A SURVEY OF ENERGY AWARE ALGORITHMS

FOR ALLOCATION OF SENSOR NODES IN WIRELESS

SENSOR NETWORK
Prashant Kumar, Prakash Bhatt, Saba Khan
Department of Information Technology
Maharaja Surajmal Institute of Technology, New Delhi,
India
ABSTRACT: Wireless Sensor Networks
(WSNs) is gaining popularity in multiple
applications over single application.
Every application has different view for
different the algorithm. This research
investigates a number of dynamic
algorithms, showing different scenarios
in which every algorithm has its
advantage and its lows according to the
application in use.

Introduction

The Wireless Sensor Network (WSN) was

a research project by UC Berkeley [1],
where
they
used
micro
electromechanical technology to make a
wireless sensor. It was initially made to
monitor
physical
or environmental
conditions like pressure, temperature
and motion. The first use of WSN came
in existence in battle field. Due to small
size, low price WSNs were used in daily
applications like fire detection, CO2
emission, bridge vibration etc. Usually a
Sensor node for example Mica2 Mote
consist
of
radio
transceiver,
an
embedded
processor,
internal
and
external memories , a power source such
as battery and one or more sensors [3].In
WSN, strength of signals are inversely
proportional to the distance [1]. Larger the
distance, higher is energy consumption,
thus affecting the lifetime of network. To
fully utilize the power of sensor nodes for
continuous data transmission is itself an
important research topic. Basically, there

are 2 ways of transmission, direct &

indirect transmission [1]:
1.1

Direct Transmission

Each node sends data directly to the

base station. Here, the data rate is
higher since there is no data forwarding
during transmission. Implementation of
direct transmission is easy as the
application has constraint for small area,
but as we increase the application to
larger area, the sensor nodes will die
quickly due to energy consumption.
1.2

Indirect Transmission

Sensor nodes will send data using

forwarding so as to send the data to
base station, but forwarding data causes
high energy consumption in long
distance transmission. The sensor nodes
start to collect and transmit data to the
base station using multi hop forwarding.
2

Related Work

In WSN, hardware is barely able to run

multiple tasks at a time, so to overcome
this problem we implement virtual
sensors as an image of physical sensors.
It contains the Meta data about the
physical sensors and user is shown that
virtual sensor. And also virtual sensors
have processing code with this it can
respond to the complex codes. So, there
are
four
different
configurations
regarding this [6]:
2.1One-To-Many Configuration

When several request is made by users,

so every user is allocated their virtual
sensor image, and the underlying
physical sensor is shared among them

physical sensors are connected to a

single virtual sensor image.

2.2Many-To-One Configuration
In this configuration, when every user
requires aggregated data then many
2.3MANY-TO-MANY CONFIGURATION
A physical sensor can correspond to
many virtual sensors, and it can also be
a part of a network that provides
aggregate data for a single virtual
sensor.
2.4DERIVED CONFIGURATION

Figure

It is a combination
1 of the multiply
configurations.
In
the
derived
configuration,
the
virtual
sensor
communicates with multiple sensor
types; in the other three configurations,
the virtual sensor communicates with
the same type of physical sensors.
3

Hardware Platform

Each
node
does
some
sensing,
processing
and
transmitting
task.
Commonly used sensor nodes are Mica2
mote,
developed
by
Crossbow
Technology [3]. It consists of a radio
transmitter, an embedded processor,
internal and external memories and a
power source.
3.1Embedded Processor
A processor is used to schedule tasks,
process data and control functionality of
other components. Different type of
embedded processors that are used
includes Microcontroller, Digital Signal
Processor (DSP) and Application-Specific
Integrated Circuits (ASIC). Above all
these processors Microcontroller is
mostly used because of its flexibility to
connect to other devices and its low
price. For example 8051 microcontroller
is used by CC2531 development board,

and Mica2 Mote

microcontroller.

uses

ATMega128L

3.2Transceiver
A Transceiver is used to communicate
with other sensor nodes. Sensor nodes is
operated on different states, they are
Transmit, Receive, Idle and sleep. Mica2
Mote uses two kinds of Radio frequency
radios: RFM TR1000 and Chipcon
CC1000, with a range of about 150
meters.
3.3Memory
Memory includes in-chip flash memory,
RAM and external flash memory. For
example, ATMega128L microcontroller
has 128 K-byte flash memory and 4-K
static RAM.
3.4Power Source
Energy
is
used
by
sensing,
communication and data processing.
Data communication needs more energy
than sensing and data processing.
Energy is stored in batteries. Some of
the batteries used are 2 AA batteries.
3.5 Sensors
Sensor is a hardware device used to
detect and measure response signal to a
change in physical environment such as
temperature, pressure and humidity. The
continues change in analog signal is
converted to digital signal by Analog to
digital signal and then sent to processor.
A sensor node can have one or more
sensors connected.

Figure
2

This algorithm takes care of both the

components (CPU, RADIO) of the sensor
nodes. It uses both the conditions
followed by DDLBA and DCLBA. Sum of
the energy consumed by CPU and radio
component is maintained and the sensor
having the minimum sum is allocated.

4 ALLOCATION ALGORITHMS
The main objective of a resource
allocation in WSN is to improve lifetime
of wireless network [2] and of a routing
algorithm is to find the better way of
transmitting the data to save energy [1].
4.1 Dynamic Random Allocation
(DRA)
The sensor nodes are allocated
randomly to various applications, where
the selection of sensor nodes has equal
probability.
4.2 Dynamic CPU Load Balanced
Allocation (DCLBA)
The DCLBA algorithm attempts to
balance the energy consumption due to
the CPU component amongst the various
sensor
nodes.
For each sensor node, the information
about the total of all execution time of
job requests that have been already
allocated to the
sensor node is
maintained. The sensor node having the
minimum sum is allocated for the
application request.
4.3 Dynamic Data Load Balanced
Allocation (DDLBA)
DDLBA algorithm balance the energy
consumed
while
transmitting
the
response messages corresponding to the
requests processed by a sensor node.
For each sensor node, information about
the sum of the size of the response
messages corresponding to the job
requests submitted to the sensor node is
maintained. The sensor node having the
minimum sum is allocated for the
application request.
4.4 Balanced Metric Allocation (BMA)

4.5 Maximum Energy First (MEF)

Energy consumption is based upon Hop
distance,
processing
power
and
transmission
energy.This
algorithm
assumes complete knowledge about the
energy levels of all sensor nodes. The
allocation of sensor is based upon their
decreasing order of energy, i.e. the
sensor node having maximum remaining
energy is allocated.

5 Comparisons of BMA, MEF,

DRA, DCLBA, DDLBA
algorithms
5.1Effect of Arrival Rate
With the increase in the arrival rate the
number of messages transmitted in the
network increases and hence the
minimum energy of a sensor node at the
end of simulation decreases. The node
that has the minimum energy is the one
for which energy consumption is
maximum. Among all algorithms the
BMA and MEF algorithms performed
the best.
5.2Effect of Execution Time
Allocation of sensor nodes is done at
random by RDA.DRLBA and DDLBA
algorithm balances only the partial
component of the energy consumption.
Since BMA and MEF algorithms do not
allocate sensor nodes having high
energy consumption or minimum energy
remaining, hence except for BMA and
MEF, for all other algorithms the
minimum energy decreases with the
increase in the execution time. Hence
among all algorithms the BMA and
MEF algorithms performed the best.
5.3Effect of Data Size

Since the energy used in transmitting a

message increased with the size of the
message and BMA and MEF algorithms
do not allocate sensor nodes having high
energy consumption or minimum energy
remaining. Hence the BMA and MEF
algorithms give the best result and
their performance increases with the
increase of the data size and compared
to others.
5.4Effect of Number of Sensors
In this the number of sensors allocated
to an application is increased. Hence
with the increase in the number of
sensors the number of messages
transmitted
in
the
network
also
increases due to which the minimum
energy decreases and since BMA and
MEF algorithms do not allocate sensor
nodes having high energy consumption
or minimum energy remaining. Hence
the BMA and MEF algorithms give
the best result.
5.5Effect of Unequal Initial Energy
of Sensor Nodes
In this case it is assumed that the initial
energy of the sensor nodes is not
uniform. Hence in MEF the sensor nodes
having higher energy are allocated
higher number of times than the sensor
nodes having less energy. So in this case
BMA gives better performance than
MEF.
6

CONCLUSION
If the initial energy of all the
sensor nodes is same then, BMA
and MEF algorithms will give the
best performance.
If the initial energy of the sensor
nodes is not same then, BMA

algorithm
performance.

gives

best

The
difference
in
the
performances of the algorithms
increases with the increase in the
arrival rate of incoming requests,
execution time of applications,
and data size of messages.

REFERENCES
1. Bo-Si Lee,Hao-Wei Lin,Wernhuar
Tarng, A Cluster allocation and
routing algorithm based on node
density for extending the lifetime
of wireless sensor networks, In
the
proceeding
of
26th
international
conference
on
Advanced information Networking
and Application Workshops 2012.
2. Navdeep Kaur Kapoor,Biswajit
Nandy,Shikharesh
Majumdar,
Dynamic allocationg of sensor
nodes in wireless sensor networks
hosting multiple applications.
3. Quinghua
Wang,
Ilangko
Balasingham, Wireless sensor
networks An introduction ,
Wireless Networks: ApplicationCentric
Design,Yen
Kheng
Tan(ed.),ISBN-978-953-307-321-7.
4. Yichao
jin,
Serdar
Vural,
Alexander Gluhak, Klas Moessner,
Dynamic task allocation in mulithop multimedia wireless sensor
networks with low mobility, Open
Access Article 16-10- 2013.
5. Malik Tubaishat ,Sanjay Madria ,
Sensor networks:An overview,
IEEE potentials 2013.
6. Sanjay Madria, Vimal Kumar, and
Rashmi Dalvi, Sensor Cloud: A
Cloud of Virtual Sensors, IEEE
Computer Society