arXiv:1307.7105v1 [cs.NI] 26 Jul 2013
M-GEAR: Gateway-Based Energy-Aware
Multi-Hop Routing Protocol for WSNs
, M. B. Rasheed
, N. Javaid
, Z. A. Khan
, Y. Maqsood
, A. Din
COMSATS Institute of Information Technology, Islamabad, Pakistan.
Faculty of Engineering, Dalhaousie University, Halifax, Canada.
Abasyn University, Peshawar, Pakistan.
Abstract—In this research work, we advise gateway based
energy-efﬁcient routing protocol (M-GEAR) for Wireless Sensor
Networks (WSNs). We divide the sensor nodes into four logical
regions on the basis of their location in the sensing ﬁeld. We
install Base Station (BS) out of the sensing area and a gateway
node at the centre of the sensing area. If the distance of a
sensor node from BS or gateway is less than predeﬁned distance
threshold, the node uses direct communication. We divide the
rest of nodes into two equal regions whose distance is beyond
the threshold distance. We select cluster heads (CHs)in each
region which are independent of the other region. These CHs are
selected on the basis of a probability. We compare performance
of our protocol with LEACH (Low Energy Adaptive Clustering
Hierarchy). Performance analysis and compared statistic results
show that our proposed protocol perform well in terms of energy
consumption and network lifetime.
Keywords: Wireless Sensor Networks; clustering; Gateway.
A key concern in WSN technology is to enhance the
network lifetime and to reduce the energy consumption of the
sensor network. Wireless sensor nodes are dispersed typically
in sensing area to monitor earthquake, battle ﬁeld, industrial
environment, habitant monitoring , agriculture ﬁeld ,
physical atmosphere conditions and smart homes. Sensor
nodes sense the environment, gather information and transmit
to BS through wireless link.
Due to escalating in Micro-Electro-Mechanical System
technology, now it is possible to set up thousands or millions
of sensor nodes. The intense deployment of WSN makes it
quite difﬁcult to recharge node batteries. Therefore, a key
subject for WSNs is to curtail power expenditure of sensor
nodes to prolong network lifetime. Many clustering based
algorithms   are proposed. Clustering is a technique
in which network energy consumption is well managed by
minimizing the transmission range of the sensors. In this
modus operandi, CH manages the group communication with
the BS. Sensor nodes no longer transmit data directly to the
BS instead CHs receive the whole group messages, aggregates
and forwards to the BS.
All nodes in cluster transmit their data to corresponding
CH. The CH issues a Time Division Multiple Access (TDMA)
schedule for its member nodes to avoid collision. Each mem-
ber node transmits its data to CH only in deﬁned allocated
time slot therefore, sensor nodes turn off their transceivers
otherwise. TDMA scheduling encourages saving energy of
sensor nodes and these nodes stay alive for longer period.
As a rule, each member node transmits its data to nearby
CH therefore; sensor nodes require minimum energy for data
transmission. CHs perform computation on collected data and
ﬁlter out the redundant bits, it reduces the amount of data that
has to forward to the BS. Consequently, transmission energy
of sensors reduce to signiﬁcant amount. In this research work,
we design a gateway based energy-aware multi-hop routing
The impulse behind this work is to trim down the energy
consumption of sensor nodes by logically dividing the network
into four regions. We use different communication hierarchy in
different regions. Nodes in one region communicate directly to
BS while nodes in region 2 communicate directly to gateway
node. Nodes in other two regions use clustering hierarchy and
sensor nodes transmit their data to gateway node through their
CHs. Gateway node assists in deﬁning clusters and issues a
TDMA schedule for CHs. Each CH issues its own TDMA
schedule for its member nodes.
The rest of the paper is ordered as follows: section 2 brieﬂy
review the related work. In section 3, we describe motivation
for this work. Section 4 describes the network model. Proposed
algorithm is explained in section 5. In section 6, we deﬁne
the performance parameters and show the performance of
our proposed protocol by simulations and compare it with
LEACH. Finally, section 7 gives conclusion.
II. RELATED WORK
Energy consumption and network lifetime are the most
important features in the design of the wireless sensor net-
work. This study present clustering based routing for WSNs.
Many clustering based protocols are homogeneous, such as
LEACH  PEGASIS  and HEED . CHs collect data
from its members or slave nodes, aggregate and than forward
to faraway located BS. This process overloads the CH and
it consumes lot of energy. In LEACH, the CHs are selected
periodically and consume uniform energy by selecting a new
CH in each round. A node become CH in current round on the
basis of probability p. LEACH performs well in homogenous
network however, this protocol is not considered good for
heterogeneous networks as shown in .
In  author presented another clustering protocol(TL-
LEACH). This protocol describes two level clustering scheme
which performs well in terms of minimum energy consump-
tion of network. There are two levels of CHs, level one
CHs and level two CHs. Level one CHs connect with their
corresponding member sensor nodes. CHs at second level
create clusters from CHs of level one. TL-LEACH scheme is
potentially more dispense therefore; the load of the network
on the sensors is well shared which results in long lived sensor
In PEGASIS  nodes form a chain to transfer data from
source to sink. In chain formation process each node connect
with next node. The chain formation process require global
knowledge of sensor nodes, hence, it is very difﬁcult to
implement this topology.
Another clustering based protocol is HEED in which CHs
are selected on the base of a probability. The probability of
a node to become CH is related to the residual energy of the
node. In HEED, it is possible that the nodes with minimum
residual energy acquire larger probability to become CH.
A PEGASIS based mobile sink scheme is proposed in .
The sink moves along its trajectory and stays for a sojourn
time at sojourn location to guarantee complete data collection.
A similar sink mobile based technique is proposed in .
SEP protocol is designed for heterogeneous nodes. Nodes
in SEP are heterogenous in terms of their initial energy, called
normal nodes and advance nodes. The probability to become
CH depends on the initial energy of the node. Performance of
SEP in multi level Heterogeneous networks is not good.
An Energy Efﬁcient Unequal Clustering (EEUC) protocol
is presented which tries to balance the energy consumption
of the network. EEUC divide the network ﬁeld into unequal
clusters. In EEUC, there are some nodes in network that are
not associated with any cluster, therefore, they are isolated
inside the network.
On adaptive energy-efﬁcient scheme for transmission
(EAST) is proposed in . This scheme use open-looping
feedback process for temperature-aware link quality estima-
tion, whereas closed-loop feedback process divides network
into three logical regions to minimize overhead of control
packets. In  Quadrature-LEACH (Q-LEACH) for ho-
mogenous networks is proposed. This scheme maximize the
throughput, lifetime of network and stability period of the
Latif et al.  presented Divide-and-Rule (DR) scheme.
DR technique used for static clustering also for the selection
of CH. This scheme avoids probabilistic selection of CH
instead it elects ﬁxed number of CH. Away Cluster Head
(ACH) prtocol for WSN is proposed in . This protocol
efﬁciently maximize the stability period and throughput. J.
Kulik et al.  proposed sensor Protocols for Information
Via Negotiation (SPIN). In SPIN, a node advertise its sensed
data to its neighbors about the kind of the data it sensed. An
interested neighboring node will send a request for a copy
of data to originating node. In this way, the entire nodes in
the network acquire this data. The drawback of this approach
is that, there is no guarantee of data delivery to each node
in the network because if the node is interested in data from
distant source node then data will not deliver to interested
node. This protocol is not suited for applications where reliable
data delivery priority is on top.
A hybrid protocol Hybrid Energy Efﬁcient Reactive Protocol
for WSN is proposed in . In this protocol, CH is selected
based on the residual energy of node and average energy of
Due to the fact that clustering protocols consume less
energy, these protocols for WSNs have gained extensive
acceptance in many applications. Many state of the art WSN
protocols exploit cluster based scheme at manifold levels
to minimize energy expenditures. CH in most cluster based
protocols is selected on the base of probability. It is not
obvious that CHs are distributed uniformly throughout the
sensor ﬁeld. Therefore, it is quite possible that the selected
CHs concentrate in one region of the network. Hence, a
number of nodes will not get any CHs in their environs.
Similarly some protocols used unequal clustering and try to
use recourses proﬁciently.
Multiple level clustering hierarchy has following major
• In multiple level schemes, one CH forward data to other
CH which relays data to BS. If relay CH is faraway,
than it is necessary for forwarder CH to transmit data
with high power.
• In clustering protocols, a member node decides itself
whether to become CH or not. It is possible that some
distant nodes are selected as CHs. Therefore, these nodes
consume lot of energy to forward data to BS. Hence,
these nodes will die early
In this article, our goal is to design a gateway based energy
aware multi-hop routing protocol. This approach meets the
• Network is divided into regions and aid of gateway
node reduces the average transmission distance. Hence,
it saves network energy and prolong network lifetime.
• CH selection in each region is independent of other
regions so, there is deﬁnitely a CH exist in each region.
IV. NETWORK MODEL
In this article, we assume S sensors which are deployed
randomly in a ﬁeld to monitor environment. We represent
the i-th sensor by s
and consequent sensor node set S= s
.We assume the network model shown in ﬁg 1.
Fig. 1: Network Model
• We deploy the BS faraway from the sensing ﬁeld. Sensor
nodes and the BS are stationary after deployment.
• A gateway node is deployed in the same network ﬁeld at
the centre of the network.
• Gateway node is stationary after deployment and
• We use homogeneous sensor nodes with same computa-
tional and sensing capabilities.
• Each sensor node is assigned with a distinctive identiﬁer
We use ﬁrst order radio model as used in  and . This
model represents the energy dissipation of sensor nodes for
transmitting, receiving and aggregating data. The transmitter
dissipates more energy then receiver as it requires more energy
for the transmitter electronics and ampliﬁer. On the other hand,
in receiver, only electronic circuit dissipate energy, as shown
in ﬁg 2.
Fig. 2: Radio Model
The energy required to transmit a data packet of k bits to a
distance d and to receive a data packet of k bits, is given as:
(k, d) = E
(k) + E
(k, d) = E
× k + E
× k × d
(k) = E
(k) = E
(k) = E
× k (2)
V. THE M-GEAR PROTOCOL
In this section, we present detail of our proposed protocol.
Sensor nodes have too much sensed data for BS to process.
Therefore, an automatic method of combining or Aggregating
the data into a small set of momentous information is
required  . The process of data aggregation also
termed as data fusion. In order to improve network lifetime
and throughput, we deploy a gateway node at the centre of
the network ﬁeld. Function of gateway node is to collect data
from CHs and from nodes near gateway, aggregation and
sending to BS. Our results ensure that network lifetime and
energy consumption improved with the expense of adding
gateway node. We add rechargeable gateway node because
it is on ground fact that the recharging of gateway node is
much cheaper than the price of sensor node.
A. Initial Phase
In M-GEAR, we use homogenous sensor nodes that are
dispersed randomly in network area. The BS broadcast a
HELLO packet. In response, the sensor nodes forward their
location to BS. The BS calculates the distance of each node
and save all information of the sensor nodes into the node
data table. The node data table consists of distinctive node
ID, residual energy of node, location of node and its distance
to the BS and gateway node.
B. Setup Phase
In this section, we divide the network ﬁeld into logical
regions based on the location of the node in the network. BS
divide the nodes into four different logical regions. Nodes
in region-one use direct communication and transmit their
data directly to BS as the distance of these nodes from BS
is very short. Similarly nodes near gateway form region-two
and send their data directly to gateway which aggregates
data and forward to BS. These two regions are referred to as
non clustered regions. All the nodes away from the gateway
node and BS are divided into two equal half regions. We call
them clustered regions. Sensor nodes in each clustered region
organize themselves into small groups known as clusters.
C. CH Selection
Initially BS divides the network into regions. CHs are
elected in each region separately. Let r
represent the number
of rounds to be a CH for the node S
. Each node elect itself
as a CH once every r
= 1/p rounds. At the start of ﬁrst
round all nodes in both regions has equal energy level and
has equal chance to become CH. After that CH is selected
on the basis of the remaining energy of sensor node and with
a probability p alike LEACH. in each round, it is required to
have n x p CHs. A node can become CH only once in an
epoch and the nodes not elected as CH in the current round
feel right to the set C. The probability of a node to (belongs
to set C) elect as CH increases in each round. It is required to
uphold balanced number of CHs. At the start of each round,
a node S
belongs to set C autonomously choose a random
number between 0 to 1. If the generated random number for
is less than a predeﬁned threshold T(s) value then
the node becomes CH in the current round.
The threshold value can be found as:
T (S) =
if s ∈ C
where P = the desired percentage of CHs and r = the current
round, C = set of nodes not elected as CH in current round.
After electing CHs in each region, CHs inform their role
to neighbor nodes. CHs broadcast a control packet using a
CSMA MAC protocol. Upon received control packet from
CH, each node transmits acknowledge packet. Node who ﬁnd
nearest CH, becomes member of that CH.
When all the sensor nodes are structured into clusters, each
CH creates TDMA based time slots for its member nodes.
All the associated nodes transmit their sensed data to CH in
its own scheduled time slot. Otherwise nodes switch to idle
mode. Nodes turn on their transmitters at time of transmission.
Hence, energy dissipation of individual sensor node decreases.
E. Steady-State Phase
In steady state phase, all sensor nodes transmit their sensed
data to CH. The CH collects data from member nodes,
aggregates and forwards to gateway node. Gateway node
receives data from CHs, aggregates and forwards to BS.
VI. PERFORMANCE EVALUATION
We assess the performance of our proposed protocol and
compare it with existing protocol in WSN, known as LEACH.
A. Simulation Setting
In order to appraise the performance of our proposed proto-
col, we simulated our protocol using MATLAB. We consider a
wireless sensor network with 100 nodes distributed randomly
in 100m X 100m ﬁeld. A gateway node is deployed at the
centre of the sensing ﬁeld. The BS is located faraway from the
sensing ﬁeld. Both gateway node and BS are stationary after
deployment. We consider packet size of 4000 bits. We compare
our protocol with LEACH protocol. To assess performance of
our protocol with LEACH, we ignore the effects caused by
signal collision and interference in the wireless channel. Table
1 presents the radio parameters.
B. Performance Parameters
In this subsection, we present performance metrics. In this
work, we evaluated three performance parameters given below.
1) Network lifetime: It is the time interval from the start
of the network operation till the last node die.
2) Throughput: To evaluate the performance of throughput,
the numbers of packets received by BS are compared with
the number of packets sent by the nodes in each round.
3) Residual Energy: The residual battery energy of network
is considered in order to analyze the energy consumption
of nodes in each round. Residual energy ensures graceful
degradation of network life.
C. Simulation Results and Analysis
In this subsection, we show the simulation results. We run
extensive simulations and compare our results with LEACH.
Next subsections give detail of each metric.
1) Network Lifetime: In ﬁg 3, we show the results of the
network lifetime. Nodes are considered dead after consum-
ing 0.5 joule energy. M-GEAR protocol obtains the longest
network lifetime. This is because the energy consumption
is well distributed among nodes. Network is divided into
logical regions and two of them are further sub divided
into clusters. M-GEAR topology balance energy consumption
among sensor nodes. On the other hand, in LEACH, nodes die
quickly as stability period of network ends. It is not evident
that predestined CHs in LEACH are distributed uniformly
throughout the network ﬁeld. Therefore, there is a possibility
that the selected CHs will be concentrated in one region of the
network. Hence, some nodes will not have any CHs in their
environs. Fig 3 shows interval plot of network lifetime with
99% conﬁdence interval. we note that, the results of M-GEAR
protocol are statically different and perform well.
0 500 1000 1500 2000 2500 3000
Percent of allive nodes
Fig. 3: Interval plot- Analysis of network lifetime
2) Throughput: Average packets sent to BS are assessed
through extensive simulations. Simulation results of M-GEAR
protocol illustrate increased throughput. Interval plots of M-
GEAR and LEACH in ﬁg 4 clearly depicts performance of
both protocols. To calculate throughput, we assume that CHs
can communicate freely with gateway node. Simulation results
show an increase throughput of 5 times then LEACH. Sensor
nodes near gateway send their data directly to gateway; simi-
larly nodes near BS transmit data directly to BS. Sensor nodes
in both regions consume less transmission energy therefore,
nodes stay alive for longer period. More alive nodes contribute
to transmit more packets to BS.
0 500 1000 1500 2000 2500 3000
Fig. 4: Interval plot- Analysis of Throughput
3) Residual Energy: Fig 5 shows average residual energy
of network per round. We assume that a node has 0.5 joule
energy. The total energy of 100 node network is 50 joule.
M-GEAR protocol yields minimum energy consumption than
LEACH. Fig 5 clearly depicts that our protocol outperforms
LEACH routing protocol in terms of energy consumption per
round. Deployment of gateway node at the centre and high
probability of CHs in all regions ensures minimum energy
0 500 1000 1500 2000 2500 3000
Fig. 5: Interval plot- Analysis of remaining energy
Emp 0.0013 pJ/bit/m4
Eda 5pJ/ bit
Message size 4000 bits
TABLE I: Network parameter
VII. CONCLUSION AND FUTURE WORK
We describe an energy-efﬁcient multi-hop routing protocol
using gateway node to minimize energy consumption of sensor
network. In this work, we divide the network into logical
regions. Each region use different communication hierarchy.
Two regions use direct communication topology and two
regions are further sub-divided into clusters and use multi-hop
communication hierarchy. Each node in a region elects itself
as a CH independent of other region. This technique encourges
better distribution of CHs in the network. Simulation results
shows that our proposed protocol performs well compared
to LEACH. In this work, we study the three performance
metrics: Network lifetime , Residual energy and throughput. In
future, we will study ETX link metrics and we will implement
this metric in our scheme as implemented and demonstrated
in   .
 Mainwaring, Alan, et al. “Wireless sensor networks for habitat monitor-
ing.” Proceedings of the 1st ACM international workshop on Wireless
sensor networks and applications. ACM, 2002.
 Burrell, Jenna, Tim Brooke, and Richard Beckwith. “Vineyard comput-
ing: Sensor networks in agricultural production.” Pervasive Computing,
IEEE 3.1 (2004): 38-45.
 Ye, Mao, et al. “EECS: an energy efﬁcient clustering scheme in wire-
less sensor networks.” Performance, Computing, and Communications
Conference, 2005. IPCCC 2005. 24th IEEE International. IEEE, 2005.
 Li, Chengfa, et al. “An energy-efﬁcient unequal clustering mechanism
for wireless sensor networks.” Mobile Adhoc and Sensor Systems
Conference, 2005. IEEE International Conference on. IEEE, 2005.
 Heinzelman, Wendi Rabiner, Anantha Chandrakasan, and Hari Balakr-
ishnan. “Energy-efﬁcient communication protocol for wireless microsen-
sor networks.” System Sciences, 2000. Proceedings of the 33rd Annual
Hawaii International Conference on. IEEE, 2000.
 Lindsey, Stephanie, and Cauligi S. Raghavendra. “PEGASIS: Power-
efﬁcient gathering in sensor information systems.” Aerospace conference
proceedings, 2002. IEEE. Vol. 3. IEEE, 2002.
 Younis, Ossama, and Sonia Fahmy. “HEED: a hybrid, energy-efﬁcient,
distributed clustering approach for ad hoc sensor networks.” Mobile
Computing, IEEE Transactions on 3.4 (2004): 366-379.
 Smaragdakis, Georgios, Ibrahim Matta, and Azer Bestavros. “SEP: A
stable election protocol for clustered heterogeneous wireless sensor
networks”. Boston University Computer Science Department, 2004.
 Loscri, V., G. Morabito, and S. Marano. “A two-levels hierarchy for
low-energy adaptive clustering hierarchy (TL-LEACH).” IEEE Vehicular
Technology Conference. Vol. 62. No. 3. IEEE; 1999, 2005.
 Jafri, Mohsin Raza, Nadeem Javaid, Akmal Javaid, and Zahoor Ali
Khan. “Maximizing the Lifetime of Multi-Chain PEGASIS Using Sink
Mobility.” World Applied Sciences Journal 21, no. 9 (2013): 1283-1289.
 M. Akbar, N. Javaid, A. A. Khan, Z. A. Khan, U. Qasim, “On
Modeling Geometric Joint Sink Mobility with Delay-tolerant Cluster-
less Wireless Sensor Networks”, 4th IEEE Technically Co-Sponsored
International Conference on Smart Communications in Network Tech-
nologies (SaCoNet’13) 2013, Paris, France.
 Tahir, M., et al. “On Adaptive Energy-Efﬁcient Transmission in WSNs.”
International Journal of Distributed Sensor Networks 2013 (2013).
 B. Manzoor, N. Javaid, O. Rehman, M. Akbar, Q. Nadeem, A. Iqbal, M.
Ishfaq, “Q-LEACH: A New Routing Protocol for WSN”, International
Workshop on Body Area Sensor Networks (BASNet-2013) in conjunc-
tion with 4th International Conference on Ambient Systems, Networks
and Technologies (ANT 2013), 2013, Halifax, Nova Scotia, Canada,
Procedia Computer Science, Volume 19, 2013, Pages 926-931, ISSN
 K. Latif, A. Ahmad, N. Javaid, Z. A. Khan and N. Alrajeh, “Divide-
and-Rule Scheme for Energy Efﬁcient Routing in Wireless Sensor
Networks”, The 4th International Conference on Ambient Systems,
Networks and Technologies (ANT 2013), 2013, Halifax, Nova Scotia,
Canada, Procedia Computer Science, Volume 19, 2013, Pages 340-347,
ISSN 1877-0509, http://dx.doi.org/10.1016/j.procs.2013.06.047. Proce-
dia Computer Science.
 N. Javaid, M. Waseem, Z. A. Khan, U. Qasim, K. Latif and A. Javaid,
“ACH: Away Cluster Heads Scheme for Energy Efﬁcient Clustering
Protocols in WSNs”, 2nd IEEE Saudi International Electronics, Commu-
nications and Photonics Conference (SIECPC 13), 2013, Riyadh, Saudi
 Heinzelman, Wendi Rabiner, Joanna Kulik, and Hari Balakrishnan.
“Adaptive protocols for information dissemination in wireless sensor
networks.” Proceedings of the 5th annual ACM/IEEE international
conference on Mobile computing and networking. ACM, 1999.
 N. Javaid, S. N. Mohammad, K. Latif, U. Qasim and Z. A. Khan,
M. A. Khan, “HEER: Hybrid Energy Efﬁcient Reactive Protocol for
Wireless Sensor Networks”, 2nd IEEE Saudi International Electronics,
Communications and Photonics Conference (SIECPC 13), 2013, Riyadh,
 Heinzelman, Wendi B., Anantha P. Chandrakasan, and Hari Balakr-
ishnan. “An application-speciﬁc protocol architecture for wireless mi-
crosensor networks.” Wireless Communications, IEEE Transactions on
1.4 (2002): 660-670.
 Mcmullen, Sonya A. “Mathematical Techniques in Multisensor Data
Fusion 2nd Edition.” (2004).
 Klein, Lawrence A. “Sensor and data fusion concepts and applications.”
Society of Photo-Optical Instrumentation Engineers (SPIE), 1993.
 N, Javaid. A, Khan. I. A, Djouani. K, “Performance study of ETX
based wireless rout- ing metrics,” 2nd IEEE International Conference on
Computer, Control and Communications (IC4-2009), Karachi, Pakistan,
 Javaid, N.; Bibi, A.; Djouani, K., ”Interference and bandwidth adjusted
ETX in wireless multi-hop networks,” GLOBECOM Workshops (GC
Wkshps), 2010 IEEE , vol., no., pp.1638,1643, 6-10 Dec. 2010.
 Javaid, N.; Ullah, M.; Djouani, K., ”Identifying Design Requirements
for Wireless Routing Link Metrics,” Global Telecommunications Con-
ference (GLOBECOM 2011), 2011 IEEE , vol., no., pp.1,5, 5-9 Dec.