Content uploaded by Fuseini Jibreel
Author content
All content in this area was uploaded by Fuseini Jibreel on Jul 22, 2019
Content may be subject to copyright.
International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 42, February 2019
43
Gateway–based Threshold Distributed Energy Efficient
Clustering (G-TDEEC)
F. Jibreel
Computer science department
Tamale Technical University
Tamale, Ghana
ABSTRACT
Wireless sensor network basically is made up of several
sensor nodes that sense physical quantities such as
temperature and pressure in a physical environment, capture
these quantities and relay the data to another node called the
Base station(BS). The transmission of the sensed information
from the deployment area to the BS has been observed to
drain the limited energy resource of the sensor nodes. Some
researchers are of the view that, placing the BS at the centre of
the sensing field will sufficiently reduce the energy
consumption during data transmission. Base on this, some of
the descendants of DEEC protocol such as TDEEC
(Threshold Distributed Energy Efficient Clustering) protocol
also placed the BS at the centre of the deployment field to
conserve network energy. So what happens to TDEEC scheme
if it was to be deployed at a place such as military surveillance
where the BS may be far from the sensing field? In this
research work, a gateway-based TDEEC, G-TDEEC protocol
is proposed. The new scheme introduced a gateway at the
centre of the sensing area and then installed the BS far away
from the sensing field. The cluster heads relay their data to the
gateway which will then aggregate the data and then send the
final report to the BS. Simulation was performed to assess the
performance of the proposed protocol and the TDEEC scheme
using MatLab 2017a. The simulation results showed that, the
proposed protocol performed better than the existing scheme
in terms of stability period, throughput, residual energy and
the network lifetime.
General Terms
Algorithm-Routing protocol
Keywords
G-TDEEC Protocol; Network lifetime; Centre, Outside;
Gateway; MatLab simulation.
1. INTRODUCTION
Improvements in technology has made the shifting from wired
networks to wireless networks possible where fix
infrastructure is not needed. Among these wireless networks is
a wireless network that is composed of thousands of sensor
nodes mostly deployed to collect measurement quantities such
as temperature and humidity in any physical environment.
This kind of network is referred as wireless sensor networks.
When the sensor nodes in these networks sensed data from the
environment, they convey the information wirelessly to a
well-resourced node called the Base station (BS). Once the
data gets to the BS, the user can access and analysis it
satisfactorily. However, the nodes in this network are very
tiny, constrained in terms energy (they are battery operated),
limited in ability to process, store and even transmit large data
over a long distance. And their ability to transmit data over a
longer distance is very crucial in a very hostile environment.
Although, in some of the heterogeneous routing protocols, the
Base stations are mostly placed at the centre of network to
reduce energy consumption, other strategy can be adopted to
still allow them function very well in areas where the BS must
be far from the deployment areas. Several energy saving
methods have been proposed. Among them is the energy
efficient cluster-based routing algorithm for both
heterogeneous and homogeneous networks. These routing
protocols are gaining more attention from the researchers.
This is proven in some of the following literature.
Heinzelman et al. [1] explained the earliest single-hop
clustering routing protocol called Low-Energy Adaptive
Clustering Hierarchy (LEACH). The scheme conserved the
network energy better compared to the non-cluster-based
routing schemes such as the direct transmission protocol and
since then, several clustering algorithms were developed
based on LEACH.
A centralized form of LEACH (LEACH_C) was suggested by
Heinzelman et al. [2]. In this algorithm, the Base station
manages the affair of the network. In the setup phase, all the
sensor nodes send a message containing, their energy level
and location details to the BS. The Base station then select the
cluster heads based on the node location details. It then
introduces the selected cluster heads and their IDs to all the
nodes. However, the usage of GPS receiver in each round
affects the performance of the protocol.
Kaur and Kaur [3] described Enhanced M-Gear scheme for
Wireless Clustering System. In this protocol, the number of
gateway nodes were increased to ensure evenly distribution of
load among them. The network was divided into a number of
sections and each section is assigned a gateway node. The
nodes of a region will transmit their data to their assigned
gateway node which will then send to either BS or nearest
gateway to the BS. Although, the simulation results proved
that, the proposed scheme performed better than MGEAR in
terms of throughput, energy consumption and network
lifetime the cost of the network will high because of the
number of gateway nodes.
Author in [4] presented an improved version of M-Gear
Protocol for homogeneous wireless sensor network. The
protocol modified the threshold for choosing cluster heads by
taking into account the distance between the nodes and the
gateway as well as their residual energy. The scheme also
introduced hard and soft thresholds to reduce unnecessary
transmission of data to the Base station. The simulation results
showed that, the scheme performed better than M-Gear in
terms of stability period, throughput, residual energy and
network life time.
Qing et al. [5] proposed a heterogeneous routing scheme,
DEEC which has now become the basis upon which several
heterogeneous routing protocol are developed and continue to
be developing. The protocol select cluster head based on the
ratio between residual energy of each node and the average
energy of the network. The algorithm uses two type of nodes,
International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 42, February 2019
44
the normal and advanced nodes in its two level hierarchy.
However, the challenge in the scheme is the continuous
discrimination against the advanced nodes when their residual
energy becomes equal to the normal nodes in the network.
Elbhiri et al. [6] suggested a new form of DEEC algorithm for
heterogeneous network. DDEEC, presented a better solution
to the main problem that was observed in DEEC scheme. The
problem exist where the advanced nodes are constantly being
punished when their residual energy becomes at equal level
with that of the normal nodes. To solve this, DDEEC
algorithm introduced threshold residual energy value,
for all the nodes based on which the average probability of
each node is determined.
Another version of DEEC model, E-DEEC for heterogeneous
networks has been described by Saini et al. [7]. The scheme
added another sensor nodes, super nodes to the normal and
advanced nodes. So, E-DEEC is made up of three types of
sensor nodes with different initial energies. The nodes with
highest initial energy is the super nodes followed by advanced
nodes and normal nodes with the lowest initial energy.
Simulation was conducted to see the performance of the
scheme and the simulation results showed that, E-DEEC
outperforms DEEC in terms of stability period and network
life enhancement.
Authors in [8] proposed improved form of DEEC protocol,
TDEEC algorithm. This model also adopted three level of
sensor nodes with different initial energies as in [7]. The
algorithm however modified the value of the threshold, upon
which a node decides to be a cluster head or not. The
threshold is based on ratio of residual energy and average
energy of that round with respect to the optimum number of
cluster heads. This is to make the nodes with more energy to
become the cluster head. TDEEC also suggested the
probabilities for two levels, three levels and multilevel
heterogeneity. The simulation results proved that, TDEEC
performed better in terms stability period and network lifetime
especially, in its three levels and multilevel.
The remainder of this research is organized as follows:
Section 2 described the methodology used, simulation results
and analysis are discussed in Section 3 and conclusion is then
drawn in Section 4
2. METHODOLOGY
DEEC protocol was the basis upon which many
heterogeneous routing protocols have been proposed. The
protocol places the Base station at the centre of the sensing
field for simplicity according to [5] but to large extend is to
sufficiently reduce the energy consumption in the network.
So, TDEEC also placed the BS at the centre with the aim of
minimizing energy depletion in the network. It was observed
that, when the BS is placed outside the sensing area, the
TDEEC protocol performed poorly in term of throughput and
energy conservation. This means that, when it comes to
applications where the BS must be placed outside the
network, TDEEC cannot performed well. In order to solve this
problem, a new algorithm called gateway-TDEEC (G-
TDEEC) is proposed. In this new scheme, a gateway node is
introduced and placed at the centre and re-installed the BS far
from the deployment field. The cluster heads which receive
information measurements from the normal sensor nodes relay
the data to the gateway. The gateway then aggregate the data
and then convey the report to the BS far from the sensing
field.
3. SIMULATION RESULTS AND
ANALYSIS
To assess the performance of G-TDEEC protocol and T-
DEEC scheme, MatLab 2017a was used for simulation. In this
experiment, a random network of 100 nodes is used in 100m
x100m cross-sectional area. The gateway is placed at the
middle of the deployment area (50m, 50m) and the Base
station installed outside the field (50m, 200m). For the
composition of the nodes used, 20 advanced nodes were
deployed with 1.5 times more energy than normal nodes and
30 super nodes with 3 times more energy than the normal
nodes ( ). Other
parameters used in the simulation are shown in the Table1.
Table I: Simulation Parameters
S/N
Parameter
Values
1
elect
E
50nJ/bit
2
fs
E
10pJ/bit/m2
3
mp
E
0.0013pJ/bit/m2
4
0.5J
5
4000
6
100
7
0.1
5nJ/bit/message
Table 2: Round vs Node death during simulation process
Protocol
Death count
Round
TDEEC
50
256
100
1750
G-TDEEC
50
538
100
2375
Table 3: Round vs Residual energy during simulation
process
Protocol
Round
Residual energy
TDEEC
20
0.3
50
0.0
G-TDEEC
20
0.42
460
0.0
Round numbers of deaths of all the nodes during the
simulation process for both the proposed and existing
protocols have been gathered. To ensure effective comparison,
data for death of half of the sensor nodes and all the sensor
nodes in both protocols have been shown in Table 2 and the
detailed plot of the whole data displayed in Fig 2. Table 3 also
shows the residual energies of the two protocols with the
detailed plot of the whole data in Fig 4. From the Table 2 and
International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 42, February 2019
45
Table3, it is clear that, the proposed algorithm has better
network lifetime and residual energy than the existing scheme.
Figure 1 shows the number of alive nodes per round during
simulation process for the G-TDEEC protocol and T-DEEC
scheme. It can be observed from the graph that, the lifetime of
the network has been extended in G-TDEEC compared to T-
DEEC. The nodes in T-DEEC survived up to 1750 rounds and
vanished but remained alive up to 2375 rounds in G-TDEEC
before disappearing. This shows that, nodes remain a live for
longer time in G-TDEEC and hence better lifetime than T-
DEEC routing scheme. The longer lifetime of the new
algorithm is as a result of the multi-hop communication
method adopted in the protocol. The cluster heads send their
data to the gateway which then relay it to the BS. So energy of
the cluster heads are conserved as well as other nodes in the
network.
Figure 1: Number of the Alive nodes per round
Figure 2 shows the number of dead nodes per round for the G-
TDEEC protocol and the exiting scheme. It was again realised
from the graph that, the death rates in G-TDEEC is lower
compare to that of T-DEEC as seen in Figure 2. At 1750
rounds, all the nodes in T-DEEC are dead where as in G-
TDEEC, it was 2375 rounds. Also, the new scheme has better
stability period than the T-DEEC scheme. As early as 100
rounds, T-DEEC first node died where as in G-TDEEC, it is
in 400 rounds. This shows that the proposed scheme has
effectively reduced the number of dead nodes resulting into a
better network lifetime and stability period.
Figure 2: Number of the dead nodes per round
Figure 3 also shows the magnitude of data sent to the BS per
round in both G-TDEEC and the existing protocols. It can be
seen that, the amount of data sent to the BS by T-DEEC
increases from 0 to approximately 40000 at the end of the
experiment sending less amount of data to the BS. In the new
scheme, large quantity of data was observed being conveyed
to the BS which is more than even 50000. This performance is
as result of the multi-hop communication mode used and
reduction of burden on the cluster heads. The heads are
supposed to aggregate the data received from the normal
nodes in the existing scheme but in the proposed protocol, the
gateway rather perform such function thereby reducing the
energy that would have been used by the heads for such
purpose. So the heads have transmitted more data with less
energy expenditure.
Figure 3: Number of packet to the BS per rounds
Figure 4 displays energy dissipation of the network in both
routing algorithms. As early as 100 rounds, the existing
protocol has drained its energy. Though this is understandable
since T-DEEC was not designed for such long distant Base
station. The new algorithm on the other hand shows relatively
reduction in energy consumption because of the presence of
International Journal of Computer Applications (0975 – 8887)
Volume 182 – No. 42, February 2019
46
the gateway. It manages the energy consumption of G-
TDEEC until 400 rounds. This shows that, the energy
remaining per round in the proposed model is better than the
T-DEEC protocols.
Figure 4: Remaining Energy per round
4. CONCLUSION
In this work, gateway-TDEEC (G-TDEEC) protocol for
heterogeneous networks is proposed. In the model, gateway
node was introduced at the centre of the network while
installing the Base station outside far from the deployment
area. The gateway receives the measurement data from the
cluster heads, aggregate it and convey the final report to the
Base station. This has reduced the energy expenditure of the
cluster head which they could have used for the data fusion.
The scheme also adopted multi-hop communication from the
normal sensor nodes to the Base station. And this has also
reduced the energy consumption in the network. The
simulation results showed that, the proposed protocol
performed better than the T-DEEC in terms of coverage,
stability period, and throughput and network life time.
5. REFERENCES
[1] [W.R. Heinzelman, A. Chandrakasan, “Energy efficient
Communication Protocol for Wireless Microsensor
Networks”, In: IEEE Computer Society Proceedings of
the 33rd Hawaii International Conference on System
Sciences (HICSS '00),Vol. 8, 2000, pp. 8020.
[2] W. B. Heinzelman, A. P. Chandrakasan, H.
Balakrishnan, “An application-specific protocol
architecture for wireless microsensor networks,” IEEE
Trans on Wireless Communications, Vol. 1, No. 4, 2002,
pp. 660-670.
[3] G. Kaur, and S. Kaur, “Enhanced M-Gear Protocol for
Lifetime Enhancement in Wireless Clustering System”,
International Journal of Computer Applications (0975 –
8887) Vo. 147 – No.14, 2016 pp.30-34
[4] F. Jibreel, ” Improved- Gateway-Based Energy-Aware
Multi-Hop Routing Protocol for WSNs”,International
Journal of Innovative Science and Research
Technology, Volume 3, Issue 12, 2018, pp.625-630
[5] L. Qing, Q. Zhu, and M. Wang, “Design of a distributed
energy-efficient clustering algorithm for heterogeneous
wireless sensor networks”, Computer Communications,
vol. 29, no.12, 2006, pp.2230-2237.
[6] B. Elbhiri, R, Saadane, S. El Fkihiand and D.
Aboutajdine, “Developed Distributed Energy-Efficient
Clustering (DDEEC) for heterogeneous wireless sensor
networks”, I/V Communications and Mobile Network
(ISVC), 5th International Symposium on, vol., no., 2010,
pp.1-4
[7] P. Saini and A. K. Sharma, “E-DEEC- Enhanced
Distributed Energy Efficient Clustering Scheme for
heterogeneous WSN”, 1st International Conference on
Parallel, Distributed and Grid Computing, 2010, pp. 205-
210.
[8] P. Saini, and A. K. Sharma, “Energy Efficient Scheme
for Clustering Protocol Prolonging the Lifetime of
Heterogeneous Wireless Sensor Networks”, International
Journal of Computer Applications 6(2), 2010, pp. 30-36.
IJCATM : www.ijcaonline.org
