ArticlePDF Available

Hierarchical Cluster-Based FIFO Asynchronous Data Transfer Technique for Reducing Congestion for Energy Efficient State Wireless Sensor Network-HAEEW

Authors:

Abstract and Figures

The applications of WSN can be quiet numerous. In applications like battlefield monitoring, grid power generation, health systems, sensors are deployed on large scale. During such deployment, energy efficiency must be proficient, which requires clustering, in the WSN architecture. Clustering architecture requires maintenance of sensor nodes due to malfunctioning of sensor which becomes depleted of energy. As some nodes leaves and some are being replaced, congestion is introduced in the network due the limited processing capability of memory, computations, and bandwidth condition. This paper proposes one of the energy efficient clustering techniques (HAEEW), using asynchronous data transfer (ADT), which has been modeled from data transfer technique (EEHCR), and using hierarchical clustering. Our model uses synchronization in clock time queries in one and each iterations round time, to determine cluster head, and head-set member formation, using Ad hoc on-demand energy aware routing protocols (AOERP) to make decision. In each iteration, the head-set members receives message request from neighboring nodes to confirm their average distance estimation, in which to transmit aggregated data to the base station. In a sensor deployment, which is aimed for data collection, control and management of sensor nodes, play a vital role, where nodes can be adjusted to boost energy in the network life time. We used matlab for simulations analysis of our result.
Content may be subject to copyright.
International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 622
ISSN 2229-5518
IJSER © 2015
http://www.ijser.org
Hierarchical Cluster-Based FIFO Asynchronous Data
Transfer Technique for Reducing Congestion for
Energy Efficient State Wireless Sensor Network-
HAEEW
Samuel Erskine
Computer Science and Engineering
University of Bridgeport
Bridgeport, USA
serskine@bridgeport.edu
Khaled Ellethy
Computer Science and Engineering
University of Bridgeport
Bridgeport, USA
kelliethy@bridgeport.edu
Linfeng Zhang
Computer Science and Engineering
University of Bridgeport
Bridgeport, USA
Lzhang@bridgeport.edu
Abstract The applications of WSN can be quiet numerous. In applications like battlefield monitoring, grid power generation, health
systems, sensors are deployed on large scale. During such deployment, energy efficiency must be proficient, which requires clustering,
in the WSN architecture. Clustering architecture requires maintenance of sensor nodes due to malfunctioning of sensor which
becomes depleted of energy. As some nodes leaves and some are being replaced, congestion is introduced in the network due the
limited processing capability of memory, computations, and bandwidth condition.
This paper proposes one of the energy efficient clustering techniques (HAEEW), using asynchronous data transfer (ADT), which has
been modeled from data transfer technique (EEHCR), and using hierarchical clustering. Our model uses synchronization in clock time
queries in one and each iterations round time, to determine cluster head, and head-set member formation, using Ad hoc on-demand
energy aware routing protocols (AOERP) to make decision. In each iteration, the head-set members receives message request from
neighboring nodes to confirm their average distance estimation, in which to transmit aggregated data to the base station. In a sensor
deployment, which is aimed for data collection, control and management of sensor nodes, play a vital role, where nodes can be
adjusted to boost energy in the network life time. We used matlab for simulations analysis of our result.
KeywordsWSN architecture, energy efficiency, clustering, asynchronous data transfer, base station, application
—————————— ——————————
1 INTRODUCTION
Recent emerging technology in Wireless Sensor Network
(WSN) has become so ubiquitous such that, requirements in
application delivery such as battlefield monitoring, smart power
grid and health monitoring issues are critical. Based upon that,
the application delivery design in data reliability practices must
be robust. The WSN data transfer a technique which is deployed
in multihop transfer, usually is not coped with time
synchronization of clocks with the sensor data transmissions.
Asynchronous data transfer, uses FIFO buffer techniques in the
sensor nodes which use synchronized clocks for reducing
congestion, as a result of node changes in clustering architecture.
This does not depend on point-to-point address-based data
transfer, which incurs a lot of overhead message and introduce
congestion. Based upon this the need to reduce congestion for
energy efficient sensor network due to the changes made to the
data transfer, must be considered, especially from a particular
sensor node distance to base station is required. Based upon the
fact that sensor nodes must be deployed in larger scale, it is also
an accepted fact that the sensor is a small device with limited
memory, power and computational capabilities.
In order for more collaboration for efficient data delivery,
energy efficient techniques using hierarchical clustering and
asynchronous data transfer, is required, in clustered deployment
of the sensors. This must have a significant impact in the design
such that reliability in packet flow must overcome any
congestion for a particular network deployment connection.
Moreover, the wireless connection is envisaged as more prone to
errors due to much vulnerability in the air [1]. Therefore,
connectivity issues in the wireless multi-hop environment must
not be underestimated. Sufficient number of packet generation
must be constantly estimated. Sensor base station (sink)
communication should be continuous in order to overcome any
congestion encounter, that would constrain the limited memory
resources due to limited bandwidth issues in the connection.
We propose Asynchronous Data Transfer phase model
technique, in a clustered network, where the Wireless Sensor
node deployment, is capable of transmitting data, based upon
revised base station distance estimation technique, in transmitting
data from the sensor, which enhances energy efficient and data
control practices in clustered network architecture. In WSN
congestion reducing must benefit the application to secure more
data reliability for energy efficiency. Based upon this, the sensor
memory must be pruned for more bandwidth resolution.
IJSER
International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 623
ISSN 2229-5518
IJSER © 2015
http://www.ijser.org
The rest of the work shall be observed as follows: Section I
include introduction and the problem identification phase, Section
II consist of the related work in various congestion mitigation
algorithms in fast data transfer, and section III consist of the
proposed system design, which include Section V as the
description of hierarchical clustering architecture, section VI is
the Asynchronous data transfer model description, section VII is
result discussion, section VIII is conclusion, and section X is
reference.
1.1 Problem Identification
Wireless Sensor Network (WSN) technology has become
much ubiquitous, so that congestion control in a clustered
architecture, using cluster head, head-set members in architecture
deployment of large scale sensor node, is a widely accepted issue
which must be reviewed. Therefore, a sensitive application such
as battlefield surveillance, smart power grid systems, and health
system deployments, requires sufficient bandwidth for efficient
data transmissions. Random traffic generation improvement [2]
over packet delivery, including intermediate sensor nodes should
consequently enhance a packet delivery in all base station (sink)
transactions [3].
A major cause of congestion in WSN can be attributed to
the fact that the sensor node is smaller in size with corresponding
limited memory, power, and computational capability [4]. The
batteries which should be supplied to the sensor nodes usually
run out so quickly that they are depleted in no time. This means
they are not able to withstand long period of operations based on
particular sensor deployment situation. Based upon this sensor
deployment must adapt energy efficiency techniques, such as
hierarchical clustering. In applications such as the battle field
monitoring, smart grid power, etc. can be under threat if not well
secured. Data reliability and delivery techniques are important
for energy boost. Usually, in such situation, in order for more
data reliability to be achieved, energy efficient techniques, using
clustering architecture for sensor deployment for energy
efficiency must be proficient and proactive. When this is
achieved enough bandwidth will continuously occur in the
memory of the sensor with enhanced computational capability, in
order to sustain efficient energy state for the WSN.
2 RELATED WORK
In this section, we categorize the literature review under
five aspect of wireless sensor data control techniques and
algorithms. Our aim is to identify and investigate the issues
relating to those proposed algorithms, and try to deal with any
unresolved issues, that would not assist in achieving efficient and
reliable fast packet delivery. The aim is also to provide loss data
recovery in the network environment, that does not include
estimation of average precision distance to the base station,
regarding data transmissions. Therefore, it is important that fast
data control techniques that we describe, in relation to those
algorithms must address any congestion issue in the network, due
to replacements of new nodes in clustering formation, which are
depleted of their energy. Based upon this our related work can be
reviewed as follows.
Retransmission timeout (RTO) [5] is one of the algorithms
that can be used in data transfer in clustering network
architecture. RTO is proposed for enhancement in packet
delivery, and to increase the RTO value. Based upon this, Round
Trip Time (RTT) for packets uses a timeout mechanism to trigger
retransmission of packet which have not yet been acknowledged
(ACK) in packet transmission, after an expired timer occurs.
Based upon the standard de facto used, the transmission
algorithm tracked average RTT in addition to, and its product
with RTT mean deviation which represent RTO value for
subsequent packet derivation which has value 4 typically
assigned to.
If SRTT (k) is average smoothed RTT, then
 (1)
 (2)
(1) And (2) represents the smoothed average RTT and the
RTT mean deviation, which is also used in determining
the new RTO value.
2.1 RTO Determination:
Based upon the above description, in order for the next
packet to be generated, the timeout value need to be set by the
RTO. The RTT measurement should possibly be estimated
afterwards, which subsequently update the RTO based on above
two Eqns. (1) and (2).
This means, it is expected that timestamp must importantly
be activated for RTT to be tracked for every packet that was
transmitted. It was however
realized that only outstanding packet amongst the remaining
tracked RTT at all times, whilst the remaining other packet could
not be tracked. Based upon this, we ask an important question
that, why should insignificant number of packet could only be
used for tracking the RTO? This means, a timestamp issue
occurs, which causes not all packets to be tracked by all RTT.
This poses a conflicting design issue that need to be investigated.
Although, equations (1) and (2) were proposed to resolve the
issue, however the issue remains unresolved. We investigate
about that and try to resolve the issue.
Fuzzy Based Algorithm for Congestion Control (FBACC)
[2] was proposed for packet drop reduction, which occur over
intermediate nodes. Moreover, at the same time, this also
maximizes packets sent to the sink with source traffic rate
regulation known as Fine tuning the Fuzzy Bucket Token [6].
Based upon this, the average packet delivery represents QoS
which is an important metric factor that overcomes burst traffic.
An important characteristic of FBACC is that it is very useful for
applications that have limited memory, energy constraint, and
with main objectives of performing very well in fast data
transmission based on accepted packet level drop.
One important aim of this research is to investigate about
fast reliable data, using effective hierarchical clustering
congestion control technique, including asynchronous data
transfer model. This also applies to time sensitive and fast data
delivery application, such as battlefield monitoring of equipment
and personnel. It can also be applicable in health system and
smart power applications. This means that, an urgent requirement
of our design is that, it must be capable to prevent occurrence of
any congestion such as packet drop issue in clustered
architecture. Moreover, in critical condition like health
application situation, it would be unacceptable that a WSN
application should transmit data that still incur appreciable level
IJSER
International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 624
ISSN 2229-5518
IJSER © 2015
http://www.ijser.org
of congestion such as packet drop. This is because applications
like battlefield health, and power systems, which are critical and
time sensitive, require continuous data delivery from sensor to
base station, in order to maintain the energy efficiency state
using clustering architecture. In view of this, we deem it very
urgent to investigate about the issue of packet drop in our WSN
fast data transfer algorithm description.
APPLICATION
LAYER
TRANSPORT
LAYER
NETWORK
LAYER
MAC LAYER
PHYSICAL LAYER
MULTIHOP DATA
TRANSFER
PACKET
FORWARDING
FRAME
TRANSMISSION
Fig.1 HAEEW Protocol
Distributed Control Algorithm (DCCA) [7] proposed for the
WSN is congestion level detection based on queue detection
scheme in the MAC layer. This adopts Hop-by-Hop (HBH)
feedback notification scheme, with appreciable level of packet
drop, which also gives high overhead. This also merges with the
Transport Control Protocol Layer (TPL). Our fast data transfer,
and congestion mitigation technique, which ensures a reliable
data transmissions with energy efficiency design in mind, (Fig.1)
has been designed such that, it is based on feedback congestion
notification in a multihop WSN environment, and forward packet
instantly without any congestion encounter such as packet drop
from the Network Layer (NWL). Instantly, this is required to
transmit fast data immediately to the remaining sensor nodes via
the TPL. Meanwhile the NWL that is proposed in DCCA consist
of packet forward and packet drop [7]. Packet drop issue is a
severe congestion encounter in clustered network, are hostile to
design environment such as our protocol design. We therefore
seek to investigate about packet drop issue in our application that
should be addressed in our protocol layer design.
Furthermore, due to limited memory and size of wireless
sensor nodes, it must be carefully designed such that it protocol
layered approach must be unique, compared to wired network
layered approach. Based upon the description in WSN protocol
layered approach used in [8], it was designed according to
requirement of wired network layered protocol approach.
Therefore, channel estimation for the WSN is not generated to
suit the application. Since the reliable fast data application in the
wired/traditional network is end-to-end [8], it does not include
requirement in multihop feedback loss congestion notification
and loss recovery for the WSN. We require that channel
reliability in WSN must include use of multihop clustered
architecture which is energy efficient compared to hop-by-hop,
in order to solve the long distance latencies issue created in the
end-to-end wired layered approach. Channel estimate for sending
and receiving of control information must synchronize with same
time clock, in order to serve its usefulness. Therefore, we find it
urgent to investigate about any conflicting design issues, that
would not to suit channel applicability situation in WSN
transmissions.
In Energy Efficient Reliable Transport Protocol (EERT)
Hadamard coding techniques is explored whereby a source node
encodes a data packet. The packet encoded is finally delivered to
the base station (sink) using Ad-hoc on Demand Distance Vector
Routing Protocol (AODV) [9]. The technique of data encoding is
that, Hadamard coding scheme maps a message length K-bits
into 2K bits codeword, which is transmitted to the receiver node.
At the receiver Cyclic Redundancy Check (CRC) is performed
on the data with K-bits length in the transmitter and 2K-bit
codeword in the receiver. Subsequently, it is detected that only
few errors occurred [9], whiles in most occasion errors are rarely
discovered in the transmission. Meanwhile, data encoding is
done continuously, over and over. We consider using frame
iterations time in clustered network, synchronized, should
subsequently sent data to neighbors. We anticipate continuous
delivery of such encoded data, without clustered architecture
result in loss or inaccurate data transmission. We believe also
this can give much data overhead which need to be investigated.
3 SYSTEM DESIGN
The system design comprises of description of various WSN
energy efficiency, and data transfer clustering techniques,
outlined below:
3.1 Sensor FIFO Buffer Queue Technique
FIFO [10] stands for first in, first out, and FIFO
asynchronous data transfer (ADT) buffer queue, is a hardware
component design that is capable of storing data, and also treat
the data as part of exchange between a numbers of processors.
FIFO ADT buffer queue can also be designed in queue block
such that simulation in the buffers in software is possible. Based
upon the requirement of the design, processors are required to be
driven by a clock. The job of the clocks is required to provide
synchronization between all processor hardware in the design,
which intends to provide synchronization by all clocks and the
sensor nodes, that are found in the design.
In Fig.2 is the model that represents two processors A and
B. The processors are designed with their corresponding clock
attached to the sensor FIFO ADT buffer. In the processors are
found entities which have connections with the FIFO Queue
blocks. Entity connections and entities are formed such that,
implicitly, they move and unite along with each other, and also
enforce blocking between blocks which occur in the design.
Subsequently, FIFO queue blocks are developed which buffer
and regulate the movement of data between Processors A and
Processor B. For the purpose of our model design, processors
represent sensor nodes.
IJSER
International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 625
ISSN 2229-5518
IJSER © 2015
http://www.ijser.org
V. REFERENCES
[1] Single path or Multipath Stochastic Reliability inWireNetBiLinFenyu
[2] Jaiswal, S. and A. Yadav (2013). F basedadap ContemporaryC Con
[3] Anurag Aeron, “Fine TokenBucketSchemefoCongesiHighSpeeNetwo
[4] A., F. Akyildzi (2010). Wireless Sensor Network
.
V. Hierarchical Clustering Architecture
One of the effective and commonly used backbones for
organizing WSN is hierarchical clustering architecture [12]. As
shown in Fig.3 below, hierarchical architecture for WSN for
asynchronous data transfer is developed. This includes sensors
transmitting data to the cluster head and subsequently transmit
data to the base station (BS). Clustering architecture has been
modeled from LEACH, but with a different concept in using
head-set, rather that cluster head concept. By using clustering in
our concept, election of head-set, in a large sensor nodes
deployment in the network is developed, and divided into groups
forming clusters, through partitioning. The head-set members,
are required to transmit messages to distant BS. As the Wireless
Sensor Network (WSN) is limited with bandwidth resources due
to its limited memory and computation capability, and limited
energy withholding capability, it is recommended that spatial
reuse of bandwidth provision must be made continuously to
sensor node. In addition, efficient energy-aware routing
capability must be developed to aid in message transmission.
3.12 Reducing Congestion in Clustering Backbone
Constant clustering formation of the network consists of
deployment in large scale sensor node in a multihop
environment. As a result a lot of congestion is introduced in the
network, due to replacements of new nodes that have sufficient
energy that can replace nodes depleted out of energy. As such the
network must be subdivided, with each given a separate base
station. Due to congestion, there is contention for all data source,
competing with each other, which access the sensor nodes from
each cluster subdivision, reporting to the base station (sink). In
energy efficient desiring protocol such us our model, contention
of accessing the sensor nodes results in congestion, when each
data reporting to the base station results in high energy depletion,
and excessive energy holes. In order to introduce new batch
sensor node in constant cluster formation, sensors must be
capable of estimating twice as precision distance, in relation to
data transmissions to the base station . The ADT FIFO buffer
uses its synchronized clocks in each time iteration round, to
determine the average distance between all the sensors due to
implementation of block queuing in the deployment, in relation
to the base station. New node with high enough energy should be
successful to replace old sensor nodes, due to high connectivity
protection developed in the asynchronous data transmission in
the network.
Fig. 3 Energy Efficiency Protocol
3.2 Radio Model Design
A new radio model design, synonymous to the one described
in [13] is recommended in our model . However, the radio model
in [13] uses only shorter transmissions distance in the clusters,
which is not able to achieve synchronization of the sensor times.
But in reference to our new proposed model, and for our purpose
of attaining full synchronization in the sensor processors, we aim
to design radio transmission that is capable of extending it scope,
in order to achieve average precision sensor distance. The aim is
to explore a diverse transmission range, based upon clustered
network formation using energy consumption in transmit
amplifier, proportional to . This means, irrespective of the
nodes coverage distance, average distance estimation between
long or short transmission ranges must be determined by the
head-set formed, in relation to the transmissions of the base
station; which is also capable of exploring the location of its
position. Based upon this, we estimate the energy consumption to
be . Estimating the energy consumed for 1-bit message
transmit in this new radio model, which include estimating an
average between short and long distance is necessary, which
include obtaining average precision distance with the
corresponding energy as follows:
 (3)
Similarly, to transmit 1-bit message in shorter transmission
range, the energy consumed is:
 (4)
Furthermore, in order to receive 1-bit message, the energy
consumed should be:
 (5)
Eqn. 3, must incur a reduced cost assessment in our new
sensor hardware design, that is capable of reducing energy
consumption in the new receiver hardware . We show the
constants in our new radio model as in Table 1 as below :
1
Sink
Cluster Head
Sensor Node
Sink
Cluster Head
Sensor Node
IJSER
International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 626
ISSN 2229-5518
IJSER © 2015
http://www.ijser.org
Description
Cost
Amplifier energy
consumption for
Average distance
data transfer
0.0026pj/bit/
Amplifier energy
consumption for
short distance data
transfer
20pJbit/
Electronic circuit
energy consumption
in received signal
50nJ/bit
Energy consumed
for new sensor
2nJ/bit
Table 1: Parameter values in our radio model for quantitative
measurement analysis
3.3 Cluster Head Election Phase
In our algorithm description of cluster head election phase,
we determine optimal sensor cluster number , required in our
protocol for sensor nodes. We anticipate that nodes have
acquired average initial energy. The energy consumed is
averagely estimated to be the same for all cluster formation.
Before election phase begins, substantial amount of cluster head
is randomly identified to the base station. A typical initial action
is that, cluster head are required to disseminate message to all
neighborhood sensors. After that, cluster heads sends message to
sensors, and sensors depend on the received signal strength to
choose their cluster head. Subsequently, sensors make decision
and transmit to their corresponding cluster heads. Finally, cluster
heads are able to identify messages that originate from their
corresponding sensor nodes. These corresponding cluster head
are capable of choosing associate set, based on each cluster
head, dependent on the signal well analyzed .
In order to maintain uniformity in cluster distribution, we
anticipate that each cluster should have
nodes. Eqns. 3 and 4
can be used in determining energy consumption in each cluster
head as follows:


(6)
In Eqn. (6), the first part ( is used to show the
advertisement of message transmit consumed energy, which
typically represents average precision distance of sensor energy
consumption dissipation model. Whilst in the second part,
represent received message energy consumption.
Subsequently, we simplify Eqn. 6 as follows:


 (7)
The energy consumption of non-cluster head sensor nodes
election, can be estimated by Eqns. 3 and 4 as follows:
 (8)
The first part of Eqn. 8, indicates the received cluster head
consumed energy message. Based upon this, we anticipate
message is received by sensor node from all clusters. In the
second part of Eqn. 8, the consumed energy for transmitting
message decision to the corresponding cluster head is shown.
Simplifying, Eqn. 8 is as follows:
 (9)
4 Asynchronous Data Transfer Model in WSN
We use asynchronous data transfer in our proposed model as
outlined below. Asynchronous data transfer(ADT) in senor
FIFO buffer queue phase in WSN is vital for storing data
efficiently, and exchange data between corresponding numbers
of sensors in a multihop clustered network environment. ADT is
modeled according to data transfer phase model technique in
[13]. With the data transfer phase model[13], hierarchical
clustering technique is used in large scale sensor deployment,
that enhances energy efficiency state of the clustered WSN. The
technique further describes an algorithm for data transfers, which
require that nodes must be capable of transmitting messages to
their cluster head. Subsequently, aggregated message is
transferred to a distant base station. With asynchronous data
transfer, aggregate message transmission, which includes base
station, and data gathering techniques, is not applicable in the
data transfer phase. Therefore, we develop algorithm to fulfill
that requirement, which fulfill deployment requirement in
aggregated data transmission. This should include the energy
consumed by each cluster head, in a multhop clustered network
environment based on the new designed hardware transceiver.
Note, however that based upon description, already outlined
in system design section III, ADT uses uniqueness attribute in
sensor FIFO buffer queue software simulation attribute, that must
be maintained. The software uniqueness emphasize on the
capability of our algorithm, using blocking queue development
within queuing block, to maintain constant connection.
Subsequently, these form connections of data transfer protection,
that should be capable to prevent occurrence of congestion in
continuous clustered formation. Therefore, we model energy
consumed based on each cluster head-set, as according to data
transfer model in [13]. Reemphasizing, our simulation in
software, which enables us use software simulation analysis
development in our model is applicable in this new radio
hardware data transmission test, and we compare our result to the
data transfer phase mode[13]. However, our model uses energy
efficient routing, such as ad-hoc on-demand energy-aware
routing (AOER), for energy efficiency state, evaluated as
follows:
IJSER
International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 627
ISSN 2229-5518
IJSER © 2015
http://www.ijser.org

 (10)
From the foregoing analysis, energy development equations,
which relate to average precision distance estimation of sensor
nodes, in relation to base station, with cluster head formation, is
given in Eqn. 10, which shows the energy consumed for
transmitting message to distant base station. Based upon this
energy requirement of our model can be satisfied. In the second
part of the Eqn. 10, is energy consumed in order to receive
message from the remaining sensor node is (
. Note that
this does not form part of the head-set.
Simplifying Eqn. 10 is as follows:



(11)
In order to transmit sensor data to the cluster head, energy
consumed by non-cluster head is determined as follows :
 (12)
Determination of energy consumed is based on uniform
distribution of circular clusters for sensor node, which uses
network diameter . Recalling, the average value distance
measurement for , least cost energy requirement which is
needed in our proposed model is given as:



i.e. 
 (13)
Simplifying Eqn. 12 is given as below:

 (14)
5 Circular Data Iteration
In circular data events for data distribution, data frame
transmission occurs in one iteration. Therefore, the transmitted
frame by each cluster should be . In all we estimate that
uniform partition of  frame occur amongst  cluster
nodes. In order for each cluster head frame transmission to
occur,
non-cluster head frame fraction determination is
required.
We determine equations for frames transmissions fractions
 as given below:
Now,

(15)



(16)
Therefore, we estimate that asynchronous data transfer (ADT)
frame of each cluster energy consumption will be determined as:
 (17)
 (18)
6 Start Energy for Data Transfer Rounds
Start energy for data transfer round is denoted by .
This is the initial energy of the sensor node, with initial start
time. In order to successfully transfer data in rounds, 
should be sufficient for at least one round in transmitting data.
During data transfer, it is required for a node to become head-set
member for one time, in a data transfer round, and a non-cluster
head for
times.
 is estimated finally to be:


(19)
Hence, 

7 Optimum Cluster Number Determination
Optimum cluster number , used in minimum energy
consumption, can be determined based on[13] as:

(20)
The optimum value of is determined based on Eqns. used in
[13] for minimum frame energy dissipation, evaluated and
summarized as :

 (21)
8 Description of Time Completion during One Round Data
Transfer
In data transfer phase, message transmission has been
specified in [13], which is based only on TDMA schedule.
Asynchronous Data Transfer(ADT) phase message transmission
is based on combination of message transfer in the network layer
(NWL), transport layer (TPL), and MAC layer, shown in Fig.1.
Furthermore, ADT FIFO buffer queue is specified with design
requirement, that uses connections in sensor ad hoc behavior to
maintain constant formation in queue. These queue are formed
such that, implicitly, they move and unite along with each other,
and also enforce blocking between blocks, which prevent
occurrence in any irregular contention of access in the sensor
buffer queue development medium. The design requirement also
includes TDMA technique for data transfer in one round. In
addition, this design need fulfill the requirement that address the
need for finding a suitable protocol resolution for Wireless
sensor network protocol layer transmissions.
Based upon this, frame time , which include different
message transmission times for all sensor cluster nodes, should
be determined. Consequently, sensor nodes should be giving
equal time processing capability to synchronize their clock times
IJSER
International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 628
ISSN 2229-5518
IJSER © 2015
http://www.ijser.org
due to routing in NWL, which uses ad-hoc on-demand energy-
aware routing (AOER) protocol. During this stage, data transfer
rate  can be estimated and average precision
message length of , is developed in time to transfer
message  time as follows:

(22)
Note, here that message transmission in one frame, include non-
cluster heads nodes and head-set active member. We estimate
also that in one round time, appreciable number of sensor head-
set members should be active. Therefore, as this requirement is
fulfilled in our design, no additional requirement will be needed,
in order for one to give account in overhead data transmission,
that might occur as inactive head-set member in our design.
Based upon this we estimate one frame time as below:


 
(23)
Based upon the description in Eqn. 23 above, the first part uses
 messages from non-cluster head nodes. The second part,
indicates sensor head-set active member message transmissions.
The justification of our model form [13] is based on estimating
same frame time message transfer for all the sensors, based on
uniform clock synchronization transmission of sensor device in
our model. Therefore, Eqn. 23 can further be simplified as
follows:

 (24)
Until now, we believe there should not be a concern, that should
make one to misunderstand that, frame have been used for
data transmission in one iteration. Therefore, we determine one
iteration time  required as below:
 (25)
With Eqns. 23 and 24 combined, we summarize the time for
iteration  as:

(26)
Note here also that with
 iterations, occurring in one round in
each data transmission, we estimate one round time  as
below: 


(27)
9 Results Discussion
We analyze our result discussion in our proposed
Hierarchical Clustering based FIFO asynchronous data transfer
technique, for Energy Efficient WSN (HAEEW). We also
compare the result to the original model, Energy Efficient
Hierarchical Cluster-based Routing (EEHCR) protocol- (which is
modeled from LEACH) protocol, using quantitative measure,
based on the radio communication model ,and software
simulations parameter settings outlined in section V.
9.1 Optimum Cluster Setting Analysis
Based upon Eqn. 21, we analyze result in the optimum
cluster number setting in the Asynchronous Data Transfer (ADT)
technique (HAEEW), and compare the result to the original data
transfer model technique (EEHCR). Fig.4 depicts the two graphs
The graph for Energy Efficient Hierarchical Cluster-based
Routing Protocol (EEHCR) and the graph for Hierarchical
Clustering-based FIFO asynchronous data transfer technique for
Energy efficient WSN (HAEEW)- our model. These two graphs
are both evaluated based upon the optimum cluster number
variation, and head-set size for sensor nodes in software
simulation implementation in matlab, with parameters settings:
sensor nodes number n=1200, with base station location at
d=150m.
For improved result of our proposed model, even though, we
varied the number of sensor node from n=1000 to n= 1200. The
head-set was required not to exceed between 1 and 6
theoretically. We maintained that status (seeFig.4). Our
algorithm HAEEW shows remarkable improvement in the head-
set, and does not exceed the estimated theoretical setting range of
1 and 6, which does not also incur message overhead.
Fig.4
: Maximum optimum cluster number in Head - set
Fig. 5 Maximum optimum cluster number
0 2 4 6 8 10 12
0
5
10
15
Head-Set Size
Number of Clusters
HAEEW
EEHCR
0 2 4 6 8 10 12 14 16 18 20
0
1
2
3
4
5
6
Number Of Clusters
Energy (J)
IJSER
International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 629
ISSN 2229-5518
IJSER © 2015
http://www.ijser.org
Fig.5 shows energy consumption, estimated in relation to the
cluster number. Based upon this graph, the energy consumption
is expected to reduce, with increased cluster number, using
Hierarchical Clustering-based FIFO asynchronous data transfer
technique for Energy Efficient WSN (HAEEW). However, based
on our improved techniques in Asynchronous Data Transfer, the
number of clusters used have been doubled, which also
introduced average cluster number estimation. This include
increased in cluster number as shown Fig.6, compared to that
data transfer technique, in Energy Efficient Hierarchical Cluster-
based Routing Protocol (EEHCR) shown in Fig.5. Therefore, it is
expected that with more increased in the cluster number, due to
perfect achievement in more clock synchronization, the
difference in energy consumption reduction is huge with our
technique, when more clusters are introduced in the network.
Fig. 6: Improved maximum optimum cluster number
9.2 Frame Iteration Synchronization Time
Frame iteration time is used for completing one iteration.
This is averagely estimated based on the precision distance
determination on sensor nodes transmissions, in relation to the
base station; with respect to cluster head formation, which forms
part of the head-set. Average time determination in one iteration
is also estimated such that, each clock time for all sensor nodes
becomes synchronized in relation to the network diameter
estimation.
Fig.7:Iteration time for Head-set size on average diameter
estimation
Fig.7 illustrate frame iteration time variation graph, based on
synchronized clocks of the sensor, which has significant effect in
cluster diameter, and head-set size estimation. In our model
(HAEEW) improved diameter of 1.5m is estimated, which gives
reduced cluster diameter, and correspondingly less one round
iteration time, as compared to the diameter estimated in
(EEHCR) which gives 3.0m. The three axes, are represented
respectively by sensor cluster diameter, the head-set size, and
synchronized clock time completion in one iteration. The sensor
cluster percentage is used for determining the head-set size.
Constant start energy  is applied in all cases, and can be
useful for long duration, with the head-set size estimated to be
more than 50% of the cluster size. Based upon the increased
percentage (>50%) in the head-size, sufficient data transmission
is allowed in each iteration; this reduces overhead in the number
of iteration in HAEEW; this would be impossible without
improved time synchronization technique in Asynchronous Data
Transfer, used for the percentage cluster size determination, and
subsequently for estimating the head-set.
10 CONCLUSIONS
Our proposed model Hierarchical Clustering-based FIFO
asynchronous data transfer technique for Energy efficient WSN
(HAEEW) has been modeled in reference to, Energy Efficient
Hierarchical Cluster-based Routing Protocol (EEHCR)-this is
modeled from
LEACH algorithm. In EEHCR, quantitative analysis was used to
indicate result for energy consumption, which decrease
systematically; by inclusion of many sensor nodes in cluster
head-set formation. In another development, with same number
of data collecting sensor nodes, it was analyzed that, controlled
number of nodes, and management number of nodes could
possibly be adjusted, based upon the network prevailing
condition. Our modeled technique, HAEEW, has even showed
remarkable improved result in energy efficiency, with
consistency in more energy consumption reduction. Thus when
we consider using greater number of sensor nodes, in the cluster
0 5 10 15 20 25 30
0
1
2
3
4
5
6
Number Of Clusters
Energy (J)
050 100 150 200
0
50
100
0
0.5
1
1.5
2
x 104
Network Diameter
Head-Set Size (%)
Time for one iteration (sec)
Head-Set
Size=1.5
Head-Set
Size=1.5
IJSER
International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 630
ISSN 2229-5518
IJSER © 2015
http://www.ijser.org
head-set formation, than that used in EEHCR (original model).
Consequently, referring to our model, Asynchronous Data
Transfer data collection of sensor nodes, shows much controlled
number of sensor nodes, and management number of sensor
nodes, which can be adjusted even more easier, irrespective of
the prevailing network condition, example of which is greater
number of sensor nodes that can be deployed for extended
network life.
Future work will concentrate on estimating the cluster head-
set size formation in mixed network environment of single-hop
and multihop network, with non-uniform cluster distribution also
taken into account.
11 ACKNOWLEDGEMENT
The authors wish to express their profound gratitude to the
Wireless Sensor Lab and Electrical Engineering Lab of
University of Bridgeport, for allowing them to use resources for
the research.
12 REFERENCES
[1] Xiong, Chuang Lin, Feng-yuan Ren, Wei Yan .Single path or Multipath
Stochastic Reliability in Wireless Sensor Networks Bin-bin
[2] Jaiswal, S. and A. Yadav (2013). Fuzzy based adaptive congestion control in
wireless sensor networks. Contemporary Computing (IC3), 2013 Sixth
International Conference on.
[3] Anurag Aeron, “ Fine Tuning of Fuzzy Token Bucket Scheme for Congestion
Control in High Speed Network s.” 2nd ICCEA, vol 1, pp. 170-174, 2010
[4] A., F. Akyildzi (2010). Wireless Sensor Network
[5] Energy Efficient Congestion Control Operation in WSNs Adel Gaafar
[6] Anurag Aeron, “Fine Tuning of Fuzzy Token Bucket Scheme
for Congestion Control in High Speed Networks.” 2nd ICCEA, vol 1, pp. 170-
174, 2010
[7] Zhang, M., W. Cai, et al. (2012). Hop-to-Hop Congestion Feedback
Mechanism for Sink Bottleneck Problem in WSNs. Intelligent Networks and
Intelligent Systems (ICINIS), 2012 Fifth International Conference on.
[8] P. Gilesh R. C. Hansdah (2010). An Adaptive Reliable Transport Protocol
Based on Automatic resend request (ASQ) Technique for Wireless Sensor
Networks M.
[9]http://www.eng.auburn.edu/cse/classes/comp8700/papers/OPNETWORKPAP
ERS/Simulation Modeling
[10] wwww.matwork.com/ sademo_function_mdl
[11] Cambridge Mobile ATM, p 2.
[12]A. Amiya, I Stojmenovic (2010).
Wireless Sensor and Actuator Networks.p36
[13] S. Hussain, and A. W. Matin. (2005)Energy
Efficient Hierarchical Cluster-Based Routing for Wireless
Sensor Networks Technical Report. p 11-27
Authors Biography:
Samuel Kofi Erskine is a PhD Candidate and graduate research
assistant at Department of Computer Science and Engineering
University of Bridgeport CT USA. He obtained his Masters of
Science in Telecommunications at George Mason University,
Virginia USA. Samuel has authored and published more than
three research papers in Computer Science & Engineering, and
also in business Technology management in international
journals of repute.
Dr. Elleithy is the Associate Vice
President for Graduate Studies and
Research at the University of Bridgeport.
He is a professor of Computer Science
and Engineering. He has research
interests in the areas of wireless sensor
networks, mobile communications, network security, quantum
computing, and formal approaches for design and verification.
He has published more than three hundreds research papers in
international journals and conferences in his areas of expertise.
Dr. Elleithy is the editor or co-editor for 12 books by
Springer. He is a member of technical program committees of
many international conferences as recognition of his research
qualifications. He served as a guest editor for several
International Journals. He was the chairperson for the
International Conference on Industrial Electronics, Technology
& Automation, IETA 2001, 19-21 December 2001, Cairo
Egypt. Also, he is the General Chair of the 2005, 2006, 2007,
2008, 2009, 2010, 2011, 2012, 2013, and 2014 International Joint
Conferences on Computer, Information, and Systems Sciences,
and Engineering virtual conferences.
Dr. Linfeng Zhang is Assocoate
Professor at university of Bridgeport.
Dr. Zhang received his PhD at Wayne
State University, Michigan USA. His
current research interest lies in Power
electronics, Susutainable energy,
Controls, Data acquisition, Fuel cells,
sensors, and virtual instrumentation.
IJSER
... Events can be served depending on FIFO (First Come First Served), Earliest Deadline First (EDF), Priority Queuing disciplines and many more. The FIFO queue is the simplest among other schemes, but introduces large waiting time for the late arrived events in the queue even if their priority of servicing is high [8,9]. Earliest deadline first (EDF) scheme requires priority assignment for each event and its tasks which to be processed and process for the early deadline [3]. ...
Conference Paper
Full-text available
In wireless sensor networks, congestion control is a challenging task because it causes the loss of information and reduction of energy efficiency. The congestion control scheme is necessary to be carried out to detect congestion precisely and regulate it fairly. To achieve this objective, Fuzzy Based Adaptive Congestion Control (FBACC) is proposed which uses buffer occupancy, participants and traffic rate. FBACC provides a fuzzy logic based congestion estimation, smart way to drop packet in case of congestion up to acceptable quality level and most important it regulate traffic rate. It improves performance of network over other protocols which are presently available. Performance of FBACC is evaluated and compared with existing schemes e.g. Event-to-Sink Reliable Transport (ESRT), Fuzzy Logic Based Congestion Estimation (FLCE) and Congestion Control Scheme Based on Fuzzy Logic (CCSFL) with respect to congestion detection, packet loss, end-to-end delay and energy. FBACC detects the network congestion more preciously and uniformly and it also adapts to current traffic rate with respect to product of previous participants with previous traffic rate to reduce the packet loss. Finally, FBACC saves energy of retransmission because packet loss is minimal due to traffic adaption.
Conference Paper
Wireless Sensor Networks (WSNs) is a most promising method for future environmental observation. The problem of congestion in WSNs is significantly different from conventional ad-hoc networks and has not been studied to any great extent thus far. Congestion in sink bottleneck of WSNs not only causes packet loss, but also leads to excessive energy consumption. Therefore, congestion feedback and control of WSNs will improve wireless channel quality dramatically, reduce energy consumption and avoid starvation, as a result increase network efficiency. In this paper, we proposed a distributed congestion feedback algorithm innovatively named as DCCA to solve sink bottleneck problem for multi-hop WSNs, which uses queue based congestion level detection scheme in the MAC layer and then adopts hop-by-hop feedback notification scheme in the transport layer. Finally, whether forward or drop packet will be decided by each sensor node in the network layer. Therefore, the proposed DCCA is a cross-layer design philosophical approach.
Conference Paper
In Wireless Sensor Networks, it has been proved that the reliable transmission protocols sending redundant packets to the upstream neighbor hop-by-hop have advantage on energy efficiency compared with those using end-to-end error recovery and control scheme. While transmitting the redundant packets over error prone wireless links, there are two different ways could be used, which are single path and multipath respectively. In this paper, we proposed a comprehensive study on the energy efficiency constrained with desired reliability for all these approaches. A finite state Markov Chain model and hidden Markov model combined theoretical analysis method is presented to evaluate and compare the performance of these delivery processes. Our analysis indicates that when the packet loss rate of the wireless channel is relatively low, the single path scheme reveal better energy efficiency than the multipath one. However, the performance of single path scheme deteriorates rapidly with decreasing of the channel quality. When the bit error rate of the channel passes through a certain threshold, the multipath scheme will come to be the most energy efficient one.
Article
In ATM networks, due to the dynamic nature of the bursty traffic, severe network congestion may occur. Therefore, policing as the preventive congestion control is required for properly guaranteeing QoS. The leaky bucket control mechanism, which is the most widely and only implemented one, has some drawbacks. In this paper the principles of fuzzy logic are deployed to tackle the problem of congestion control. This paper implements the fuzzy logic based token bucket scheme in which token generation rate is controlled by Fuzzy Logic. This work studies how to implement token rate predicting and policing in a high speed (ATM) network using fuzzy token bucket. The fuzzy token rate predictor facilitates token rate predicting and dynamic policing. It is able to achieve lower delay, less cell loss and higher throughput by Fuzzy token bucket scheme. A simulation study was performed for the fuzzy logic predictor using MATLAB-Simulink Simulator. The simulation results show that the fuzzy logic predictor gave reasonable token rate prediction compared to a non-fuzzy Token bucket scheme.
Chapter
Position information enables development of localized routing methods where greedy routing decisions are made at each node, based solely on knowledge of positions of neighbors and the destination, with considerable savings in the communication overhead. Power consumption can be taken into account in the routing process. This chapter will survey existing flooding-based and positionbased routing schemes. It also describes a general cost to progress ratio-based approach for designing routing protocols under a variety of metrics, such as hop count, power, remaining energy, delay, and others. The chapter also describes routing with guaranteed delivery for unit disk graphs and an ideal medium access control (MAC) layer. Gabriel graph, as a localized planar and connected structure needed for such solutions is described. Solutions are expanded toward beaconless behavior, where nodes are not aware of their neighborhood. Georouting with virtual coordinates is based on hop distances to some landmarks. This chapter also discusses the physical layer aspects of georouting, routing in sensor-actuator networks, and the load-balancing issue in routing.
Conference Paper
There are many wireless sensor network(WSN) applications which require reliable data transfer between the nodes. Several techniques including link level retransmission, error correction methods and hybrid Automatic Repeat re- Quest(ARQ) were introduced into the wireless sensor networks for ensuring reliability. In this paper, we use Automatic reSend reQuest(ASQ) technique with regular acknowledge- ment to design reliable end-to-end communication protocol, called Adaptive Reliable Transport(ARTP) protocol, for WSNs. Besides ensuring reliability, objective of ARTP protocol is to ensure message stream FIFO at the receiver side instead of the byte stream FIFO used in TCP/IP protocol suite. To realize this objective, a new protocol stack has been used in the ARTP protocol. The ARTP protocol saves energy without affecting the throughput by sending three different types of acknowledgements, viz. ACK, NACK and FNACK with seman- tics different from that existing in the literature currently and adapting to the network conditions. Additionally, the protocol controls flow based on the receiver's feedback and congestion by holding ACK messages. To the best of our knowledge, there has been little or no attempt to build a receiver controlled regularly acknowledged reliable communication protocol. We have carried out extensive simulation studies of our protocol using Castalia simulator, and the study shows that our protocol performs better than related protocols in wireless/wireline networks, in terms of throughput and energy efficiency. Keywords-Transport Protocol; Communication Architecture; Reliability;
Article
The existing problem of indoor wireless network is the supply of energy. This project aims to find solutions to the energy problem by looking into harvesting of ambient energy namely piezoelectric harvesting and solar energy harvesting. To achieve that, tasks of finding information on wireless network and energy harvesting were undertaken in order to conduct tests and experiments for analysis. This report begins by introducing the various technologies and history behind the operations of the wireless sensor network and the energy harvesting techniques of piezo and solar. Then the structure of both systems will be explained using diagrams. Calculations and observations are then compared and analyzed in terms of effectiveness and power to show how well the power harvesting unit supports the wireless system by transmitting sensor data. At the end of the project we conclude the feasibility of ambient power harvesting applications and what are the limitations at the current technological level and also discuss what the future outlook of such applications is. ELECTRICAL and ELECTRONIC ENGINEERING
Energy Efficient Hierarchical Dr Elleithy is the editor or co-editor for 12 books by Springer. He is a member of technical program committees of many international conferences as recognition of his research qualifications. He served as a guest editor for several International Journals
  • S Hussain
  • A W Matin
S. Hussain, and A. W. Matin. (2005)Energy Efficient Hierarchical Dr. Elleithy is the editor or co-editor for 12 books by Springer. He is a member of technical program committees of many international conferences as recognition of his research qualifications. He served as a guest editor for several International Journals. He was the chairperson for the International Conference on Industrial Electronics, Technology & Automation, IETA 2001, 19-21 December 2001, Cairo – Egypt.
Wireless Sensor Network [5] Energy Efficient Congestion Control Operation in WSNs Adel Gaafar
  • F Akyildzi
A., F. Akyildzi (2010). Wireless Sensor Network [5] Energy Efficient Congestion Control Operation in WSNs Adel Gaafar