Abstract:
Wireless Sensor Network (WSN) consist of a unique set of resources like on-board
battery and wireless communication devices with limited bandwidth. WSN offers a
wide range of applications for monitoring space or targets. WSN is capable of
performing simple processing tasks like tracking, detection of an event, or
classification and may consist of multiple nodes that can process the information and
communicate with nearby nodes in real-time for environmental monitoring, event
detection, surveillance, object tracking, battlefield situation monitoring, and data
collection etc. However, there are certain limitations in deploying WSN efficiently
such as in terms of limited power resource for a single node in WSN, limited
processing capability and varying network life time. It had been shown that wireless
communication to and from sensor nodes consumes significantly more battery power
in comparison to power expended in sensing, computation and memory access
procedures. One of the possible solutions to this problem is to communicate as
sparingly as possible through efficient sleep/wake scheduling for WSN nodes to
extend node and network lifetime. A major research issue in WSN is to develop an
energy efficient MAC protocol that not only provides increase in network lifetime
but also addresses latency. In this research, a new MAC protocol is designed using
sleep/wake scheduling for WSN. Though energy consumption in WSN is
unavoidable due to communication necessity and for different stages like idle
listening, retransmission, channel sensing and overhearing; this proposed protocol
will help in decreasing this energy consumption. Energy efficiency and latency of the
proposed sleep/wake scheduling scheme is evaluated and compared with the state of
the art research. An AEL (Accounting for Energy and Latency) factor is introduced
which is the deciding element for active and sleep cycles of the node. This said AEL
factor defines the minimum duty cycle among the network nodes and is specified
prior to nodes deployment depending upon application requirements. In the proposed
protocol, the nodes adjust their duty cycles according to this AEL factor depending
upon traffic load, their position and their connectivity in the network. The effect of
this AEL factor on energy efficiency and delay is evaluated for different network
densities in this research thesis. While incorporating sleep/wake scheduling for
energy efficiency, delays are added in the network to route packet from node towards
sink. Therefore, it is necessary to address latency for MAC protocol especially for
delay constrained applications. This research thesis focuses on sleep/wake
scheduling scheme ensuring energy efficiency, decreased latency and increased
network lifetime by selecting an appropriate AEL value. The research includes the
comparison of the proposed protocol with state of the art research and has shown
significant percentage improvements in energy efficiency and delay from S-MAC
and Anycast protocols.
