a As vividly shown in Fig. 3, when the value of P is small,
100 m × 100 m field with 100 nodes randomly distributed which represents a small number of CH nodes, the nodes
in the network area. Each node is set to have 2 J as its initial begin to die in a very short time. This result might be due
energy. Furthermore, the BS is located in the sensor network. to the shortage of small Ps, although the lifetime of the entire
The detailed parameters that were used in our simulation are network becomes longer. This results in large parts of the
listed in TABLE I. In addition, the simulations among LEACH, network being not under monitoring in the later rounds. In
LEACH-C, GEEC and EECRP are also implemented using the addition, Fig. 3 indicates that the average data transmission of
ns-2 simulator. networks with fewer CH nodes is larger, which means that the
The following three performance measurements are consid- data integrity is poor.
ered: the total number of nodes that are alive, the total energy However, when there is a large number of CH nodes in the
dissipation and the number of messages received at the BS. sensor network, all aspects of the sensor network are in poor
The lifetime of the network is defined as the round number condition. In detail, shorter lifetime, smaller data transmission
when the remaining nodes are not sufficient to form clusters. and quicker energy dissipation become the shortages of the
networks with large Ps, as shown in Figs. 3, 4, and 5.
B. Simulation Results Taking all aspects into consideration, setting the percentage
The results of part 1 are shown in Figs. 3, 4, and 5. To of CH nodes to 5% leads to a good condition of the network.
evaluate the setting of the value of P in Eq. (1), we consider The energy of nodes dissipates slowly, and as the nodes die,
the number of nodes that are alive, the size of transmitted network area that is not being monitored can hardly be found.
data and the total energy dissipation through the entire network In other words, considering the influence on lifetime and the
with the BS located in the network area. The value of P ranges rate of energy dissipation, 5% should be the best setting of
from 1% to 12%. CH nodes.

2169-3536 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2749606, IEEE Access

JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 8

Fig. 6: The number of sensor nodes that are alive in the four
protocols
Fig. 5: The total energy dissipated versus different values of
Ps

C. Comparison Results
EECRP and the other three protocols (LEACH, LEACH-
C and GEEC) are simulated in this part. Each of the four
protocols is analyzed from three different aspects, which are
the number of sensors still alive, the number of data signals
received by the BS and energy dissipation.
In this part, as illustrated in Figs. 6, 7, and 8, several metrics
are taken into consideration to evaluate the performances of
LEACH, LEACH-C, GEEC and EECRP when the BS is
located inside of the network. The detailed comparison of the Fig. 7: The number of messages received by the BS in the
four protocols is as follows. four protocols
The number of nodes that are alive: The number of nodes
that are alive indicates the lifetime of WSNs, which is the
most important parameter of routing protocols. As shown in is less than the MAX-dist. The details about this protective
Fig. 6, there are considerable differences in the number of mechanism are presented in section III.
sensor nodes that are alive among EECRP, LEACH, LEACH-C Energy dissipation: As shown in Fig. 8, the speed of total
and GEEC. First, the first node dies at approximately the 100th energy dissipation of LEACH-C remains at a very high level.
round in LEACH-C. Regarding LEACH, GEEC and EECRP, In other words, LEACH-C performs not as well as LEACH,
the first node dies at approximately the 400th round. Second, GEEC and EECRP in this aspect. In addition, the speed of
the nodes in LEACH-C die in a very short time due to the energy consumption of EECRP is slower than that of LEACH
large amount of communicating control messages, whereas the after the 400th round, which means that nodes in EECRP can
network using EECRP has a very long lifetime. The lifetime monitor the network for a long period. In addition, GEEC
of EECRP is longer than that of LEACH by approximately and EECRP have similar performance in terms of energy
100 rounds and longer than that of GEEC by approximately consumption.
50 rounds.
The number of messages received by the BS: Fig. 7
shows that LEACH-C delivers considerably fewer messages D. Comparison of EECRP and the Conventional Protocols
to the BS than any other protocols. The numbers of messages The performance and comparison of the characteristics of
received by the BS in the network using EECRP are equal LEACH, LEACH-C, GEEC and EECRP are presented in
to those using LEACH and GEEC before the 400th round. TABLE II. The characteristics used to compare the protocols
However, after the 400th round, EECRP0 s packet delivery are life cycle, scalability, computation and communication
slowly increases, which is due to the setting of MAX-dist overhead, path selection, location awareness and mobility.
in EECRP. When increasingly more nodes die, the distance As shown in TABLE II, EECRP performs better than
between two nodes in the network increases. The node will LEACH, LEACH-C and GEEC in terms of life cycle, scala-
stop sending information to the CH node and the BS to save bility, computation and communication overhead and location
energy when the distance to the CH node or BS exceeds the awareness.
value of MAX-dist. Nodes will store the data packets in their However, with static sensor nodes and a single hop between
storage spaces. However, the messages that are stored will be the BS and CH nodes, EECRP is suitable for the networks
sent in the following round until the communication distance where sensor nodes do not need to change locations. Finally,

2169-3536 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2749606, IEEE Access

JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 9

TABLE II: Comparison among EECRP, LEACH and LEACH-C

Protocol Properties LEACH LEACH-C GEEC EECRP

Life cycle 540 rounds 480 rounds 670 rounds 720 rounds
Scalability Bad Bad Good Good
Computation & communication Setting up and Setting up and Setting up and Setting up and
overhead maintaining cluster maintaining cluster maintaining cluster maintaining cluster
Path selection Single hop Single hop Single hop Single hop
Location awareness No Yes Yes Yes
Mobility Fixed Fixed Fixed Fixed

R EFERENCES
[1] C. Floerkemeier, M. Langheinrich, E. Fleisch, F. Mattern, and S. E.
Sarma, “The internet of things,” Electronics World, vol. 297, no. 6, pp.
949 – 955, 2017.
[2] M. Az, Mart, N. Cristian, and B. Rubio, “State-of-the-art, challenges,
and open issues in the integration of internet of things and cloud
computing,” Journal of Network and Computer Applications, vol. 67,
no. C, pp. 99–117, 2016.
[3] S. Misra, M. Maheswaran, and S. Hashmi, “Securing the internet of
things,” Computer Fraud and Security, vol. 2016, no. 4, pp. 15–20,
2016.
[4] T. Alhmiedat, A. Abu Taleb, and M. Bsoul, “A study on threads detection
and tracking systems for military applications using wsns,” International
Journal of Computer Applications, vol. 40, no. 15, pp. 12–18, 2012.
[5] G. Xing, M. Li, T. Wang, W. Jia, and J. Huang, “Efficient rendezvous
algorithms for mobility-enabled wireless sensor networks,” IEEE Trans-
Fig. 8: The total energy dissipation in the four protocols actions on Mobile Computing, vol. 11, no. 1, pp. 47–60, 2012.
[6] A. Liu, J. Ren, X. Li, Z. Chen, and X. S. Shen, “Design principles and
improvement of cost function based energy aware routing algorithms
for wireless sensor networks,” Computer Networks, vol. 56, no. 7, pp.
it is very suitable for us to deploy EECRP in networks where 1951–1967, 2012.
the BS is located inside the network. [7] J. Shen, H. Tan, J. Wang, J. Wang, and S. Lee, “A novel routing protocol
providing good transmission reliability in underwater sensor networks,”
Journal of Internet Technology, vol. 16, no. 1, pp. 171–178, 2015.
[8] C. F. Lai, S. Zeadally, J. Shen, and Y. X. Lai, “A cloud-integrated
VI. C ONCLUSION appliance recognition approach over internet of things,” Journal of
Internet Technology, vol. 16, no. 7, pp. 1157–1168, 2015.
In this paper, we propose a new energy-efficient centroid- [9] J. Zhu, J. Liu, Z. Hai, and Y. G. Bi, “Research on routing protocol facing
based routing protocol (EECRP) to manage the energy of to signal conflicting in link quality guaranteed wsn,” Wireless Networks,
vol. 22, no. 5, pp. 1739–1750, 2016.
WSN-assisted IoT by solving the problem of forming clusters, [10] Q. Chi, H. Yan, C. Zhang, Z. Pang, and L. D. Xu, “A reconfigurable
which is based on the distance to the energy centroid. We smart sensor interface for industrial wsn in iot environment,” IEEE
also propose an optimization algorithm based on the number Transactions on Industrial Informatics, vol. 10, no. 2, pp. 1417–1425,
2014.
of dead nodes and the number of cluster head nodes. From [11] J. Shen, T. Zhou, D. He, Y. Zhang, X. Sun, and Y. Xiang, “Block
the simulation results, when the BS is located in the network, design-based key agreement for group data sharing in cloud computing,”
the EECRP could transmit a considerable amount of data with IEEE Transactions on Dependable and Secure Computing, 2017, DOI:
10.1109/TDSC.2017.2725953.
very low energy dissipation. Meanwhile, the network lifetime [12] J. Shen, J. Shen, X. Chen, X. Huang, and W. Susilo, “An efficient
of EECRP is longer than that of LEACH, LEACH-C and public auditing protocol with novel dynamic structure for cloud data,”
GEEC. In future work, we want to improve the protocol by IEEE Transactions on Information Forensics and Security, 2017, DOI:
10.1109/TIFS.2017.2705620.
finding the multi-hop path from the CH nodes to the BS. The [13] W. R. Heinzelman, A. Chandrakasan, and H. Balakrishnan, “Energy-
multi-hop path is used by CH nodes to transmit data packets. efficient communication protocol for wireless microsensor networks,”
We hope that our future protocol can perform well when the in Proc. 33rd HICSS, 2000, pp. 10–20.
BS is located outside of the network. [14] A. Yektaparast, F. H. Nabavi, and A. Sarmast, “An improvement on
leach protocol (cell-leach),” in Proc. IEEE ICACT, 2012, pp. 992–996.
[15] Y. Wu, S. Fahmy, and N. B. Shroff, “Energy efficient sleep/wake
scheduling for multi-hop sensor networks: Non-convexity and approxi-
ACKNOWLEDGMENT mation algorithm,” in Proc. IEEE INFOCOM, 2007, pp. 1568–1576.
[16] D. Kandris, P. Tsioumas, A. Tzes, G. Nikolakopoulos, and D. D. Ver-
This work is supported by the National Science Foundation gados, “Power conservation through energy efficient routing in wireless
sensor networks,” Sensors, vol. 9, no. 9, pp. 7320–7342, 2009.
of China under Grant No. 61672295, No. 61300237 and No. [17] C. Lung and C. Zhou, “Using hierarchical agglomerative clustering in
U1405254, the State Key Laboratory of Information Security wireless sensor networks: An energy-efficient and flexible approach,” Ad
under Grant No. 2017-MS-10, the 2015 Project of six per- Hoc Networks, vol. 8, no. 3, pp. 328–344, 2010.
[18] B. Wang, X. Gu, L. Ma, and S. Yan, “A variable threshold-value
sonnel in Jiangsu Province under Grant No. R2015L06, the authentication architecture for wireless mesh networks,” International
CICAEET fund, and the PAPD fund. Journal of Sensor Networks, vol. 23, no. 4, pp. 265–278, 2017.

2169-3536 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2017.2749606, IEEE Access

JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 10

[19] P. Raja and P. Dananjayan, “Game theory-based efficient energy con-

sumption routing protocol to enhance the lifetime of wsn,” International
Journal of Information and Communication Technology, vol. 8, no. 4,
pp. 357–370, 2016.
[20] J. Shen, S. Chang, J. Shen, Q. Liu, and X. Sun, “A lightweight
multi-layer authentication protocol for wireless body area
networks,” Future Generation Computer Systems, 2016, DOI:
10.1016/j.future.2016.11.033.
[21] P. Marappan and P. Rodrigues, “An energy efficient routing protocol for
correlated data using cl-leach in wsn,” Wireless Networks, vol. 22, no. 4,
pp. 1415–1423, 2016.
[22] J. Zhang, C. K. Jeong, G. Y. Lee, and H. J. Kim, “Cluster-based
multi-path routing algorithm for multi-hop wireless network,” Future
Generation Communication and Networking, vol. 1, pp. 67–75, 2007.
[23] J. Shen, A. Wang, C. Wang, Y. Ren, and X. Sun, “A rfid based
localization algorithm for wireless sensor networks,” in Proc. ICCCS,
2016, pp. 275–285.
[24] M. Chen, V. C. Leung, and S. Mao, “Directional controlled fusion in
wireless sensor networks,” Mobile Networks and Applications, vol. 14,
no. 2, pp. 220–229, 2009.
[25] J. Shen, D. Liu, J. Shen, Q. Liu, and X. Sun, “A secure cloud-assisted
urban data sharing framework for ubiquitous-cities,” Pervasive and
Mobile Computing, 2017, DOI: 10.1016/j.pmcj.2017.03.013.
[26] Y. Yuan, Z. He, and M. Chen, “Virtual mimo-based cross-layer design
for wireless sensor networks,” IEEE Transactions on Vehicular Technol-
ogy, vol. 55, no. 3, pp. 856–864, 2006.
[27] J. Shen, A. Wang, C. Wang, Y. Ren, and J. Wang, “Performance
comparison of typical and improved leach protocols in wireless sensor
network,” in Proc. IEEE CCITSA, 2015, pp. 187–192.
[28] Y. Zhang, X. Sun, and B. Wang, “Efficient algorithm for k-barrier
coverage based on integer linear programming,” China Communications,
vol. 13, no. 7, pp. 16–23, 2016.
[29] A. Liu, Z. Zheng, C. Zhang, Z. Chen, and X. Shen, “Secure and energy-
efficient disjoint multipath routing for wsns,” IEEE Transactions on
Vehicular Technology, vol. 61, no. 7, pp. 3255–3265, 2012.
[30] S. Lindsey and C. S. Raghavendra, “Pegasis: Power-efficient gathering
in sensor information systems,” in Proc. Aerospace Conference, 2002,
pp. 1125–1130.
[31] V. Loscri, G. Morabito, and S. Marano, “A two-levels hierarchy for
low-energy adaptive clustering hierarchy (tl-leach),” in Proc. IEEE VTC,
2005, pp. 1809–1813.
[32] D. Wei, Y. Jin, S. Vural, and K. Moessner, “An energy-efficient clustering
solution for wireless sensor networks,” IEEE Transactions on Wireless
Communications, vol. 10, no. 11, pp. 3973–3983, 2011.
[33] A. Razaque, M. Abdulgader, C. Joshi, F. Amsaad, and M. Chauhan, “P-
leach: Energy efficient routing protocol for wireless sensor networks,”
in Proc. IEEE LISAT, 2016, pp. 1–5.
[34] A. Razaque, S. Mudigulam, K. Gavini, and F. Amsaad, “H-leach:
Hybrid-low energy adaptive clustering hierarchy for wireless sensor
networks,” in Proc. IEEE LISAT, 2016, pp. 1–4.
[35] D. Lin and Q. Wang, “A game theory based energy efficient clustering
routing protocol for wsns,” Wireless Networks, pp. 1–11, 2016.
[36] J. Yick, B. Mukherjee, and D. Ghosal, “Wireless sensor network survey,”
Computer Networks, vol. 52, no. 12, pp. 2292–2330, 2008.
[37] J. Shen, C. Wang, A. Wang, X. Sun, S. Moh, and P. C. K. Hung,
“Organized topology based routing protocol in incompletely pre-
dictable ad-hoc networks,” Computer Communications, 2016. DOI:
j.comcom.2016.07.009.
[38] M. Khosrowpour, “Network simulator ns-2,” Journal of the Institute of
Image Information and Television Engineers, vol. 65, pp. 946–949, 2011.

2169-3536 (c) 2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.