EM-SEP: An efficient modified stable election protocol

Abstract

Abstract— Recently wireless sensor networks (WSN) became an interesting topic because of its increasing usage in many fields; medical systems, environment monitoring, military applications and video surveillance. Usually sensors are placed in the desired locations to gather information frequently and then transfer it to the observers. WSN consists of a collection of application specific sensors, a wireless transceiver and a simple general purpose processor. In heterogeneous wireless sensor network, researchers found many challenging issues including the limited energy, the efficient usage of the energy, and the problem with the hierarchy of the network as imbalance network. Many studies indicated that the node clustering is a promising solution for such issues. Clustering has been shown to increase the efficacy of the energy consumption where clusters are formed dynamically with neighboring sensors and the power is assumed to be distributed equally among nodes. One of the nodes is considered as the cluster head that is responsible for transferring data among the neighboring sensors. In this work, we propose a modification based on SEP protocol. EM-SEP aims to prolong the stable period of the sensor network by maintaining balanced energy consumption. This means that we choose the advanced nodes to become cluster heads more often than the normal nodes as the case with SEP protocol. Furthermore, EM-SEP takes in account the number of nodes that are associated with each cluster head. Another important enhancement of EM-SEP protocol that if there are more than one sensor available to be a cluster head at certain round, we choose the sensor with highest energy.

Full text

Proceedings of 2014 Zone 1 Conference of the American Society for Engineering Education (ASEE Zone 1)
978-1-4799-5233-5/14/$31.00 ©2014 IEEE
Abstract— Recently wireless sensor networks (WSN) became
an interesting topic because of its increasing usage in many
fields; medical systems, environment monitoring, military
applications and video surveillance. Usually sensors are placed in
the desired locations to gather information frequently and then
transfer it to the observers. WSN consists of a collection of
application specific sensors, a wireless transceiver and a simple
general purpose processor. In heterogeneous wireless sensor
network, researchers found many challenging issues including
the limited energy, the efficient usage of the energy, and the
problem with the hierarchy of the network as imbalance
network. Many studies indicated that the node clustering is a
promising solution for such issues. Clustering has been shown to
increase the efficacy of the energy consumption where clusters
are formed dynamically with neighboring sensors and the power
is assumed to be distributed equally among nodes. One of the
nodes is considered as the cluster head that is responsible for
transferring data among the neighboring sensors. In this work,
we propose a modification based on SEP protocol. EM-SEP aims
to prolong the stable period of the sensor network by maintaining
balanced energy consumption. This means that we choose the
advanced nodes to become cluster heads more often than the
normal nodes as the case with SEP protocol. Furthermore, EM-
SEP takes in account the number of nodes that are associated
with each cluster head. Another important enhancement of EM-
SEP protocol that if there are more than one sensor available to
be a cluster head at certain round, we choose the sensor with
highest energy.
Keywords__ Wireless Sensor, Clusters, SEP, Energy
Consumption, EM-SEP
Manuscript received February 7, 2014.
A. Abu Malluh is with the Computer Science and Engineering Department,
University of Bridgeport, Bridgeport, CT 06604, USA (e-mail:
aabumall@bridgeport.edu ).
K. M. Elleithy is the Associate Dean for Graduate Studies in the School of
Engineering at the University of Bridgeport. He is with the Computer Science
and Engineering Department, University of Bridgeport, Bridgeport, CT
06604, USA (e-mail: elleithy @bridgeport.edu ). He is IEEE senior member.
Z. Qawaqneh is with the Computer Science and Engineering Department,
University of Bridgeport, Bridgeport, CT 06604, USA (e-mail:
zqawaqne@bridgeport.edu ).
R. J. Mstafa was with the Computer Science Department, University of
Zakho, Duhok, Iraq. Now he is with the Computer Science and Engineering
Department, University of Bridgeport, Bridgeport, CT 06604, USA (e-mail:
rmstafa@bridgeport.edu ).
A. Alanazi is with the Computer Science and Engineering Department,
University of Bridgeport, Bridgeport, CT 06604, USA (e-mail:
aalanazi@bridgeport.edu ).
I. INTRODUCTION
WSN is an emerging technology that helps to bring
enormous changes in data collection, processing and
dissemination for different environments and applications [1,
2]. Nodes have limited power batteries which shorten its
lifespan and battery replenishment is often not practical.
Increasing the network lifetime depends on management of
the sensing nodes energy resource. The wireless sensors can
be deployed in large different places to form wireless sensor
networks for a wide variety of purposes. Such deployment can
be random that can be dropped from an airplane or installing
fire alarm sensors in a building [3]. The lifetime of the sensor
is limited by power that every sensor has in the network and
will determine the validity of the sensing task. In [4],
distributed cluster-based routing protocol for sensor network
called Low-Energy Adaptive Clustering Hierarchy (LEACH)
is proposed. The authors developed Low-Energy Adaptive
Clustering Hierarchy (LEACH) protocol which works for
micro sensor networks that link the ideas of energy-efficient
cluster-based routing and media access together with
application-specific data aggregation to result in better
performance in terms of system lifetime, latency, and
application-perceived quality. Leach contains a distributed
cluster which helps to organize a large numbers of nodes.
Also, it includes the algorithms used for adapting clusters and
rotating heads positions to distribute equally the power
between all the nodes in the network. Also, Leach enables the
signal distribution among nodes in order to increase the
communication performance. It is shown that the Leach
increases the lifetime of the system by an order of magnitude
compared to other approaches.
To design an efficient protocol for wireless sensor
networks, the following properties of wireless sensor network
must be taken into consideration:
A) Architecture of the deployment: wireless sensor
networks consist of hundreds of nodes deployed in different
environments. Such environments might not be easy to reach
weather they are dangerous or located in distant places. In
such cases, sensors enable users to communicate and exchange
data with other users even if there is no existing network
infrastructure.
B) The lifetime of the system: the network should remain
alive as long time as possible to have better performance since
EM-SEP: An Efficient Modified Stable Election
Protocol
Arafat Abu Malluh, Khaled M. Elleithy, Zakariya Qawaqneh, Ramadhan J. Mstafa, Adwan Alanazi

recharging or changing the batteries might not be feasible. All
nodes must be designed to prolong the lifetime of the system
for better efficient.
C) Latency: latency means how much time the data takes to
be retrieved from the other nodes. In wireless sensor network,
the time is sensitive and data must be received within certain
time to have scalable system.
D) Quality: the user does not need redundant information
because it is connected to information of neighboring nodes
and user needs only the information from a higher-level
description of events occurring in the monitored environment.
In general, the quality of the networks depends on the
quality of the data which means the protocols must be
designed guarantee a specific level of quality of the data.
The rest of the paper is organized as follows. Section 2
presents some of the related work. Section 3 gives an
overview of the main characteristics of the Leach protocol
architecture and how it works. Section 4 we addresses the
Optimal Clustering. Section 5 presents SEP protocol and the
stable election protocol based on the energy of normal and
advanced sensors. Section 6 introduces performance measures
that we have used. Section 7 provides the design details of
EM-SEP protocol. Section 8 shows a comparison between
SEP and EM-SEP. Section 9 provides the conclusions.
II. RELATED WORK
Clustering algorithms for wireless sensor networks can be
categorized in two broad types : (1) probabilistic algorithms
that run in synchronized rounds which are mostly used for
static networks [4], and weight-based algorithms [5], which
are used in dynamic networks[6]. In this section, we explain
the weight-based clustering algorithms.
Weight-based algorithms can be constructed as overlapping
cluster where every node can belong to more than one cluster.
Also, it can be constructed as disjoint cluster where a node
only belongs to one cluster. In the overlapping cluster, the
used algorithms choose set of nodes, termed gateway or
border nodes. Each set is connected to at least two adjacent
clusters. In [7], the LCA clustering algorithm is discussed
where every node chooses its own cluster head from the
neighboring cluster head with the smallest ID and any node
that can hear two or more cluster heads is called as a gateway.
Disjoints clusters can be constructed where each node must
share its information with its cluster head (such as the sensed
date or sensor id). The cluster head uses this information of
the node to select the cluster head nodes some protocols make
their selections based on the completion of the information for
certain number of hops or for the whole network. In[8], the
max-min D cluster algorithm uses the d-hop information for
cluster head selection. After finishing the selection process,
the nodes are at most d hops away from the cluster head.
In [6], DMAC is presented that partitions the nodes of a
fully mobile network (ad hoc network) into clusters, thus
giving the network a hierarchical organization. Nodes are
grouped by following new weight-based criteria that allows
the choice of the nodes that coordinate the clustering process
based on node mobility-related parameters. Like DSA, nodes
choose the role based on the information from the neighboring
hop. However, DMAC is not implemented in different phases.
Every node responds locally to the surrounding topology. The
topology can be represented by addition and deletion of links
to its neighboring nodes. The cluster head is the node with
highest weight between its unassigned neighbors.
In [9], the authors generalized DMAC as G-DMAC, where
a new node that joins the cluster head with the highest weight
in its one-hop neighborhood (like DMAC). The topology is
changeable but the node is still considered as member of this
cluster head v as long as there is no other neighboring cluster
head u with weight w(u) > w(h) + h, given the parameter h ≥
0. Another parameter k is introduced that defines the
maximum number of cluster head neighbors that a cluster head
is allowed to have.
The C4SD clustering algorithm[10] is proposed for mobile
network where the node’s capability and the degree of
dynamics represent the weight. In such a network, each node
chooses the neighbor with the highest capability grade as
parent. If such a node does not exist, this node itself is
considered as a cluster head. The result of this algorithm can
be represented by disjoint clusters where cluster heads form an
independent set.
In[8], a new algorithm called Tandem is proposed for
spontaneous clustering of mobile wireless sensor nodes facing
similar context (such as moving together) where every node
runs a shared-context recognition algorithm, which provides a
number on a scale, representing the confidence value that two
nodes are together. Every node periodically makes sure it
shares the same context with its neighbors. The selection of
cluster heads is weight-based: the node with the highest
weight among its neighbors with which it shares a common
context is considered as a cluster head. A regular node joins
the cluster head with the highest weight and the one that
shares a common context with it.
III. LEACH PROTOCOL ARCHITECTURE
Leach protocols work in wireless micro sensor networks to
monitor a remote environment. Since data is correlated among
neighboring nodes in the network, the user doesn't need the
redundant data but requires a high-level function of the data
that describes the events occurring in the environment [4, 11].
The data signal is strongly correlated among nodes specially
the one close to each other. The clustering aspect is considered
as the main principle for Leach. This enables the data to be
transferred and processed among nodes in the same cluster
which leads to decreasing the data set that must be transferred
to another user. Data aggregation techniques is an option to
combine different correlated information signals to be smaller
data sets that results effective and efficient data [12]. In
addition, this technique reduces the amount of transmitted data
between the cluster and the base station.

In Leach, the cluster is formed locally by the nodes and one
of the nodes is elected to be the head of the cluster. In each
round there is a new selected head for the cluster which means
distributing the load equally among the nodes in the same
cluster. The rest of the nodes that are not cluster head transfer
their data to the head and they implement a signal procession
functions on the data and transfer the data to the base station.
The cluster head node consumes more power than other nodes
since is responsible for the commutation in the cluster. The
power load should be distributed evenly between the nodes in
the same cluster. If the heads are selected in advance during
the system lifetime, this causes wastage of power and that
node loses its total energy in short time. If the node loses its
energy, then the node is no longer operational and the non-
head nodes in the same cluster will not able to communicate
with each other. For this reason, the selection of the head
should not be fixed throughout the lifetime of the systems and
should be randomly selected to reduce the energy
consumption of the sensors in the network.
LEACH operates in rounds. Every round starts with a set-up
phase when the clusters are organized, then a steady-state
phase when the information is transferred between the nodes
and the head and between the head and the base station. The
nodes are considered to be homogeneous when all the nodes
have the same initial power. However, in Leach protocol every
node will be a head for the cluster every 
 round which is
called as epoch for the cluster wireless network[13].
It is assumed that every node can be a cluster head with a
probability of popt. In every round there is a new cluster head
instead of the one in the previous round. The set G contains
the rest of the nodes that are non-head nodes to keep the
cluster head constant for each round. Every round r, a new
head is selected when every node s  G selects a number
randomly in [0,1] . The node is considered as a head for the
cluster if that number is less than the threshold T(s).
T(s) = 
 

 if s€ G (1)
IV. OPTIMAL CLUSTERING
Many studies have been discussed the optimal clustering
with different results. In [7], the authors proposed a new
distributed randomized clustering algorithm for organizing the
sensors in the wireless sensor network in clusters while
maintaining the hierarchy of cluster heads. It is assumed that
the communication environment is contention-based and
without error. The results show that total energy usage is
reduced with increasing the number of levels in the cluster
while transmitting the data to the data processing center.
Another algorithm for optimal clustering is proposed in[14] to
organize the sensors in a wireless sensor network into clusters
to minimize the consumed power in the system while
exchanging the data in the network. The authors evaluate the
number of cluster heads and cluster diameters that minimize
the energy consumed in the network. The power used to
collect information at a central node from all sensors in a
clustered network is found to be significantly lower than the
Max–Mind cluster algorithm. Also, the results show that
energy used in the cluster with five levels can be as low as
25% of the power consumption in a non-clustered network
over a large area. This clustering is optimal when energy is
evenly distributed among all sensors and the total energy
consumption is minimal.
Fig.1 [7] shows the radio energy dissipation model and to
get acceptable Signal-to-Noise Ratio (SNR) in transmitting an
L−bit message over a distance d, the energy expended by the
radio is shown below:
ETx(l, d) =L · Eelec + L · _fs· d2 if d < d0 (2)
L · Eelec + L · _fs · d2 if d < d0
Where Eelec is the energy consumed per bit to transmit data
between sender and receiver, _fs and _mp depend on the
transmitter amplifier model we use, and d represents the
distance between the sender and the receiver.
Fig. 1. Radio Energy Dissipation Model [7]
Where Eelec is the energy consumed per bit to transmit data
between sender and receiver, _fs and _mp depend on the
transmitter amplifier model we use, and d represents the
distance between the sender and the receiver.
By assuming that d = d0, we can get:
d0 = 
 (3)
We assume an area A = M×M square meters, n is the
number of nodes which are randomly distributed in that area
and the sink is in the center of the field. Also, it is assumed
that the distance between the sink and any node is ≤ d0.
The energy consumed in the head node for each round can
be represented in Equation (4):
ECH = L·Eelec (
 -1) + L .EDA 
· _fs toBS (4)
Where k represents the number of clusters, EDA is the
processing (data aggregation) cost of a bit per signal, and
dtoBS is the distance between the cluster head and the sink.

The energy used in a non cluster head node can be computed
as:
EnonCH = L · Eelec + L · _fs · d2 toCH (5)
Where dtoCH represents the distance between a cluster
member and its cluster head. Assuming that the nodes are
uniformly distributed, it can be shown that:
E[d2toCH] =  
 (6)
where ρ(x, y) represents the node distribution.
The energy consumed in a cluster per round can be
computed as in the following formula:
Ecluster ≈ ECH + 
 (7)
The total energy in the network is :
Etot = L.(2n Eelec + nEDA + € fs(k . toBS + 
 )) (8)
The optimal number of constructed clusters is equal to:
Kopt = 

 

 (9)
Because the average distance from a cluster head to the sink
as in [7]:
E[dtoBS] =  


 (10)
The optimal probability of a node to become a cluster head,
popt, can be calculated as follows:
Popt = 
 (11)
V. SEP PROTOCOL
SEP protocol proposed a new solution which is called the
stable election protocol. SEP is based on the energy of normal
and advanced sensors. SEP approach proposes two different
(Popt) weighted optimal election probability; one for the
normal nodes and the other for the advanced nodes. SEP
defines the (Pnrm) which is the weighted election probability
for normal nodes, and the (Padv) the weighted election
probability for the advanced nodes. Therefore, the weighted
probabilities of normal and sensors are formed as following
respectively.
 
 (12)
 
 (13)

SEP define two different thresholds as well. . T(Snrm) is the
normal nodes threshold. On the other hand, T(Sadv) the
threshold for advanced nodes. Hence, the normal and
advanced nodes equation threshold consequently is
  
 


   (14)
  
 


  (15)
where r is the current round, G’ is the set of the normal
nodes that have not become cluster heads within the last
(1/Pnrm) round.
VI. PERFORMANCE MEASURES
We have identified the following parameters to measure the
performance of the clustering protocols:
1. Stability period: known as a stable region and it
refers to the duration of time when the network is
operational.
2. Instability Period: known as unstable region and it
refers to the time between the death of the first and last
node
3. The lifetime of the network: the time between the
beginning of the network and the death of the last node.
4. Number of cluster head per round: refers to the
number of nodes that can communicate with sink from
their cluster members.
5. Alive node every round: refers to the number of
nodes that still have not used their total energy.
6. Throughput: refers to the rate of the data sent in the
network, the data exchanged between the nodes and
cluster head and between the cluster head and the sink.
There is a relationship between the lifetime of system and
reliability. If there is at least one node alive, we can receive an
update regarding the sensor’s field but these updates may not
be reliable because we cannot guarantee that the existence of
one cluster head per round in the last round.
VII. EM-SEP: AN EFFICIENT MODIFIED STABLE ELECTION
PROTOCOL
In this section we describe EM-SEP protocol, which
improves the stable period of clustering hierarchy in the sensor
networks. The modification is based on the SEP algorithm as
described in the previous section. This modification aims to
prolong the stable period of sensor network by maintaining
well balanced energy consumption. This means that we choose
the advanced nodes to become cluster heads more often than
the normal nodes as the SEP protocol does. Furthermore, we
take into account the number of nodes that are associated with
each cluster head. Our modification tries to evenly distribute
the nodes between the selected cluster heads. Another
important enhancement in our work that if there are more than
one sensor available to be a cluster head at certain round, our

technique chooses the sensor with the highest energy. Those
two factors prolong the stable period of the sensor network as
shown in the result section.
VIII. ANALYSIS OF EM-SEP PROTOCOL
EM-SEP have been simulated over 100 * 100 m2 field with
100 nodes sensor network randomly located and for other
experiments we increases the number of nodes to 150. For the
sink location, we simulated the environment where the sink
located in center of the area and for other scenarios it was
placed in the corner of the area. Hence, the location of each
sensor is randomly selected and uniformly distributed over the
field dimensions. All sensors sense and transmit 4000 bits
packet every round. The nodes aggregate the data they receive
from other nodes with their own data, and produce only one
packet regardless of the number of received packets. The base
station (sink) is considered as a cluster head. Therefore each
sensor sends the sensed data to the closest cluster head. All
sensors start with 0.5 J. The results are compared with SEP
algorithm of clustering heterogeneous sensor networks using
the following factors:
• Stability period (the time where the first
sensor dies).
• Number of nodes alive per round.
• Number of dead nodes per round.
• Number of packets sent to base station.
• Number of packets sent to cluster head.
A. Stable Period
From the analysis of our work results, it is found that EM-
SEP prolongs the time interval before the death of the first
node (stability period) more than the SEP algorithm. EM- SEP
extended the stable period compared to SEP by 5%. Fig. 2
shows results for EM- SEP and SEP where m = 0.1 and a = 1,
and the base station is located in the center of 100 m2 area.
Fig. 2 demonstrates that EM-SEP prolongs the life time of
the network by 5%. To verify the results, we repeated the
previous experiment by changing the location of the base
station to be located in the corner of the area (90x90). Fig. 3
shows that EM-SEP algorithm result is better than the SEP
algorithm. Fig. 4 shows the comparison between EM- SEP and
SEP while increasing the number of nodes.
A. Throughput
Fig. 5 and Fig. 6 show that the throughput of EM-SEP is
close enough to the throughput of SEP in most of the rounds.
We have measured the total number of packets sent to both the
base station and the cluster heads. It is found that EM- SEP
throughput is little bit less than SEP throughput, but EM-SEP
is much better to lengthen the sensors time. The number of
alive nodes per round and the number of dead nodes per round
are shown in Fig. 7 and Fig. 8 respectively. EM-SEP has
better number of alive nodes per round, and the number of
dead nodes is less than the SEP algorithm.
It is important to note that for more than 50 tests of EM-
SEP, the results showed that no sensor is dead before round
950. Meanwhile, SEP results showed that every 10 tests there
are roughly 2 tests where the first dead sensor takes place in
800s.
Fig. 2. First dead node (m=0.1, a=1 and BS located 50 X 50)
Fig. 3. First dead node (m=0.1, a=1 and BS located 90 X 90)
Fig. 4. First dead node (m=0.1, a=1 and BS located 90 X 90)

Fig. 5. Number of packets to BS (m=0.1, a=1 and BS located 50 X 50)
Fig. 6. Number of packets to CH (m=0.1, a=1 and BS located 50 X 50)
Fig. 7. Number of alive nodes (m=0.1, a=1 and BS located 50 X 50)
Fig. 8. Number of dead nodes (m=0.1, a=1 and BS located 50 X 50)
IX. CONCLUSIONS
In this paper, we propose a modification to the SEP
protocol to reduce the communications overhead by saving
power and extending the network life time. EM-SEP protocol
deals with the network as a number of clusters while
introducing an efficient mechanism for communications
among nodes. EM- SEP protocol increases the stable period
of the sensor network by evenly distributing the power
consumption among the nodes in the cluster. In addition, we
propose a solution in case there are a number of candidate
nodes to be the cluster head at any given point of time. EM-
SEP protocol suggests that the node with the highest energy
should be the cluster head.
REFERENCES
[1] P. Santi, "Topology Control in Wireless Ad Hoc and Sensor Networks,"
Istituto di Informatics e Telematica del CNR, Italy, 2005
[2] I. F. Akyildiz, et al., "A survey on sensor networks," Communications
Magazine, IEEE, vol. 40, pp. 102-114, 2002.
[3] E. J. Duarte-Melo and L. Mingyan, "Analysis of energy consumption and
lifetime of heterogeneous wireless sensor networks," in Global
Telecommunications Conference, 2002. GLOBECOM '02. IEEE, 2002,
pp. 21-25 vol.1.
[4] W. B. Heinzelman, et al., "An application-specific protocol architecture
for wireless microsensor networks," Wireless Communications, IEEE
Transactions on, vol. 1, pp. 660-670, 2002.
[5] M. Chatterjee, et al., "WCA: A Weighted Clustering Algorithm for Mobile
Ad Hoc Networks," Cluster Computing, vol. 5, pp. 193-204, 2002.
[6] S. Basagni, "Distributed Clustering for Ad Hoc Networks," presented at
the Proceedings of the 1999 International Symposium on Parallel
Architectures, Algorithms and Networks, 1999.
[7] A. Ephremides, et al., "A design concept for reliable mobile radio
networks with frequency hopping signaling," Proceedings of the IEEE,
vol. 75, pp. 56-73, 1987.
[8] R. Niati, et al., "A key management scheme for dividing/merging cluster
based wireless sensor networks," in Networks, 2005. Jointly held with the
2005 IEEE 7th Malaysia International Conference on Communication.,
2005 13th IEEE International Conference on, 2005, p. 5 pp.
[9] A. D. Amis, et al., "Max-min d-cluster formation in wireless ad hoc
networks," in INFOCOM 2000. Nineteenth Annual Joint Conference of
the IEEE Computer and Communications Societies. Proceedings. IEEE,
2000, pp. 32-41 vol.1.
[10] E. M. Saad, et al., "Adaptive and Energy Efficient Clustering
Architecture for Dynamic Sensor Networks," in Soft Computing
Applications, 2007. SOFA 2007. 2nd International Workshop on, 2007,
pp. 221-225.
[11] W. R. Heinzelman, et al., "Energy-efficient communication protocol for
wireless microsensor networks," in System Sciences, 2000. Proceedings
of the 33rd Annual Hawaii International Conference on, 2000, p. 10 pp.
vol.2.
[12] D. L. Hall, "Mathematical Techniques in Multisensor Data Fusion,"
1992.
[13] I. M. Georgios Smaragdakis, Azer Bestavros, "SEP: A Stable Election
Protocol for clustered heterogeneous wireless sensor networks," Technical
Report BUCS-TR, 2004.
[14] S. Bandyopadhyay and E. J. Coyle, "Minimizing communication costs in
hierarchically clustered networks of wireless sensors," in Wireless
Communications and Networking, 2003. WCNC 2003. 2003 IEEE, 2003,
pp. 1274-1279 vol.2.

Arafat Abu Mallouh
Arafat Abu Mallouh is originally from Jordan. He is pursuing his
Doctorate in Computer Science and Engineering at the University of
Bridgeport in Bridgeport, Connecticut, USA. He received his Bachelor’s
degree in Computer Science from The Hashemite University, Zarqa, Jordan.
Mr. Abu Mallouh received his Master’s degree in Computer Science from
Amman Arab University for Graduate Studies, Amman, Jordan. His research
interests include artificial intelligence, image processing, Machine Learning,
and Data Mining.. Currently Mr Abu Mallouh works on new techniques for
voice processing.
Khaled M. Elleithy
Dr. Elleithy is the Associate Dean for Graduate Studies in the School of
Engineering at the University of Bridgeport. He has research interests are in
the areas of network security, mobile communications, and formal approaches
for design and verification. He has published more than two hundred and fifty
research papers in international journals and conferences in his areas of
expertise.
Dr. Elleithy is the co-chair of the International Joint Conferences on
Computer, Information, and Systems Sciences, and Engineering (CISSE).
CISSE is the first Engineering/Computing and Systems Research E-
Conference in the world to be completely conducted online in real-time via
the internet and was successfully running for six years. Dr. Elleithy is the
editor or co-editor of 12 books published by Springer for advances on
Innovations and Advanced Techniques in Systems, Computing Sciences and
Software.
Zakariya Qawaqneh
Zakariya Qawaqneh, originally from Ajloun, Jordan, is pursuing his
Doctorate in Computer Science and Engineering at the University of
Bridgeport in Bridgeport, Connecticut, USA. He received his Bachelor’s
degree in Computer Science and Application from The Hashemite University,
Zarqa, Jordan. Mr. Qawaqneh received his Master’s degree in Computer
Science from Jordan University of Science and Technology, Irbid, Jordan. His
research interests include artificial intelligence, image processing, computer
languages, and network security.
Ramadhan J. Mstafa
Ramadhan Mstafa is originally from Dohuk, Kurdistan Region, Iraq. He is
pursuing his Doctorate in Computer Science and Engineering at University of
Bridgeport, Bridgeport, Connecticut, USA. He received his Bachelor’s degree
in Computer Science from University of Salahaddin, Erbil, Iraq. Mr. Mstafa
received his Master’s degree in Computer Science from University of Duhok,
Duhok, Iraq. His research interests include image processing, mobile
communication, security and steganography.
Adwan Alanazi
Adwan Alanazi is originally from Saudi Arabia He is pursuing his
Doctorate in Computer Science and Engineering at the University of
Bridgeport in Bridgeport, Connecticut, USA. He received his Bachelor’s
degree in Computer Science from University of Hail, Hail, Saudi Arabia. Mr.
Alanazi received his Master’s degree in Computer Science from University of
Missouri Kansas City. His research interests include Wireless Sensor
Networks and Network Security.