Clustering Based Routing for Wireless Sensor Networks in Smart Grid Environment

Full text

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
DOI:10.5121/ijassn.2018.8301 1
C
LUSTERING
-B
ASED
R
OUTING FOR
W
IRELESS
S
ENSOR
N
ETWORKS
I
N
S
MART
G
RID
E
NVIRONMENT
Laila Nassef
1,2
, Reemah El-Habshi
1
and Linta Jose
1
1
Department of Computer Science, Faculty of Computing and Information Technology,
King Abdulaziz University, Jeddah, Saudi Arabia
2
Department of Computer Sciences and Information, Institute of Statistical Studies and
Research, Cairo University, Egypt
A
BSTRACT
Wireless Sensor Networks (WSN) is widely deployed in different fields of applications of smart grid to
provide reliable monitoring and controlling of the electric power grid. The objective of this paper is
simulate and analyze impact of various smart grid environments on performance of four different WSN
routing protocols namely the Low Energy Adaptive Clustering Hierarchy (LEACH) and Centralized
LEACH (LEACT-C) as well as other two conventional protocols namely Minimum Transmission Energy
(MTE) and Static Clustering. This analysis would be beneficial in making the correct choice of WSN
routing protocols for various smart grid applications. The performance of the four protocols is simulated
using NS-2 network simulation on Ubuntu. The results are analyzed and compared using number of data
signals received at base station, energy consumption, and network lifetime as performance metrics. The
results show that the performance of various protocols in the smart grid environments have deteriorated
due log normal channel characteristics and consequently network lifetime have decreased significantly.
The results also indicate that clustering based routing protocols have more advantageous over
conventional protocols; MTE and static clustering. Also, centralized clustering approach is more effective
as it distributes energy dissipation evenly throughout the sensor nodes which reduce energy consumption
and prolong the networks’ lifetime. This approach is more effective in delivering data to base station
because it has global knowledge of the location and energy of all the nodes in the network.
K
EYWORDS
Wireless Sensor Networks; Clustering; Energy Consumption, Network Lifetime.
1.
I
NTRODUCTION
The current power grid suffers from lack of effective communications, monitoring, fault
diagnostics, and automation, which increases the possibility of region wide system breakdown.
Smart grid [1] is a new generation of electric power network that modernizes electric power grid
network using advanced sensors, and distributed computing technologies to improve the
efficiency, reliability and safety of power delivery. Electric power grid consists of three main
subsystems: power generation, power transmission and distribution, and customer facilities.
Smart grid needs online monitoring, diagnostics and protection to ensure better control by
incorporating automation and self-healing capabilities. Recently, Wireless Sensor Networks
(WSNs) [1]have been recognized as a promising technology to achieve seamless, energy
efficient, reliable and low cost monitoring and control of the smart grid. WSNs are applied
through the three main subsystems of the smart grid. Though WSNs brings about many
advantages to the smart grid technology, it also brings up many challenges because of unique
characteristics, resource constraints, and the harsh and complex electric-power environment.

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
2
Therefore this paper investigate performance of various cluster based routing protocols in
different smart grid environments.
Sensors are small, inexpensive, self powered devices that can sense and communicate with each
other for the purpose of collecting local information to make global decisions about their physical
environment [2]. Information is aggregated to a local processing and control system, which in
turn can communicate with a remote system using any of the various external networks available.
In WSNs, nodes can send their data to the Base Station (BS) using routing protocols that are
broken down into three categories. First, direct communication (DC) [3], which is the simplest
protocol, where sensor nodes send data directly to the BS. In direct communication protocol,
sensor nodes, which are far from the BS, dissipate faster than others do because they send their
data to the BS directly. The second category involve Minimum Transmission Energy (MTE)
protocols [4], where nodes route data to the BS through intermediate nodes, each node acting as a
router for the other nodes [5].
The limited energy supplies of WSN nodes, unreliable communication due to the wireless
medium, and the need for self-configuration impose constraints on network’s lifetime. These
constraints along with limited computational and memory storage represent a challenge to keep
network’s protocols as light weight as possible [6]. These energy constraints necessitate the need
for energy efficiency routing protocols and various energy efficient protocols have been proposed
to solve problems in conventional routing protocols. One of the energy efficient techniques to
extend the lifetime of WSNs is clustering [7]. Therefore, the third category are made up of
clustering protocols, where sensor nodes are organized into independent clusters, each with a
Cluster Head (CH) node and a number of member nodes to collect data from their environment
and forward it to the CH. CH collect and aggregate information from sensors in their own cluster
and forward information to the BS, which collects and processes data in order to act either as a
supervisory control station or as an access point for a human interface or to act as a gateway to be
connected to remote stations. For efficiently maintain, the routing path between the BS and sensor
nodes, various types of clustering protocols can be used. Nodes in static clustering are organized
into clusters that communicate with a local BS that transmit the data to the global BS, where it is
accessed by the end-user [8].
The Low Energy Adaptive Clustering Hierarchy (LEACH) [9] is a cluster based hierarchical
algorithm. It is a cross layer protocol architecture that integrates energy efficient cluster based
routing and media access together with application specific data aggregation to achieve good
performance [10]. LEACH forms clusters by using a distributed algorithm, where nodes make
autonomous decisions without any centralized intervention. In LEACH, distributed cluster
formation can be done without knowing the exact location of any of the nodes in the network and
any sensor node can act as CH. The selection of CH rotate among nodes to distribute energy
evenly in the whole network. Once a node declares itself the CH, the nearby nodes join he cluster
and send their data to CH in their assigned time slot.
Unlike the LEACH, LEACH-C [11] utilizes a central BS for the formation of CHs. During set-up
phase of LEACH-C, each node sends information about its current location and energy level to
the BS. In addition to determining good clusters, the BS needs to ensure that the energy load is
evenly distributed among all the nodes. The BS computes the average node energy, and
whichever nodes have energy below this average cannot be cluster-heads for the current round.
The nodes transmit their data to the CH node during each frame of data transfer and the CH
aggregates the data and sends the resultant data to the BS. When the CH node’s energy is
depleted, the nodes in the cluster lose communication ability with the BS and are essentially dead.
Many radio propagation models have been used to predict performance of these protocols using
simplified propagation models[12]. They have analyzed and performed simulation using free
space model and two ray radio propagation model that fails to reflect actual performance of these

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
3
routing protocols. Therefore, proper modeling of smart grid real environment is needed for proper
evaluation of performance of these protocols for WSNs for monitoring and control of the smart
grid.
The power grid environments have been modeled using IEEE 802.15.4 compliant WSNs which
showed that log normal shadowing path loss model is more accurate channel models to be used in
comparing these routing approaches.
The objective of this paper is to simulate and analyze impact of various smart grid environments
on performance of four different WSN routing protocols namely two clustering protocols namely
the Low Energy Adaptive Clustering Hierarchy (LEACH) and Centralized LEACH (LEACT-C)
as well as other two conventional protocols namely Minimum Transmission Energy (MTE) and
Static Clustering. The results revealed that smart grid’s radio propagation environments have
strong impact on the performance of all protocols. The rest of the paper is organized as follows.
Section 2 provides description of wireless channel propagation models. Section 3 presents
overview of WSN’s routing protocols. Section 4 presents simulation of different protocols.
Section 5 provides analysis and comparison of the results. Finally, Section 5 provides the
conclusion and future work.
2.
W
IRELESS
S
ENSOR
N
ETWORKS
P
ROTOCOLS
WSN’s routing protocols are responsible for finding the most efficient path for the messages to
travel on its way to a destination. In the following, four protocols namely Minimum Transmission
Energy (MTE) and Static Clustering, LEACH and LEACH-C are explained.
2.1.
M
INIMUM
T
RANSMISSION
E
NERGY
(M
TE
)
Minimum Transmission Energy (MTE) [13] is a multi-hop routing protocol where sensors
communicate directly with other sensors that are within the radio transmission range. Data is
passed to each node’s next hop neighbor until the data reaches the base station. When a node dies,
all of that node’s upstream neighbors send their data to the node’s next hop neighbor. In this
way, new routes do not need to be computed whenever a node dies. Nodes adjust their transmit
power to the minimum required to reach their next hop neighbor. This reduces interference with
other transmissions and reduces the nodes’ energy dissipation. Communication with the next hop
neighbor occurs using a Carrier Sense Multiple Access protocol (CSMA) MAC protocol, and
when collisions occur, the data are dropped. When a node receives data from one of its upstream
neighbors, it forwards the data to its next hop neighbor. This continues until the data reaches the
base station. In MTE, each node sends its message to the closest node until it reaches the base
station in which sensor nodes serve as routers for other sensor nodes. Therefore, the energy of the
sensor nodes that are near to the base station is rapidly consumed.
2.2.
S
TATIC
C
LUSTERING
In order to enable communication between sensors not within communication range, sensors form
clusters using the simulated annealing algorithm as in LEACH-C [14]. Clusters are chosen a
priori, each with a cluster head that is selected at the beginning and remain unchanged throughout
the lifetime of the network. The static clustering protocol is identical to LEACH except the
clusters are chosen a-priori and fixed. Static clustering includes scheduled data communication
from the cluster members to the cluster-head and data aggregation at the cluster head. As the
cluster head and the cluster remain same for each round, the selected static cluster head die
quickly and thus the network lifetime cannot be maximized.

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
4
2.3.
L
OW
E
NERGY
A
DAPTIVE
C
LUSTERING
H
IERARCHY
(L
EACH
)
In cluster based routing protocols, clustering algorithms are used to divide the network into
several clusters based on some stipulated rule. Each cluster is controlled by one or set of node
known as Cluster Head (CH). Data is transmitted from member nodes to CH which fused the
collected data and send aggregated data to the central node known as Base Station (BS). LEACH
periodically elects CHs randomly from all the nodes in the networks and rotates this role to
balance the energy dissipation of the sensor nodes in the networks. The CH nodes fuse and
aggregate data arriving from nodes that belong to the respective cluster at regular intervals.
Network lifetime is divided into discrete and disjoint time intervals called rounds, and every
round is composed of two phases, the setup phase and the steady state phase [14]. The setup
phase allows clusters to be formulated and CHs to be elected. In every round, a stochastic
algorithm is used by each node to determine whether it will become a CH or not. If a node
becomes a CH once, it cannot become a CH again for a given number of rounds.
The selected CH broadcast an advertisement message to the entire network declaring itself as a
new CH. Every node after receiving advertisement message decides its CH based on received
signal strength of advertisement message. After selection of CH, member nodes send message to
register with the CH of their choice. CH send the time schedule to the registered nodes so that
they can send their data using Time Division Multiple Access (TDMA) in MAC layer to allow
nodes to turn off their radio component until their allocated time slots. The schedule is
broadcasted back to all nodes in the cluster. In the next phase known as steady state phase, sensor
nodes can send the observed data to the CHs on their allocated time slot. CH send compressed
aggregated data to a central station usually know as Base Station (BS). After predetermined time
period, new CHs are elected again randomly. This repetition of electing new CHs is known as
new round so LEACH is based on large number of rounds. Randomization process is used in
LEACH to rotate CHs. During the steady phase, each CH collects data from all senor nodes in its
cluster in their assigned TDMA time slots, aggregates the data, and sends it to the BS along with
its cluster head identification.
2.4.
L
OW
E
NERGY
A
DAPTIVE
C
LUSTERING
H
IERARCHY
-
C
ENTRALIZED
(L
EACH
-C)
In LEACH, the probability of becoming a CH is based on the assumption that all nodes start with
an equal amount of energy, and that all nodes have data to send during each frame. This is not
applicable to scenarios where node’s energy may vary depending upon their role in the network
(.e. whether it is a cluster head or a cluster member). CH selection plays significant role in
developing energy efficient clustering algorithms [10]. Intra cluster communication distance
depends upon position of the selected cluster head and intra cluster energy consumption depends
upon intra cluster communication distance. Clusters with high intra cluster communication
distance consume more energy than other clusters having low intra cluster communication
distance. The nodes with more energy are elected more often than the nodes with less energy in
order to ensure that all nodes die at approximately the same time. This can be achieved by setting
the probability of becoming a CH as a function of a node’s energy level relative to the aggregate
energy of the cluster in the network, rather than purely as a function of the number of times, the
node has been CH as in LEACH.
The Low Energy Adaptive Clustering Hierarchy-Centralized (LEACH-C) [15] is adaptable to
non-uniform and dynamic energy distribution among the sensor nodes and the changing network
configurations. LEACH-C requires that each node transmit information about its location to the
base station at the beginning of each round to calculate the average energy of the nodes and
whichever nodes have energy below this average will not participate in cluster head selection.
The location information is obtained by using a global positioning system receiver that is

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
5
activated at the beginning of each round to get the node’s current location. Once clusters are
formed and CHs are selected, BS broadcasts a message that contains the identifications (IDs) of
all the CHs. If a node finds it’s ID in the message payload, it knows that it has to perform as a
CH; otherwise, it determines its allocated TDMA slot and goes to sleep until the transmission
time arrives [16]. The steady-state phase of LEACH-C is identical to that of LEACH [17].
3.
S
IMULATION
E
NVIRONMENT
The Network Simulator 2 (NS-2) [18] is used to simulate the four protocols. NS-2 is an open
source object oriented discrete event simulator. It has a simulation engine written in C++ to
implement protocols and extend the library, with a command and configuration interface using
Object Tool Command Language (OTcL). It is a scripting language to create and control the
simulation environment itself including the selection of output data. It supports the required
features and can be improved and incremented by means of external extensions. Several
extensions are added to NS-2 to support simulation of WSNs with various channel propagation
and energy dissipation models. IEEE 82.15.4 is not implemented in the default file of NS-2.
MAC implementation is developed as recommended. Network topologies are described using the
various primitives such as: Nodes, Links, Agents, and Applications. Once the topology has been
created, simulations can be run by starting the applications on different nodes at various points in
time. WSNs can be observed graphically by Network AniMator (NAM).
The energy, traffic and propagation models are selected to represent smart grid environments.
Table 1 gives propagation values for smart grid channel paramters as described in this table.
Extensive simulations to evaluate the performance of the four candidate protocols are simulated
for a 100 nodes located in a 200m×200m. Table 2 shows the common simulation parameters and
their respective values. The data size is 500 bytes/message plus a header of 25 bytes. The
message size to be transmitted is: X= (500 bytes + 25 bytes)/8 = 4200 bits.
Table 1: Log Normal Shadowing Channel Parameters
Propagation Environment Path Loss

Shadowing deviation

500-kv substation (LOS)
outdoor 2.42 3.12
500-kv substation (NLOS)
Outdoor 3.51 2.95
Underground network transformer
vault (LOS) 1.45 2.54
Underground network transformer
vault (NLOS) 3.15 3.19
Main power room(LOS)
indoor 1.64 3.29
Main power room(NLOS)
indoor 2.38 2.25
Non smart grid environment
Indoor environment 1.4 4
4.
P
ERFORMANCE
M
ETRICS
The most important metrics for evaluating WSNs are number of data received at base station,
energy consumption, network lifetime. Statistics are collected at periodic time intervals to
evaluate the effectiveness of the different candidate protocols.

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
6
(a) Data Received At Base Station
It is needed to know that how much data is received at the base station from the sensor
networks. If the amount of data received is enough as it is expected then the network will
perform well.
(b) Average Energy Consumption
The average energy consumption is calculated across the entire topology. It measures the
average difference between the initial level of energy and the final level of energy that is left
in each node. If


is the initial energy level of a node,


is the final energy level of a node
and

is number of nodes in the simulation, then the average energy consumption is


=
∑







(1)
This metric is important because it is proportional to the network’s lifetime. The lower the
energy consumption the longer is the network’s lifetime.
(c) Total Number of Nodes Alive
This metric indicates the overall lifetime of the network. Performance of a network depends
on the lifetime of each node, if the lifetime of the nodes increase then the network performs
well and sensors transmit more data to the base station. A lower value of the energy
consumption metric and a higher number of nodes alive at any given time indicates a more
efficient protocol.
Table 2: NS-2 Simulation Parameters
Simulator NS-2.34
Examined Protocols MTE, Static Clustering, LEACH, and LEACH-C
Simulation Duration 1000 seconds
Simulation Area 200 m x 200 m
Number of Nodes 100
Location of BS 50,175
Initial Energy 2 Joules
Number of cluster heads 5%
E
elec
50 nano Joule / bit
ε
fs
10 pico Joule/bit/m
2
ε
amp
0.0015pico Joule/ bit/m
4
Crossover Distance 87 m
Radio propagation speed 3x10
8
m/s
Radio Speed 1 Mbps
Antenna gain (
G
t
, G
r
) 1
Antenna height (
h
t
, h
r
) 1.5 m
System Loss Factor L 1
Signal Wavelength 0.328 m
Data payload 250 Bytes
Radio Bit Rate 1 Mbps
Processing Delay 50 micro seconds
5.
S
IMULATION
R
ESULTS AND
A
NALYSIS
The four protocols were simulated first in ideal two ray ground propagation and then compared
with shadowing propagation model. Although the two-ray propagation model has been widely
used WSN’s simulation, this model is inappropriate as it is based on simplified assumptions that

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
7
neglect the effect of fading, which represents the actual environments. The two ray ground has
delivered more data to base station, consumed less energy, and has longer lifetime. The results
revealed that the realistic and representative shadowing propagation model under harsh smart grid
environments have considerable impact on reliability and network lifetime of the four protocols.
Consequently, the performance deteriorated very quickly when shadowing model was taken into
account. The main reasons for this deterioration resulted from the large variation of the received
signal strength. Hence packets are not received successfully at BS due to the poor signal quality,
which causes problems to the normal operations of various protocols. The four protocols
performed quite differently and this gives a hint to the fact that simulation results for WSNs have
to be interpreted with a lot of care in order to conclude accurate results especially when reliability
and network lifetime are considered. Due to space limitations, results for two ray ground is not
shown here but it is available upon request.
Figure 1. Total Number of Data Delivered to Base Station vs. Time in (a) 500 kV Sub-Station-NLOS, (b)
Main Power Room-NLOS, (c) Underground Transformer Vault-NLOS

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
8
The aim of this paper is to simulate different WSNs protocols and analyze their performance
using the realistic and representative shadowing propagation model under harsh smart grid
environments. Due to limited space, only three environments are presented here corresponding to
a 500 kV substation NLOS, an underground transformer vault NLOS, and a main power room
NLOS environments. Results for the other four environments are available upon request. All
over the simulation, the seven different environments have considerable impact on performance
metrics and the different protocols exhibit nearly the same behavior over time in different
environments.
In Figure 1, LEACH-C delivers the highest amount of data to base station and considerably
outperforms LEACH in terms of the amount of data sent to the base station during network
lifetime as depicted in Figure 1. This is because the base station has global knowledge of the
location and energy of all the nodes in the network. Simulation time of LEACH-C ends earlier.
Both protocols outperform MTE and static clustering because both rotate selection of CHs
between nodes and adapt the corresponding clusters based on the nodes that are selected to be
cluster heads at a given time. In static clustering, nodes are organized into clusters initially by the
BS using the same method as in LEACH-C to ensure that good clusters are formed. These
clusters and cluster heads remain fixed throughout the lifetime of the network. Nodes transmit
their data to the cluster head node during each frame of data transfer and the cluster head
aggregates the data and sends the resultant data to the BS. When the cluster head node’s energy is
depleted, the nodes in the cluster lose communication ability with the BS.As soon as the cluster
head node dies, all nodes from that cluster effectively die since there is no way to deliver their
data to the base station as shown in Figure 2.
In Figure 2, the energy consumption for MTE routing increases with simulation time. Each node
runs a start-up routine to determine its next-hop neighbor, defined to be the closest node that is in
the direction of the BS. Data packets are passed along via next hop neighbors until they reach the
BS. As there is no central control in MTE routing, it is difficult to set up fixed MAC protocols
(e.g., TDMA), so each node uses CSMA to listen to the channel before transmitting data. If the
channel is busy, the node backs off; otherwise, the node transmits its data to the next-hop node.
As nodes run out of energy, the routes are recomputed to ensure connectivity with the BS. As
time increases, more nodes die and this reduce amount of energy consumed with respect to other
protocols. Figure 2 also shows that LEACH consumes less amount of energy than LEACH-C.
LEACH is completely distributed which requires no control information from the base station,
and the nodes do not require knowledge of the global network in order for LEACH to operate.
Distributing the energy among the nodes in the network is effective in reducing energy
consumption from a global perspective and enhancing network lifetime.

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
9
Figure 2: Average Energy Consumption Vs. Simulation Time in (a) 500 kV Sub-Station-NLOS, (b) Main
Power Room-NLOS, (c) Underground Transformer Vault-NLOS
Figure 3 shows the number of nodes that remains alive using LEACH is significantly larger than
that of LEACH-C. Lifetime of LEACH is longer than LEACH-C as LEACH rotates the cluster
head nodes and the associated clusters, nodes die more slowly which indicate that the protocol
can balance the nodes’ energy consumption. In addition to reducing energy consumption in last
figure, LEACH successfully distributes energy usage among the nodes in the network such that
the nodes die randomly and at essentially the same rate. However, the main problem with the
LEACH protocol lies in the random selection of CHs. There exists a probability that the CH
formation is unbalanced and may remain in one part of the network, making some part of the
network unreachable. In MTE, all node pass own data to nodes closest to BS and that nodes die
quickly due to higher energy consumption. In static clustering, as soon as the cluster head node
dies, all nodes from that cluster effectively die since there is no way to deliver their data to the
base station although member nodes have enough energy. The nodes in static clustering are
completely depleted in the first
50
seconds which indicate that static clustering performs poorly
because more energy is drained from preassigned fixed cluster heads due to message transmission
over long distances compared to other nodes in the cluster based protocols. Therefore, cluster
head node’s energy is drained quickly, ending the lifetime of all other nodes belonging to those
clusters as they lose communication with base station.
(c)

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
10
Figure 3: Total Number of Alive Nodes Vs. The Simulation Time in (a) 500 kV Sub-Station-NLOS, (b)
Main Power Room-NLOS, (c) Underground Transformer Vault-NLOS
Figure 4 shows that static clustering and MTE deliver lowest amount of data to base station while
LEACH and LEACH-C deliver the highest amount of data per unit energy, achieving energy
efficiency. LEACH-C deliver more data per unit energy than LEACH because the BS has global
knowledge of the location and energy of all the nodes in the network, so it can produce better
clusters that require less energy for data transmission. Static clustering performs poorly because
all CH nodes die quickly, ending the lifetime of all nodes belonging to those clusters. It is shown
that LEACH is almost as efficient as LEACH-C. In LEACH-C it starts with the lower energy
consumption but due to the transmission overhead it requires high energy consumptions. LEACH
is found to be most energy efficient among the four protocols. LEACH is completely distributed
which requires no control information from the base station, and the nodes do not require
knowledge of the global network in order for LEACH to operate.
Figure 5 shows the total number of nodes that remain alive per amount of data received at the BS.
LEACH can deliver more effective data than LEACH-C for the same number of node deaths.
MTE requires the highest amount of energy to send data to the BS due to collisions and lack of
data aggregation. Also MTE does not have any centralized control over when nodes transmit and
receive packets, collisions increase the amount of energy required to send each successful
message, causing more node deaths for the same amount of data delivery to end network lifetime
early. Furthermore, each message in MTE must traverse multiple hops to reach
BS, whereas each

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
11
Figure 4: Average Energy Consumption Vs. Simulation Time in (a) 500 kV Sub-Station-NLOS, (b) Main
Power Room-NLOS, (c) Underground Transformer Vault-NLOS
message in LEACH need only traverse one hop due to data aggregation at the cluster head. The
static clustering performs poorly and the cluster head nodes die quickly, ending the lifetime of all
nodes belonging to those clusters. The static clustering drains the nodes quickly as nodes die early
and loose communications with BS. As the cluster head and the cluster remain same for each
round, the selected static cluster heads die quickly and thus the network lifetime is quickly
degraded. The figures clearly show the large advantage of using LEACH and LEACH-C versus
conventional protocols in terms of network lifetime for a given amount of received data at the
base station and LEACH outperforms LEACH-C in prolonging network lifetime.

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
12
Figure 5: Number of alive nodes vs. number of data received at the base station in (a) 500 kV Sub-Station-
NLOS, (b) Main Power Room-NLOS, (c) Underground Transformer Vault-NLOS
6.
C
ONCLUSION AND
F
UTURE
W
ORK
Latest advances in computing and networking have enabled WSNs to realize ambient
intelligence, which is a vision through which electric power grid becomes smart. The main
challenge in the design of routing protocols for WSNs is to ensure effective consumption of
energy with ultimate objective to extend lifetime of network for as long as possible. The
realization of smart grid depends on communication facilities of WSNs in harsh and complex
environments of electric power grid that have been modelled using shadowing model. To achieve
reliable wireless communications within WSNs, it is essential to investigate performance of
routing protocols in different smart grid environments. Four WSNs protocols namely MTE, Static
Clustering, LEACH, and LEACH-C are simulated on NS- 2. The performance is evaluated using
number of data signals received at base station, energy consumption, and network lifetime as
performance metrics. The paper analyzed the results and finally reaches to a conclusion about the
routing protocol that can be efficient for monitoring and control applications in the smart grid.
The results have clearly showed that the smart grid environments have directly affected
performance of various protocols and energy efficient clustering approaches are more effective in
prolonging the network lifetime compared to conventional protocols. It is clear that in MTE
routing is not suitable for smart grid applications as the nodes closest to the base station will be
used to route a large number of data messages to the base station. Thus, these nodes will die out
quickly, causing the energy required to get the remaining data to the base station to increase and

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
13
more nodes to die. This will create a cascading effect that will shorten network lifetime. In
addition, as nodes close to the base station die that area of the environment is no longer being
monitored. In addition, static clustering is not suitable, where nodes are organized into clusters
that communicate with a local base station, and these local base stations transmit the data to the
global base station, where it is accessed by the end user. However, the local base station is
assumed to be ahigh-energy node; if the base station is an energy constrained node, it would die
quickly, as it is being heavily utilized.
LEACH-C achieved considerable improvement in number of data derived to base station because
it select cluster heads based on remaining energy and their geographical position. However, it
consume more energy and shorten the network lifetime. On the other hand, LEACH is completely
distributed which requires no control information from the base station, and the nodes do not
require knowledge of the global network in order for LEACH to operate. Distributing the energy
among the nodes in the network is effective in reducing energy consumption from a global
perspective and enhancing network lifetime. In all simulation, LEACH provided considerable
energy savings and prolonged network lifetime over other protocols. As the protocol is fully
distributed so there is no intervention from Base station in cluster formation and cluster head
selection. Global location knowledge of sensor nodes is not required. These features makes
LEACH a promising option for various smart grid applications.
Some shortcomings however are uneven distribution of CH and unbalance energy utilization in
the whole network. Due to random selection of CH it could not guaranteed that Cluster heads are
evenly distributed in the network. LEACH assumes that energy of all nodes is same and remains
fixed with time. It also assumes that energy consumption across the network is identical as the
cluster head selection is rotated among the nodes. These assumptions make CH selection
unrelated to the residual energy of a node and nodes with little energy are as likely to become
cluster heads as the nodes with abundance of energy. The energy deficient nodes die fast and
gradually render the network useless even though there might be numerous nodes still having
enough energy to be functional. Moreover, even though the nodes are equipped with the same
energy at the beginning, the networks cannot evolve equably for each node in expending energy,
due to the radio communication characteristics, random events such as short-term link failures or
morphological characteristics of the fields. Therefore, WSNs are possibly heterogeneous
networks and protocols should meet the need of the characteristic of heterogeneous wireless
sensor networks. Due to these drawbacks, future work will be carried out to solve these problems
and to propose an adaptive and reliable cluster based protocol to enhance energy consumption
and prolonging networks lifetime in smart grid environments.
R
EFERENCES
[1]
E. Fadel, V.C. Gungor, Laila Nassef, Nadine Akkari, M.G. Abbas Malik, Suleiman Almasri and Ian F.
Akyildiz, "A Survey on Wireless Sensor Networks For Smart Grid,"
Computer Communications, Vol.
71,
pp. 22-33, November 2015, https://doi.org/10.1016/j.comcom.2015.09.006.
[2]
M. Hammoudeh and R. Newman, "Adaptive routing in wireless sensor networks: QoS optimisation
for enhanced application performance,"
Information Fusion, Vol. 22,
pp. 03-
15, March 2015,
https://doi.org/10.1016/j.inffus.2013.02.005.
[3]
M. R. Mundada, S. Kiran, S. Khobanna, R. N. Varsha and S. A. George, "A STUDY ON ENERGY
EFFICIENTA STUDY ON ENERGY EFFICIENT SENSOR NETWORKS,"
International Journa
l of
Distributed and Parallel Systems (IJDPS), Vol.3, No.3,
pp. 311-
330, May 2012, DOI :
10.5121/ijdps.2012.3326.
[4]
F. Farazandeh, R. Abrishambaf and S. Uysal, "A Hybrid Energy-
Efficient routing protocol for
Wireless Sensor Networks," in
11th IEEE
International Conference on Industrial Informatics
(INDIN)
, Bochum, Germany, 2013, DOI: 10.1109/INDIN.2013.6622851.
[5]
P. Yuvaraj and Komanapalli Venkata Lakshmi Narayana, "EESCA: Energy efficient structured

International Journal of Advanced Smart Sensor Network Systems (IJASSN), Vol 8, No.1/2/3, July 2018
14
clustering algorithm for wireless sensor networks," in
International Conference on Computing,
Analytics and Security Trends (CAST)
, Pune, India, 19-
21 Dec. 2016, DOI:
10.1109/CAST.2016.7915024.
[6]
S. C. Sharma and S. P. Singh, "A Survey on Cluster Based Routing Protocols in Wireless Sensor
Networks,"
Procedia Computer Science, vol. 45,
pp. 687-
695, 2015,
https://doi.org/10.1016/j.procs.2015.03.133.
[7]
Z. A. Sadouq, M. El Mabrouk and M. Essaaidi, "Conserving Energy in WSN Through Clustering and
Power Control," in
Third IEEE International Collo
quium in Information Science and Technology
(CIST)
, Tetouan, Morocco, 20-22 Oct. 2014, DOI: 10.1109/CIST.2014.7016654.
[8]
N. Ramluckun and V. Bassoo, "Energy-efficient chain-
cluster based intelligent routing technique for
Wireless Sensor Networks,"
Applied Computing and Informatics,
07 March 2018,
https://doi.org/10.1016/j.aci.2018.02.004.
[9]
Pooja and S. Singh, "Improved O-
LEACH protocol: A clustering based approach in wireless
microsensor network," in
10th International Conference on Intelligent Systems and Control (ISCO)
,
Coimbatore, India, 7-8 Jan. 2016, DOI: 10.1109/ISCO.2016.7727015.
[10]
W. B. Heinzelman, "A Component
-
Based Cross
-
Layer Framework for Software Defined Wireless
Networks," in
8th IFIP International Conference on New
Technologies, Mobility and Security
(NTMS)
, Larnaca, Cyprus, 21-23 Nov. 2016, DOI: 10.1109/NTMS.2016.7792426.
[11]
M. Tripathi, M. S. Gaur, V. Laxmi and R. B. Battula, "Energy efficient LEACH
-
C protocol for
Wireless Sensor Network," in
CIIT 2013. Third
International Conference on Computational
Intelligence and Information Technology
, Mumbai, India, 18-
19 Oct. 2013, DOI:
10.1049/cp.2013.2620.
[12]
L. Nassef, "On the Effects of Fading and Mobility in On Demand Routing Protocols,"
Egyptian
Informatics Journal, Vol.11,
pp. 67-74, 2010, http://dx.doi.org/10.1016/j.eij.2010.10.003.
[13]
B. Pati, J. L. Sarkar and C. R. Panigrahi, "ECS: An Efficient Approach to Select Cluster Head in
Wireless Sensor Networks,"
Arabian Journal for Science and Engineering, VoL. 42, Issue: 2,
p. 669–
676, February 2017, https://doi.org/10.1007/s13369-016-2304-2.
[14]
R. D. Gawade and S. L. Nalbalwar, "A Centralized Energy Efficient Distance Based Routing Protocol
for Wireless Sensor Networks,"
Journal of Sensors, Vol. 2016,
2016,
http://dx.doi.org/10.1155/2016/8313986.
[15]
I. Sharma, R. Singh and M. Khurana, "Comparative study of LEACH, LEACH
-
C and PEGASIS
routing protocols for wireless sensor network," in
International Conference on Advances in
Computer
Engineering and Applications (ICACEA)
, Ghaziabad, India, 19-
20 March 2015, DOI:
10.1109/ICACEA.2015.7164821.
[16]
V. Geetha, P. V. Kallapur and S. Tellajeera, "Clustering in Wireless Sensor Networks: Performance
Comparison of LEACH & LEACH-C Protocols Using NS2,"
Procedia Technology, Vol. 4,
pp. 163-
170, 2012, https://doi.org/10.1016/j.protcy.2012.05.024.
[17]
W. Xinhua and W. Sheng, "Performance comparison of LEACH and LEACH
-
C protocols by NS2," in
Ninth International Symposium on Distribute
d Computing and Applications to Business Engineering
and Science (DCABES)
, Hong Kong, China, 10-12 Aug. 2010.
[18]
"http://www.isi.edu/nsnam/ns/," The network simulator ns
-
2. [Online].