sensors

Article
An Improved Energy-Efficient Routing Protocol for
Wireless Sensor Networks
Yang Liu 1 , Qiong Wu 2 , Ting Zhao 1 , Yong Tie 1, * , Fengshan Bai 1 and Minglu Jin 3
1 College of Electronic Information Engineering, Inner Mongolia University, Hohhot 010021, China;
[email protected] (Y.L.); [email protected] (T.Z.); [email protected] (F.B.)
2 School of Biological Science and Medical Engineering, Beihang University, Beijing 100083, China;
[email protected]
3 Faculty of Electronic Information and Electrical Engineering, Dalian University of Technology,
Dalian 116024, China; [email protected]
* Correspondence: [email protected]




Received: 18 September 2019; Accepted: 18 October 2019; Published: 21 October 2019


Abstract: Cluster-based hierarchical routing protocols play an essential role in decreasing the energy
consumption of wireless sensor networks (WSNs). A low-energy adaptive clustering hierarchy
(LEACH) has been proposed as an application-specific protocol architecture for WSNs. However,
without considering the distribution of the cluster heads (CHs) in the rotation basis, the LEACH
protocol will increase the energy consumption of the network. To improve the energy efficiency of
the WSN, we propose a novel modified routing protocol in this paper. The newly proposed improved
energy-efficient LEACH (IEE-LEACH) protocol considers the residual node energy and the average
energy of the networks. To achieve satisfactory performance in terms of reducing the sensor energy
consumption, the proposed IEE-LEACH accounts for the numbers of the optimal CHs and prohibits
the nodes that are closer to the base station (BS) to join in the cluster formation. Furthermore, the
proposed IEE-LEACH uses a new threshold for electing CHs among the sensor nodes, and employs
single hop, multi-hop, and hybrid communications to further improve the energy efficiency of the
networks. The simulation results demonstrate that, compared with some existing routing protocols,
the proposed protocol substantially reduces the energy consumption of WSNs.

Keywords: wireless sensor networks; routing protocol; energy efficiency; network lifetime

1. Introduction
Wireless sensor networks (WSNs) generally consist of considerable sensor nodes (SNs) with
limited energy. WSNs are randomly deployed in a particular region to acquire various types of
environmental parameters and transmit information to the base station (BS) for monitoring and
detecting applications [1]. They have been widely applied in forest fire detection, surveillance, military,
human health detection, etc., and thus have attracted the interest of researchers in recent years [2,3].
Because WSNs are usually deployed in hazardous environments, recharging or replacing the batteries
of the SNs is very difficult. Moreover, the manual operation of the network is highly difficult, which
brings some challenges regarding the application of WSNs [4–6]. To remedy these drawbacks, the
efficient use of the battery energy of SNs should be considered as a primary goal when researchers
design protocols and hardware architectures [7]. Therefore, several routing protocols have been
proposed to render the sensor network more energy efficient [8,9].
The cluster formation and various communication modes of transmitting data have been the most
emphasized approaches. In general, compared with non-clustering protocols, cluster-based routing
protocols can efficiently use the SNs in the network [10]. A cluster leader, called the cluster head (CH),

Sensors 2019, 19, 4579; doi:10.3390/s19204579 www.mdpi.com/journal/sensors

Sensors 2019, 19, 4579 2 of 20

is in charge of eliminating the correlated data that can decrease the final data volume. Afterwards,
the CH will transmit the aggregated data to the BS [11–13]. In cluster-based routing protocols, SNs
are divided into many clusters to decrease energy consumption for long distance communication.
The clustering can minimize the overall energy consumption and balance the nodes’ workload, which
is caused by the large difference in the energy depletion between the CHs and other nodes. Therefore,
clustering is an energy-efficient solution for increasing network longevity and improving energy
efficiency. Moreover, most clustering protocols adopt optimal CH selection to avoid the premature
death of the SNs and further extend the lifetime of the network [14–16].
According to [17], we know that SNs consume more energy during communication than during
the computation process. In contrast, some other protocols adopt multi-hop communication, and the
nodes close to the BS have excessive transmission overhead, leading to energy holes in the sensor
field [18]. To address the energy hole problem and prolong the network lifetime, many clustering
protocols have been specifically proposed for WSNs. The low-energy adaptive clustering hierarchy
(LEACH) is the most important hierarchical routing protocol in terms of saving energy compared with
traditional routing protocols [19]. In this type of protocol, the entire network is divided into several
clusters, and each cluster selects a node in a probabilistic manner as a CH that is in charge of receiving,
aggregating, compressing, and sending the information collected from other non-CH nodes to the BS.
With the objective of minimizing the energy consumption of the WSNs, the LEACH routing protocol
designates a single CH node in each cluster and selects the CH on a rotation basis. Although the
efficiency of the LEACH protocol has been closely studied, it still has some drawbacks that require
improvement—on one hand because selecting the CH is based on a random round robin, the number
of CHs is unreasonable in each round, and the nodes at the boundary of the network will be elected
as the CH. On the other hand, there is no consideration of the distribution of the CHs, the threshold
condition, and the remaining energy of each node after the end of each round. Consequently, these
issues will result in energy overheads.
To remedy this problem, based on this classical hierarchical clustering protocol, a variety of
protocols have been proposed [19]. In contrast to the LEACH protocol, a novel distributed protocol
called the scalable energy-efficient clustering hierarchy (SEECH) assigns CHs and relays roles to
different nodes [20]. This protocol generates two fitness functions to determine the capability to become
a CH or relay for each node by only taking their energy consumption into account. However, in the
SEECH protocol, every node may be elected to be a CH or a relay, including the nodes closer to the BS,
which increases the energy consumption. A learning automata-based multilevel heterogeneous routing
(LA-MHR) scheme was proposed in [21]. The LA-MHR utilizes S-model-based learning automata
for selecting CHs, while the cognitive radio spectrum is allocated by the BS. In addition, single hop
communication among different SNs is replaced by a multi-hop communication approach in the
LA-MHR scheme. Since a single communication approach is not suitable for scenes in which the CHs
are far from the SNs, this approach requires a large amount of transmission power. The LEACH-Mobile
(LEACH-M) protocol is proposed for mobile nodes to improve data transfer success rate [22]; however,
the cluster head selection and cluster formation for LEACH-M are similar to the LEACH protocol.
Because of the high control overheads, the energy dissipation of LEACH-M is greater than LEACH. The
optical LEACH (O-LEACH) protocol divides and links a series of WSNs that is around the distributed
fiber sensors (DFS) in the hybrid sensor network with a rectangular topology [23]. Compared with
LEACH, this method extends the lifetime of WSNs and exhibits good energy efficiency. However,
the energy consumption of O-LEACH is not improved. The stable energy efficient network (SEEN)
has been proposed for WSNs [24]. The selection of Cluster Head in this protocol is done on the
basis of the residual energy of nodes as well as the distance parameter. In wireless sensor networks,
few sensor nodes are equipped with higher processing and communicating capabilities. In [25],
an energy consumption-minimizing strategy called LEACH-XMP (LEACH-eXtended Message-Passing)
was proposed, which adopts a sophisticated energy consumption model for CH nodes without
oversimplification. These protocols aim to merely distribute the energy consumption between the CH
Sensors 2019, 19, 4579 3 of 20

and its assistant at the beginning of the first round. This distribution will result in a reduction in the
network lifetime.
To overcome the drawbacks of conventional methods and further prolong the lifetime of WSNs,
we propose a novel improved energy-efficient LEACH (IEE-LEACH) routing protocol in this paper.
In the proposed protocol, the threshold setting introduces four parameters including the initial energy
of nodes, residual energy of nodes, total energy of the network, and average energy of the network.
In the proposed IEE-LEACH protocol, the node closer to the BS than the CH does not take part in the
cluster formation. Thus, the protocol can balance the energy load and decrease the energy consumption.
Furthermore, the proposed IEE-LEACH protocol compares the energy consumption of single hop and
multi-hop communication modes in the data transmission phase. The communication mode with the
least energy consumption will be adopted. Therefore, the proposed approach decreases the overall
communication cost and significantly improves the network lifetime.
The remainder of this paper is organized as follows. Section 2 briefly introduces the system model.
The proposed IEE-LEACH protocol is indicated in Section 3. We demonstrate the performance
evaluation of the proposed protocol via simulations compared to the conventional protocols in
Section 4. Finally, conclusions are drawn in Section 5.

2. System Model

2.1. Network Model and Assumptions

In this section, the sensor network is introduced as the WSN model. The n SNs are uniformly and
randomly placed in L ∗ L square area and are manually deployed in a complex environment. The set
of SNs is given by
S = (s1 , s2 , ......sn ). (1)

The structure of the network is illustrated in Figure 1. Each cluster has its own CH and cluster
members. Each cluster member collects the data and forwards it to the CH. Then, all CHs with the
function of aggregating and compressing the information will process the data and then transmit the
information to the sink node.

6LQN

&OXVWHU

&OXVWHUKHDG

0HPEHUQRGH

Figure 1. Network structure.

During the deployment of the nodes, the underlying network model and the deployment of SNs
are assumed as follows:
(1) A sink node that is relatively far from the WSN is powered by manually, without considering
the energy consumption issue.
(2) The SNs and sink node are immobile.
(3) The SNs are set with the same initial energy, and each SN is assigned a unique identification in
the network.
Sensors 2019, 19, 4579 4 of 20

(4) The SNs have the ability to adjust the wireless transmission power on the basis of a
specific situation.
(5) The node senses the information of the surrounding environment and is in the state of sending
data all the time.

2.2. Energy Consumption Model

The radio hardware energy consumption model is shown in Figure 2. The radio electronics and
the power amplifier are run by the transmitter, and the radio electronics are also run by the receiver.
Both operations consume energy. The energy consumption model adopts the free space channel with
d2 energy consumption and the multipath channel with d4 energy consumption depending on the
distance between the transmitter and the receiver [16–20]. Thus, to transmit a packet of m bits at a
distance d, the energy consumption is given by the following equation:
(
m ∗ Eelec + m ∗ ε f s ∗ d2 d ≤ d0 ,
ETX = (2)
m ∗ Eelec + m ∗ ε mp ∗ d4 d > d0 ,
r .
where d0 can be calculated by d0 = ε f s ε mp . The electronic energy Eelec is used depending on
the digital coding, modulation, filtering, and spreading of the signal. The specific energy function
of receiver is the reverse of the transmitter, as depicted in Figure 2. The parameters ε f s and ε mp are
the amplification factors of the transmitting circuit when d ≤ d0 and d > d0 , where d0 is threshold.
To receive a message of m bits, the receiver expends energy as follows:

ERX = m ∗ Eelec . (3)

ETX (m, d )
d ERX (m)
m bit Transmit Electronics m bit
packet packet
spreading
digital Tx Receive
modulation filtering of the
coding Amplifier Electronics
signal

m * Eelec m * e amp * d l m * Eelec

Figure 2. Radio energy dissipation model.

3. Proposed Scheme

3.1. LEACH Protocol

The LEACH protocol plays an important role in the field of WSNs [26,27]. In the CH election
phase, the CH of every cluster compresses the data collected from the member nodes and then sends
them to the sink node [9]. The mechanism of selecting the CH is determined by the threshold value
T (n) and the rand function. The SN generates a random number M(0 ≤ M < 1). If it is satisfied with
M ≤ T (n), the node is elected as the head node of the cluster for the current round. The limit value
T (n) is calculated as follows:

p
 , n ∈ G,
T (n) = 1− p∗(r∗ mod (1/p)) (4)
0, otherwise,

where p is the ratio of the total number of CHs to SNs and represents the probability of each node
becoming CH during round 0 [4], r is the current number of rounds, G is the set of the nodes that
will not be elected as a CH in a recent 1/p round, and mod (·) denotes the modulus operator. The
power amplifier energy consumption to total energy consumption ratio versus the distance is shown
Sensors 2019, 19, 4579 5 of 20

in Figure 3. The power amplifier energy consumption to total energy consumption ratio can be
calculated by: 
m ∗ ε f s ∗ d2


m∗ Eelec +m∗ε f s ∗d2
, d ≤ d0 ,
ratio = m ∗ ε ∗ d 4 (5)
mp


m∗ E +m∗ε ∗d 4 , d > d 0 .
elec mp

The LEACH protocol adopts the concept of clustering and periodic data collection, which can
reduce the data transmission between the nodes and the BS. Therefore, this protocol can not only
reduce the energy loss, but also can extend the network lifetime. In addition, the CH uses the method
of data aggregation, which can reduce correlated data locally. This method can also optimize the
amount of data in the network and reduce energy consumption. Moreover, the time division multiple
access (TDMA) schedule used by LEACH allows the member nodes to go into sleep mode, and this
mechanism holds back the collision between clusters and extends the sensors’ battery life [28–30].
However, the density of nodes is not considered in the traditional LEACH protocol when selecting
the CH. The placement of nodes and the expected number of CHs per round are considered when
assigning CHs. Therefore, this protocol cannot ensure the uniform distribution of the CHs [31].
Additionally, the LEACH protocol does not consider the residual energy of nodes and the average
energy of all nodes when selecting the CH. This will lead to a node with a lower energy being chosen
as the CH. Thus, this protocol leads to the quick exhaustion of the node energy [31]. Finally, the CH
communicates directly with the BS by adopting a single hop communication mode. If the BS is far from
some of the CH nodes, about 80% of the energy consumption of the node comes from the power loss
of the long-distance data transmission [4]. Under the free space channel model, the power amplifier
energy consumption to total energy consumption ratio is about 80% when d ≈ 141 m. In addition,
under the multipath fading channel model, the power amplifier energy consumption to total energy
consumption ratio is about 80% when d ≈ 112 m. Therefore, looking towards real applications, it is
necessary to develop an energy-efficient protocol to decrease the energy loss of the WSN.
1
free space channel
0.9 multipath fading channel
Energy consumption ratio of amplifier

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0
0 50 100 150 200
d(m)

Figure 3. The power amplifier energy consumption to total energy consumption ratio.

3.2. Improved Energy-Efficient LEACH Protocol

To decrease the energy loss and increase the energy efficiency of the WSNs, we propose a novel
improved energy-efficient LEACH protocol called IEE-LEACH in this section.

3.2.1. Energy Consumption Model of the IEE-LEACH Protocol

The energy consumption model of the proposed IEE-LEACH protocol is introduced in this section.
Unlike the LEACH protocol, in which nodes with the same initial energy are considered, each SN in
Sensors 2019, 19, 4579 6 of 20

the IEE-LEACH protocol is assigned a different initial energy. It is assumed that there are N nodes that
are placed in an L × L region, and the proposed IEE-LEACH protocol takes two models into accounts,
i.e., the free space model and multipath model.
When the BS is close to the nodes, the energy dissipated follows the free space model (d2 power
loss) [17]. Thus, the energy consumption of the CH per round is

N N
ECH =( − 1) ∗ m ∗ Eelec + ∗ m ∗ EDA
k k (6)
+ m ∗ Eelec + m ∗ ε f s ∗ d2toBS ,

where k is the number of clusters per round in the WSN, N/k is the average node of each cluster, EDA is
the energy consumption of the CH receiving a message of 1 bit, and d2toBS denotes the expected squared
distance between the CH and BS [12]. Without the loss of generality, the node ρ( x, y) is distributed in
an arbitrary-shaped region. It is assumed that the base station is located at location ( a, b). The expected
squared distance between the CH and BS can be expressed as
Z Z
E[d2toBS ] = (( a − x )2 + (b − y)2 )ρ( x, y)dxdy (7)

( a − x )2 + ( b − y )2
Z Z
= dxdy,
A

where A is the region of distributed nodes. The energy consumption of cluster members per round is

EnonCH = m ∗ Eelec + m ∗ ε f s ∗ d2toCH , (8)

where d2toCH denotes the expected squared distance between one sensor and CH, which can be
expressed as
L2
Z Z
E[d2toCH ] = ( x2 + y2 )ρ( x, y)dxdy = . (9)
2πk
The energy consumption of each cluster is given by [12]

N
Ecluster = ECH + ( − 1) EnonCH
k
N
≈ ECH + EnonCH . (10)
k
Thus, the energy consumption with the IEE-LEACH protocol in the WSN during each round is
given by

Eround =kEcluster
=m(2NEelec + NEDA + kε f s d2toBS
+ Nε f s d2toCH ). (11)

The derivative of Eround with respect to k is equal to the optimal number of CHs, which can be
expressed as √
N L
k opt = √ . (12)
2π dtoBS

In addition, the energy dissipated follows the multipath model (d4 power loss) when the BS is far
from the nodes [21]. The energy consumption of the CH node is given by
Sensors 2019, 19, 4579 7 of 20

N N
ECH =( − 1) ∗ m ∗ Eelec + ∗ m ∗ EDA
k k (13)
+ m ∗ Eelec + m ∗ ε mp ∗ d4toBS .

The expected value of distance can be expressed as

Z Z  2
E[d4toBS ] = ( a − x )2 + (b − y)2 ρ( x, y)dxdy
 2
Z Z ( a − x )2 + ( b − y )2
= dxdy. (14)
A

In the multipath fading channel model, the EnonCH , d2toCH and Ecluster are the same as the free
space channel model mentioned above. Thus, the energy consumption of the IEE-LEACH protocol in
the WSN during each round can be expressed as

Eround =kEcluster
=m(2NEelec + NEDA + kε mp d4toBS
+ Nε f s d2toCH ). (15)

The derivative of Eround with respect to k is equal to the optimal number of CHs, which is given
by [32]
√ s
N ε fs L
k opt = √ . (16)
2π ε mp d2toBS

3.2.2. Cluster Head Selection Algorithm of the IEE-LEACH Protocol

Although there are advantages of using the LEACH protocol, it cannot guarantee the current
residual energy of the CH. The selection of the threshold T (n) used by conventional LEACH-based
protocols only consider whether the nodes will be selected to be the CH, without considering the
node’s energy. Therefore, the CHs are randomly selected, and, if a node with less energy is chosen as
CH, then they will quickly die. To balance the energy consumption and prolong the network lifetime,
we propose a new threshold T (si ), which is defined by

pi

1− pi (r mod (1/pi ))
, si ∈ G,
T ( si ) = (17)
0, otherwise,

where si is the node and i ∈ [1, N ]. The energy adjustment parameter pi is given by

p ∗ si ∗ Eri ∗ Ei
pi = , (18)
Et ∗ Ea

where p is the proportion of selecting the optimal CH, Eri is the current residual energy of the ith
node, Ei is the initial energy of the ith node, Et is the total energy of the whole network, and Ea is the
average energy of all SNs in the WSN. It can be seen from Equation (18) that the initial energy of nodes,
residual energy of nodes, total energy of the network and average energy of all nodes are all used
to calculate the energy adjustment parameter. This improvement can ensure that each node dies at
approximately the same time. Thus, the proposed protocol can balance the distribution of the energy
load among nodes and prolong the network lifetime. If the residual energy of the nodes is more than
Sensors 2019, 19, 4579 8 of 20

Ea , then an improved threshold value T (si ) can increase the possibility of selecting them as the CH.
After IEE-LEACH operates r rounds, the average energy Ea of all nodes is obtained
r
Et (1 − rmax )
Ea = . (19)
si

Initially, the nodes are randomly distributed in the WSN when the process of selecting the CH begins
to be executed. Each node generates a random number, which is compared with T (si ). If the result is
less than or equal to T (si ), then the node will become the CH.
The newly proposed algorithm can increase the probability that the node whose residual energy
is higher than the energy of the neighboring node or the average energy of the whole network becomes
the CH node in the current round. Notably, all nodes have different energies with the IEE-LEACH
protocol when selecting the CH. This process can ensure that each node dies at approximately the
same time by more frequently selecting nodes with more residual energy than nodes with less residual
energy as the CHs. Therefore, the IEE-LEACH protocol can lead to an increase in the lifetime of the
overall network.

3.2.3. Cluster Formation Algorithm of the Proposed IEE-LEACH

Once the election of CHs is completed, the CHs will inform the other nodes of the information
that they have become the CH in this round. To complete this function, each CH node will send
an advertisement message to all the other nodes in the form of broadcast with a non-persistent
carrier-sense multiple access (CSMA) MAC protocol [33].
Each member node decides whether to participate in cluster formation, according to the signal
intensity of the message that is transmitted from the BS and each CH node. Since the symmetric
propagation channel model is used by the proposed IEE-LEACH protocol, the election of a cluster
head is only related to the pure signal strength [15]. The stronger the received signal is, the closer the
distance between member nodes to the sink node or the CH is. Comparing the distance of the nodes
to the CHs and that of the nodes to the BS, the nodes closer to the base station do not participate in
cluster formation and directly send data information to the BS.
As shown in Figure 4, when the cluster head in the network is elected and the cluster head
broadcasts messages to other nodes, the non-cluster head node A finds the nearest cluster head B
according to the received signal strength, and calculates the distance between them, i.e., d AB . When
d AB > d AtoBS (d AtoBS denotes the distance between the node A and BS), node A does not join any
cluster and communicates directly with the base station during the data communication. When
d AB < d AtoBS , node A joins the cluster where cluster head B is located and acts as a member node in
the cluster.
This means that not all nodes in the WSN participate in cluster formation. A node closer to the
CH decides to join the cluster whose CH transmits the strongest signal to the nodes by comparing all
the signal intensities of the messages sent by the CHs. After each node closer to the CH determines
which CH will join, it needs to inform the CH that it has become a member node of this cluster. Each
node closer to the CH sends a join-request message (Join-REQ) to the chosen CH, which includes the
node’s identification and the CH’s identification.
Figure 5 shows the process of CH selection and cluster formation of the proposed IEE-LEACH
protocol. First, the related energy and the distance to the BS are calculated. Then, the threshold
T (si ) in each round is determined by Equation (17). Each node closer to the CH needs to generate a
random number and then compare it with T (si ). If the result is less than or equal to T (si ), then the
node will be chosen as the CH. The distance to CH is calculated. The nodes whose distance is closer
to the BS rather than the CH will not take part in the process of cluster formation and will directly
transmit data to the BS. Finally, the transmitted data adopt the method of single hop, multi-hop and
hybrid communications.
Sensors 2019, 19, 4579 9 of 20

d AB
Member B
nodes

Cluster d AtoBS
head

Base BS
station

Figure 4. Cluster formation.

Start

Initialization

Calculate d toBS

N
i£r
Y
Calculate T ( si )

N
t £ T ( si )

Y
CH ¬ ni

Calculate d toCH

d toCH < dtoBS

N
Y
Select nearby CH
Do not participate

Transfer date to BS

i = i +1

End

Figure 5. Block diagram of proposed protocol.

3.2.4. Data Transmission Scheme of the IEE-LEACH Protocol

To decrease energy consumption, the radio network of the CH and nodes is closed until the
transmission slot is assigned. The radio network of the CH and nodes that intend to send the data
need to be activated. After activating the nodes, non-CH nodes send the data to the CH, while the
CH receives these data and sends them to the BS [34,35]. A single data communication approach will
increase the energy consumption of the nodes because the best way to transfer data are to minimize the
distance between nodes. Therefore, we propose a methodology that provides a single hop, multi-hop
and hybrid communication network to minimize the distance.
It is assumed that the cluster head node transmits data to BS by (n − 1) hops and the distance of
each hop is r in the wireless sensor network with the free-space channel model. Therefore, to transmit
an m-bit message, the energy consumption for signal hop is given by
Sensors 2019, 19, 4579 10 of 20

E1 = ETX (m, n × r )
= m × Eelec + m × ε f s × (nr )2
 
= m × Eelec + ε f s × n2 × r2 . (20)

The energy consumption for multi-hop can be expressed as

E2 = n × ETX (m, r ) + (n − 1) × ERX (m)

 
= n × m × Eelec + ε f s r2 + (n − 1) × Eelec × m
h i
= m × (2n − 1) Eelec + ε f s × n × r2 . (21)

Comparing the energy consumption of two communication modes, if the energy consumption of
a multi-hop is less than that of a single hop, i.e.,
h i
m × (2n − 1) Eelec + ε f s × n × r2
 
< m × Eelec + ε f s × n2 × r2 . (22)

Then, it can be devised as

s
2Eelec
r> . (23)
nε f s

Note that Eelec = 50 nJ/bit and ε f s = 10 pJ/bit/m2 , when n = 2, it can be found that r >
70 m. Therefore, the energy consumption of multi-hop communication is less than that of single hop
communication when r > 70 m. This mechanism is beneficial to the average energy dissipation of the
network and prolongs the lifetime of network.
The Algorithm 1 of the proposed protocol is shown below:

Algorithm 1. IEE-LEACH
1. N-number of nodes, r-number of rounds.
2. The first step is to initialize the network parameters for WSN.
3. The second step is the random deployment of the nodes in the network.
4. The third step is the deployment of the BS in the network.
5. Calculate dtoBS
6. for i = 1 to r
r
7. Ea = Et (1 − rmax )/n
i
8. pi = p ∗ n ∗ Er ∗ Ei /( Et ∗ Ea )
9. T (si ) = pi /(1 − pi (r mod (1/pi )))
10. t=Random number
11. If (t ≤ T (si ))
12. CH ← ni
13. Calculate dtoCH
14. end if
15. if (dtoCH < dtoBS )
16. Select CH and join the cluster;
17. else
18. Nodes do not participate in the cluster
19. end if
20. Transfer to BS
21. i=i+1
22. go to step 6
23. end
Sensors 2019, 19, 4579 11 of 20

The conventional LEACH-based protocols do not consider the energy adjustment parameter,
which exhausts the node energy quickly. However, the proposed IEE-LEACH protocol introduces
four energy parameters when selecting the threshold, i.e., initial energy of nodes, residual energy of
nodes, total energy of the network, and average energy of all nodes. This mechanism can balance the
energy distribution of all nodes and improve energy efficiency. In addition, to decrease the energy
consumption, the node closer to the BS than the CH will not participate in the cluster formation.
Therefore, the robustness of the network is improved and the lifetime of the network is increased.
Furthermore, the proposed protocol compares the energy consumption of single hop and multi-hop
communication modes in data transmission. Then, the communication mode with less energy
consumption will be adopted. Since the distance of a single hop is less than 70 m as mentioned
above, the single hop will be adopted when the distance between CH and BS is less than 70 m.
Otherwise, the multi-hop method will be adopted. This improvement is beneficial for the average
energy dissipation that enhances the lifetime of the sensor network. Thus, the proposed IEE-LEACH
protocol shows flexibility over conventional LEACH protocols in terms of selecting cluster heads,
cluster formation, and the data transmission.

4. Simulation Results
In this section, to evaluate the performance of the proposed IEE-LEACH protocol in the
WSN, the existing LEACH, LEACH-Centralized (LEACH-C) [18], LEACH-M [22], O-LEACH [23],
energy-efficient LEACH (EE-LEACH) [36], Stable Energy Efficient Network (SEEN) [24], and
LEACH-eXtended Message-Passing (LEACH-XMP) [25]protocols are used in comparison in the
simulation experiments. Simulation experiments were carried out using MATLAB 2014a (MathWorks,
Natick, MA, USA). A wireless sensor network system model consisting of 100 homogeneous SNs [37]
randomly distributed in 100 m × 100 m square area with different initial energies is considered. The
BS is√placed at the coordinate (50, 100). Therefore, the maximum distance between the SN and the BS
is 50 5m (approximately 111.8 m). We run the simulation for 3500 rounds, and the value of the time
interval between rounds is 20 s [4]. Table 1 shows the parameter configurations in the WSN that are
employed in the simulation.
Table 1. Simulation parameters.

Parameters Values
Eelec 50 nJ/bit
EDA 5 nJ/bit/signal
Transmitter Amplifier (ε fs ) if d ≤ d0 10 pJ/bit/m2
Transmitter Amplifier (ε mp ) if d ≥ d0 0.0013 pJ/bit/m4
p 0.05
d0 87 m
Data Packet Size 4000 bits
Data Packet rate 1 packet/s

The performance metrics of the WSN that are used to evaluate the performance of the algorithm
as follows [38,39]:

• Stability Period: The time span from the beginning of network operation until the first dead
sensor occurs, which can be calculated by

Tstability = t FND − tstart , (24)

where t FND = 20 × FND, FND is the number of rounds after which the first sensor died, t FND
is the time that the first dead sensor occurred, and tstart is the time of the beginning of network
operation.
Sensors 2019, 19, 4579 12 of 20

• Network Lifetime: The time span from the beginning of network operation until the last live SN
dies, which can be calculated by

Tli f etime = t LDN − tstart , (25)

where t LND = 20 × LND,LND is the number of rounds after which all sensor nodes died, and
t LND is the time at which all sensor nodes died.
• The Amount of Transmitted Data: The total quantity of data sent from the nodes (including the
CHs and other nodes) in the WSN to the BS can be calculated by

data = dataCHtoBS + data NtoBS , (26)

where dataCHtoBS is the amount of transmitted data by the CHs to the BS, and data NtoBS is the
amount of transmitted data by nodes to BS. In particular, in this paper, data NtoBS concerns nodes
near the station of the base, and which does not contribute to the formation of clusters.
• Energy Consumption of Network: The energy consumption of wireless sensor networks mainly
includes circuit energy consumption and power amplifier energy consumption, and the latter
is dominant. The energy consumption of each round in the network can be calculated by
Equation (11) or Equation (15) in different channel models.
• Number of CHs per Round: The number of nodes that send information aggregated from their
cluster members directly to the BS is elected according to the threshold condition T (si ). This can
be calculated by Equation (12) or Equation (16) in different channel models.

In terms of the reliability and lifetime of the WSN, the stability period of the WSN is of great
value to research because the network is mainly responsible for data transmission. Furthermore, more
attention is paid to the lifetime of the WSN, the number of data transmissions (to the BS), and the
residual node energy, which objectively reflects the network lifetime situation.

4.1. Network Lifetime and Stability Period

This simulation illustrates the network lifetime of the proposed IEE-LEACH protocol compared
with the other existing protocols. The number of dead nodes versus the number of rounds in the WSNs
is shown in Figure 6. It is obvious that the nodes start to die after 1678 rounds, 1933 rounds, 2178 rounds,
1836 rounds, 2326 rounds, and 2416 rounds for LEACH, LEACH-M, O-LEACH, LEACH-C, EE-LEACH,
and SEEN protocols, respectively. However, for the proposed IEE-LEACH, the nodes do not start to
die until 2636 rounds. In addition, the number of rounds for all the nodes dead in the IEE-LEACH
protocol is 3160 rounds as compared to 2245 rounds in the LEACH protocol. From the obtained
results, the IEE-LEACH improves the network lifetime by approximately 9% to 57% compared with
the other protocols.
100
SEEN
90 LEACH
LEACH−C
80
O−LEACH
70 LEACH−M
Number of nodes dead

EE−LEACH
60 IEE−LEACH

50

40

30

20

10

0
0 500 1000 1500 2000 2500 3000 3500
Number of Rounds

Figure 6. Comparison of life cycle.

Sensors 2019, 19, 4579 13 of 20

4.2. Residual Nodes Energy

In this test, the effect of the residual nodes energy on the performance is shown. The residual
energy of all SNs in the network for the protocols is illustrated in Figure 7. The proposed IEE-LEACH
exhibits a slower decrease as the number of rounds increase than a variety of routing protocols. All
nodes in the existing LEACH, LEACH-M, O-LEACH, LEACH-C, EE-LEACH, and SEEN protocols
are dead after 2267, 2492, 2972, 2422, 3053, and 2889 rounds, respectively. However, all nodes in the
proposed IEE-LEACH are dead after 3242 rounds. It can be seen that the new proposed threshold that
considers the initial energy of nodes, residual energy of nodes, total energy of the network, and average
energy of all nodes increases the network lifetime by about 6% to 43% compared with other routing
protocols. Therefore, the strategy presented by the proposed IEE-LEACH protocol can effectively
reduce the dissipated energy and extend the lifetime of the network.

100
SEEN
90 LEACH
LEACH−C
80 O−LEACH
Average residual energy of nodes(J)

LEACH−M
EE−LEACH
70
IEE−LEACH

60

50

40

30

20

10

0
0 500 1000 1500 2000 2500 3000 3500
Number of Rounds

Figure 7. Comparison of residual energy of nodes.

4.3. The Amount of Transmitted Data

In this experiment, we employed the same data aggregation technique and the same number
of information bytes. The amount of transmitted data is determined by the number of alive nodes
in the entire network, i.e., it is related to the life cycle of nodes. In this section, several tests are
performed to evaluate the proposed protocol under different conditions, as displayed in Table 2.
Figure 8 depicts the throughput of protocols against the simulation rounds. Since the nodes closer to
the BS do not participate in cluster formation and the energy adjustment parameter is used in setting
the threshold, it is obvious that the proposed IEE-LEACH protocol is superior to other protocols.
Considering that the nodes closer to the BS do not participate in cluster formation, the IEE-LEACH-A
improves the throughput of the network; thus, the information transmission performance of the
IEE-LEACH-A protocol is better than that of the LEACH protocol. Moreover, the network throughput
of the IEE-LEACH-B is larger than that of the IEE-LEACH-A, which illustrates that, in terms of
improving the network throughput, the improvement by considering the energy adjustment parameter
when setting the threshold is more energy-efficient than when considering the nodes closer to BS
which do not participate in cluster formation.
Compared with the LEACH protocol, the IEE-LEACH increases throughput by nearly a factor of
7. Specifically, this method greatly improves the information transmission performance of the network.
In addition, compared with the IEE-LEACH-B, the IEE-LEACH has an approximately 30% increase in
network throughput. In terms of the amount of transmitted data, the proposed IEE-LEACH protocol
sends much more data to the BS in the same round than other protocols.
Sensors 2019, 19, 4579 14 of 20

Table 2. Comparison of the improvement schemes of the proposed routing protocols.

Protocol Name Improved Details

LEACH NULL
Threshold setting considering the energy adjustment parameter,
IEE-LEACH
and closer nodes to the BS do not participate in cluster formation.
Threshold setting in the same way as LEACH,
IEE-LEACH-A
but closer nodes to the BS do not participate in cluster formation.
Threshold setting considering the energy adjustment parameter,
IEE-LEACH-B
but the cluster formation is the same as that of LEACH.

4
x 10
14
LEACH
IEE−LEACH−A
12 IEE−LEACH−B
Number of received data at base station

IEE−LEACH
10

0
0 500 1000 1500 2000 2500 3000 3500
Number of rounds

Figure 8. Comparison of data transmission.

4.4. Comparison of the Average Energy Mean Square Deviation of Nodes

After 10 rounds, we calculate the average energy mean square deviation of the remaining nodes
to analyze the energy distribution in the whole network. We can obtain the optimum number of
CHs depending on the analysis of the average energy mean square deviation of nodes to transmit
information and increase the performance of the entire network. Figure 9 shows that the slope of the
LEACH is larger than the proposed protocol before the degree of energy dispersion increases to the
maximum. However, the volatility of the proposed IEE-LEACH is less than 0.1. This finding implies
that the performance of the proposed IEE-LEACH protocol is better than other protocols in terms of
the energy consumption. This improvement is because the proposed IEE-LEACH protocol extends the
lifetime of the network by balancing the node energy.

0.45
LEACH
Average energy mean square deviation of nodes

0.4 IEE−LEACH−A
IEE−LEACH−B
0.35 IEE−LEACH

0.3

0.25

0.2

0.15

0.1

0.05

0
0 500 1000 1500 2000 2500 3000 3500
Number of rounds

Figure 9. Comparison of average energy mean square deviation of nodes every 10 rounds.
Sensors 2019, 19, 4579 15 of 20

4.5. Comparison of the Number of Cluster Heads

To demonstrate that it is necessary to consider the improvement when nodes closer to the BS
do not participate in clustering formation, the simulation results of the cluster heads number for
the IEE-LEACH-B and IEE-LEACH protocols are provided in this section. Unlike the LEACH and
IEE-LEACH-A protocols, IEE-LEACH-B and IEE-LEACH consider the energy when selecting the CH.
Thus, IEE-LEACH-B and IEE-LEACH can select the optimal number of cluster heads.
Figure 10 shows the number of cluster heads versus the number of rounds. The maximum number
of cluster heads for IEE-LEACH occurs after approximately 700 rounds, and the number of cluster
heads is zero after 2600 rounds. However, the maximum number of cluster heads for IEE-LEACH-B
occurs in approximately 500 rounds, and the number of cluster heads is zero after 2300 rounds. We
find that the number of cluster heads extends up to 2600 rounds for IEE-LEACH as compared to
2300 rounds in IEE-LEACH-B, which is due to the improvement by which the nodes closer to BS
do not participate in the cluster formation. The number of cluster heads for IEE-LEACH-A and
LEACH disappears after 2050 rounds and 1800 rounds, respectively. It is obvious that IEE-LEACH
and IEE-LEACH-B prolong the network lifetime compared with IEE-LEACH-A and LEACH because
IEE-LEACH-B and IEE-LEACH consider the energy adjustment parameter when selecting the CH.
In addition, the proposed IEE-LEACH protocol generates more CHs than IEE-LEACH-B throughout
the 3500 simulation rounds, which can reduce the energy consumption of the WSN nodes. Thus, the
proposed scheme is efficient in terms of saving energy and extending the lifetime of the WSN.
25
LEACH
IEE−LEACH−A
IEE−LEACH−B
20 IEE−LEACH
Number of cluster heads

15

10

0
0 500 1000 1500 2000 2500 3000 3500
Number of rounds

Figure 10. Comparison of the number of cluster heads.

4.6. Comparison of Total Energy Consumption of the Network

To prove that the performance of the proposed protocol is better than that of the existing protocols,
the total energy consumption of the proposed protocol is compared with that of LEACH, LEACH-M,
O-LEACH, LEACH-C [18], EE-LEACH [36] and SEEN, under the same simulation conditions. Figure 11
shows the total network energy consumption of the protocols versus the number of rounds. As shown
in Figure 11, the LEACH-C protocol optimizes the cluster head selection and consumes less network
energy than the LEACH protocol in the same round. The LEACH-C prolongs the network lifetime
to 2723 rounds as compared to 2267 rounds of the LEACH protocol. In addition, the LEACH-M and
O-LEACH prolong the network lifetime to 2775 and 2972 rounds, respectively. Because the residual
energy of nodes is taken into account when selecting cluster heads and the multi-hop communication
mode is applied in EE-LEACH, the EE-LEACH further prolongs the network lifetime to 3114 rounds.
The proposed IEE-LEACH protocol not only considers the residual energy and initial energy of nodes
when selecting cluster heads, but also considers the average energy of all nodes and the total energy of
network to further optimize cluster head selection. Thus, the total network energy consumption of
the IEE-LEACH protocol is less than that of the other protocols in the same round, and the proposed
IEE-LEACH protocol prolongs the network lifetime to 3242 rounds. Moreover, because single hop,
multi-hop and hybrid communication modes are adopted, and the nodes closer to BS than to the CH
Sensors 2019, 19, 4579 16 of 20

do not participate in cluster formation, the proposed IEE-LEACH protocol further extends the lifetime
of the network.

100

90

Total Energy Consumption of the Network (J)

80

70

60

50

40 SEEN
LEACH
30 LEACH−C
O−LEACH
20
LEACH−M
EE−LEACH
10
IEE−LEACH
0
0 500 1000 1500 2000 2500 3000 3500
Number of Rounds

Figure 11. Comparison of total energy consumption of the network.

4.7. Performance for Large Sensor Field

To further demonstrate that the performance of the proposed IEE-LEACH protocol is still
prominent in large scale, we construct a WSN with randomly deployed 500 sensor nodes in a
500 m × 500 m sensor field. It is clearly from Figure 12 that, compared with the existing routing
protocols, the network lifetime is enhanced by the proposed IEE-LEACH protocol as the number
of rounds increases. Figures 13 and 14 display the chart view of residual energy and total energy
consumption in the network. All nodes in the existing LEACH, EE-LEACH, SEEN and LEACH-XMP
protocols are dead after 1328, 1534, 1731 and 2724 rounds, respectively. However, all nodes in the
proposed IEE-LEACH are dead after 2692 rounds. It can be seen that the new proposed scheme which
considers the initial energy of nodes, residual energy of nodes, total energy of the network, and average
energy of all nodes increases the network lifetime by about 5% to 75% compared with different routing
protocols. Therefore, it is apparently observed that the proposed IEE-LEACH algorithm performs
equally well in case of large network size and is scalable enough for large network areas.
As indicated in Tables 3 and 4, among the different routing protocols, our protocol can achieve
longer network lifetime and lower energy consumption than the others in various scale sensor fields,
which is of great significance for diverse practical applications.

500

450

400

350
Number of nodes dead

300

250

200

150
LEACH−XMP
100 EE−LEACH
SEEN
50 LEACH
IEE−LEACH
0
0 500 1000 1500 2000 2500 3000 3500 4000
Number of Rounds

Figure 12. Comparison of life cycles for a large sensor field.

Sensors 2019, 19, 4579 17 of 20

500
LEACH
450 EE−LEACH
SEEN
LEACH−XMP
400

Average residual energy of nodes(J)

IEE−LEACH

350

300

250

200

150

100

50

0
0 500 1000 1500 2000 2500 3000 3500 4000
Number of Rounds

Figure 13. Comparison of residual energy of nodes for a large sensor field.

500
LEACH
EE−LEACH
450
SEEN
LEACH−XMP
Total Energy Consumption of the Network (J)

400 IEE−LEACH

350

300

250

200

150

100

50

0
0 500 1000 1500 2000 2500 3000 3500 4000
Number of Rounds

Figure 14. Comparison of total energy consumption of the network for a large sensor field.

Table 3. A performance comparison between different protocols in a 100 m × 100 m sensor field.

Protocol Lifetime Residual Energy Total Energy Consumption

LEACH 1678 2267 2267
EE-LEACH 2326 3053 3114
SEEN 2416 2889 2990
LEACH-C 1836 2422 2723
O-LEACH 2178 2972 2972
LEACH-M 1933 2492 2775
IEE-LEACH 2636 3242 3242

Table 4. A performance comparison between different protocols in a 500 m × 500 m sensor field.

Protocol Network Lifetime Residual Energy Total Energy Consumption

LEACH 1361 1328 1411
EE-LEACH 1719 1534 1589
SEEN 1812 1731 1826
LEACH-XMP 3313 2724 2785
IEE-LEACH 3491 2692 2793
Sensors 2019, 19, 4579 18 of 20

5. Conclusions
In this paper, a novel clustering protocol, named IEE-LEACH, is proposed to reduce energy
consumption and improve the lifetime of WSNs. Compared with the existing routing protocols,
the threshold of the proposed IEE-LEACH protocol introduces four parameters: the initial energy
of nodes, residual energy of nodes, total energy of the network and average energy of all nodes.
This mechanism can improve the robustness of the network and extend the network lifetime. In
addition, the proposed protocol can optimize the number of CHs and their distributions, which can
effectively reduce the energy consumption. Furthermore, to decrease the energy consumption, we
consider that the nodes closer to the BS do not participate in cluster formation. Moreover, the proposed
protocol employs single hop, multi-hop, and hybrid communications instead of a single communication
mode in data transmission. Therefore, the proposed approach decreases the overall communication
cost and significantly improves the network lifetime. The simulation results demonstrate that the
proposed IEE-LEACH protocol has a better energy consumption distribution and is more reliable and
energy-efficient than some existing protocols.

Author Contributions: The algorithms proposed in this paper have been conceived by Y.L., Q.W., T.Z. and Y.T.;
designed the experiments, Y.L., Y.T., F.B. and M.J.; performed the experiments and analyzed the results, Q.W., T.Z.
and F.B.; wrote the paper, Y.L., Y.T. and Q.W.
Funding: The authors would like to acknowledge the National Natural Science Foundation of China for its
support of this research. This work was supported by the National Science Foundation of China under Grant Nos.
61761033 and 61461036, and in part by the Natural Science Foundation of Neimongol Autonomous Region of
China under Grant No. 2019MS06033.
Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations
The following abbreviations are used in this manuscript:

WSNs Wireless sensor networks

LEACH Low-energy adaptive clustering hierarchy
CHs Cluster heads
IEE-LEACH Improved energy-efficient LEACH
BS Base station
SNs Sensor nodes
SEECH Scalable energy-efficient clustering hierarchy
LA-MHR Learning automata-based multilevel heterogeneous routing
LEACH-M LEACH-Mobile
O-LEACH Optical LEACH
DFS Distributed fiber sensors
SEEN Stable energy efficient network
LEACH-XMP LEACH-eXtended Message-Passing
TDMA Time division multiple access
CSMA Carrier-sense multiple access
LEACH-C LEACH-Centralized
EE-LEACH Energy-efficient LEACH

References
1. Zhang, D.; Li, G.; Zheng, K.; Ming, X.; Pan, Z. An energy-balanced routing method based on forward-aware
factor for wireless sensor networks. IEEE Trans. Ind. Inform.. 2014, 10, 766–773. [CrossRef]
2. Sobrinho, J.L. Correctness of routing vector protocols as a property of network cycles. IEEE Trans. Netw.
2017, 25, 150–163. [CrossRef]
3. Mouapi, A.; Hakem, N. A new approach to design autonomous wireless sensor node based on RF energy
harvesting system. Sensors 2018, 18, 133. [CrossRef] [PubMed]
Sensors 2019, 19, 4579 19 of 20

4. Zhang, Y.; Liu, M.; Liu, Q. An energy-balanced clustering protocol based on an improved CFSFDP algorithm
for wireless sensor networks. Sensors 2018, 18, 881. [CrossRef] [PubMed]
5. Bahbahani, M.S.; Alsusa, E. A cooperative clustering protocol with duty cycling for energy harvesting
enabled wireless sensor networks. IEEE Trans. Wirel. Commun. 2018, 17, 101–111. [CrossRef]
6. Li, X.F.; Xu, L.Z.; Wang, H.B.; Song, J.; Yang, S.X. A differential evolution-based routing algorithm for
environmental monitoring wireless sensor networks. Sensors 2010, 10, 5425–5442. [CrossRef] [PubMed]
7. Ma, T.; Hempel, M.; Peng, D.; Sharif, H. A survey of energy-efficient compression and communication
techniques for multimedia in resource constrained systems. IEEE Commun. Surv. Tutor. 2013, 15, 963–972.
[CrossRef]
8. Shen, J.; Wang, A.; Wang, C.; Hung, P.C.K.; Lai, C. An efficient centroid-based routing protocol for energy
management in WSN-assisted IoT. IEEE Access 2017, 5, 18469–18479. [CrossRef]
9. Sohn, I.; Lee, J.; Lee, S.H. Low-energy adaptive clustering hierarchy using affinity propagation for wireless
sensor networks. IEEE Commun. Lett. 2016, 20, 558–561. [CrossRef]
10. Zhao, Z.; Xu, K.; Hui, G.; Hu, L. An energy-efficient clustering routing protocol for wireless sensor setworks
based on AGNES with balanced energy consumption optimization. Sensors 2018, 18, 3938. [CrossRef]
11. Heinzelman, W.R.; Chandrakasan, A.; Balakrishnan, H. Energy-efficient communication protocol for wireless
microsensor networks. In Proceedings of the 33rd Annual Hawaii International Conference on System
Sciences (HICSS), Maui, HI, USA, 7 January 2000; pp. 1–10.
12. Roy, N.R.; Chandra, P. A note on optimum cluster estimation in LEACH protocol. IEEE Access 2018, 6,
65690–65696. [CrossRef]
13. Hosen, A.; Cho, G. An energy centric cluster-based routing protocol for wireless sensor networks. Sensors
2018, 18, 1520. [CrossRef] [PubMed]
14. Sharma, D.; Bhondekar, A.P. Traffic and energy aware routing for heterogeneous wireless sensor networks.
IEEE Commun. Lett. 2018, 22, 1608–1611. [CrossRef]
15. Kaur, T.; Kumar, D. Particle swarm optimization-based unequal and fault tolerant clustering protocol for
wireless sensor networks. IEEE Sens. J. 2018, 18, 4614–4622. [CrossRef]
16. Behera, T.M.; Samal, U.C.; Mohapatra, S.K. Energy-efficient modified LEACH protocol for IoT application.
IET Wirel. Sens. Syst. 2018, 8, 223–228. [CrossRef]
17. Obaidat, M.S.; Misra, S. Principles of Wireless Sensor Networks; Cambridge Univ. Press: Cambridge, UK, 2014.
18. Alnawafa, E.; Marghescu, I. New energy efficient multi-hop routing techniques for wireless sensor networks:
Static and dynamic techniques. Sensors 2018, 18, 1863. [CrossRef] [PubMed]
19. Jadoon, R.; Zhou, W.; Jadoon, W.; Ahmed Khan, I. RARZ: Ring-zone based routing protocol for wireless
sensor networks. Appl. Sci. 2018, 8, 1023. [CrossRef]
20. Tarhani, M.; Kavian, Y.S.; Siavoshi, S. SEECH: Scalable energy efficient clustering hierarchy protocol in
wireless sensor networks. IEEE Sens. J. 2014, 14, 3944–3954. [CrossRef]
21. Tanwar, S.; Tyagi, S.; Kumar, N.; Obaidat, M.S. LA-MHR: Learning automata based multilevel heterogeneous
routing for opportunistic shared spectrum access to enhance lifetime of WSN. IEEE Syst. J. 2019, 13, 313–323.
[CrossRef]
22. Kim, D.S.; Chung, Y.J. Self-organization routing protocol supporting mobile nodes for wireless sensor
network. In Proceedings of the First International Multi-Symposiums on Computer and Computational
Sciences (IMSCCS’06), Hangzhou, China, 20–24 June 2006; pp. 622–626.
23. Yan, L.; Pan, W.; Luo, B.; Li, X.; Liu, J. Modified energy-efficient protocol for wireless sensor networks in the
presence of distributed optical fiber senor link. IEEE Sens. J. 2011, 11, 1815–1819. [CrossRef]
24. Priyadarshi, R.; Singh, L.; Singh, A.; Thakur, A. SEEN: Stable energy efficient network for wireless sensor
network. In Proceedings of the 2018 5th International Conference on Signal Processing and Integrated
Networks (SPIN), Noida, India, 22–23 February 2018; pp. 338–342.
25. Kang, J.; Sohn, I.; Lee, S. Enhanced message-passing based LEACH protocol for wireless sensor networks.
Sensors 2019, 19, 75. [CrossRef] [PubMed]
26. Heinzelman, W.B.; Chandrakasan, A.P.; Balakrishnan, H. An application-specific protocol architecture for
wireless microsensor networks. IEEE Trans. Wirel. Commun. 2002, 1, 660–670. [CrossRef]
27. Tümer, A.E.; Gündüz, M. Energy-efficient and fast data gathering protocols for indoor wireless sensor
networks. Sensors 2010, 10, 8054–8069. [CrossRef] [PubMed]
Sensors 2019, 19, 4579 20 of 20

28. Cheng, H.J.; Su, Z.H.; Xiong, N.X.; Xiao, Y. Energy-efficient nodes scheduling algorithms for wireless sensor
networks using Markov random field model. Inf. Sci. 2016, 329, 461–477. [CrossRef]
29. Hoang, D.C.; Yadav, P.; Kumar, R.; Panda, S.K. Real-time implementation of a harmony search
algorithm-based clustering protocol for energy-efficient wireless sensor networks. IEEE Trans. Ind. Inf. 2014,
10, 774–783. [CrossRef]
30. Noh, Y.; Lee, D. BCoPS: An energy-efficient routing protocol with coverage preservation. IET Commun. 2017,
11, 1933–1940. [CrossRef]
31. Miao, H.; Xiao, X.; Qi, B.; Wang, K. Improvement and application of LEACH protocol based on genetic
algorithm for WSN. In Proceedings of the 2015 IEEE 20th International Workshop on Computer Aided
Modelling and Design of Communication Links and Networks (CAMAD), Guildford, UK, 7–9 September
2015; pp. 242–245.
32. Singh, S.K.; Kumar, P.; Singh, J. P. A survey on successors of LEACH protocol. IEEE Access 2017, 5, 4298–4328.
[CrossRef]
33. Zhang, W.; Wei, X.; Han, G.; Tan, X. An energy-efficient ring cross-layer optimization algorithm for wireless
sensor networks. IEEE Access 2018, 6, 16588–16598. [CrossRef]
34. Zheng, H.F.; Guo, W.Z.; Xiong, N. A kernel-based compressive sensing approach for mobile data gathering
in wireless sensor network systems. IEEE Trans. Syst. Man Cybern. Syst. 2017, 48, 2315–2327. [CrossRef]
35. Cheng, H.J.; Chen, Y.Z.; Xiong, N.X.; Li, F.F. Layer-based data aggregation and performance analysis in
wireless sensor networks. J. Appl. Math. 2013, 2013, 502381. [CrossRef]
36. Arumugam, G.S.; Ponnuchamy, T. EE-Leach: Development of energy-efficient LEACH protocol for data
gathering in WSN. EURASIP J. Wirel. Commun. Netw. 2015, 2015, 76. [CrossRef]
37. Kong, H.Y. Energy efficient cooperative LEACH protocol for wireless sensor networks. J. Commun. Netw.
2010, 12, 358–365.
38. Mottaghi, S.; Zahabi, M.R. Optimizing LEACH clustering algorithm with mobile sink and rendezvous nodes.
AEU-Int. J. Electron. Commun. 2015, 69, 507–514. [CrossRef]
39. Sandhu, M.M.; Akbar, M.; Behzad, M.; Javaid, N.; Khan, Z.A.; Qasim, U. Mobility model for WBANs. In
Proceedings of the 9th International Conference on Broadband and Wireless Computing, Communication
and Applications, Guangzhou, China, 8–10 November 2014; pp. 155–160.

c 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).