[thumbnail of Thesis_v27.pdf]

Download (1MB)


A Wireless Sensor Networks (WSN) is composed of a large number of low-powered
sensor nodes that are randomly deployed to collect environmental data. In a WSN,
because of energy scarceness, energy efficient gathering of sensed information is one
of the most critical issues. Thus, most of the WSN routing protocols found in the
literature have considered energy awareness as a key design issue. Factors like
throughput, latency and delay are not considered as critical issues in these protocols.
However, emerging WSN applications that involve multimedia and imagining sensors
require end-to-end delay within acceptable limits. Hence, in addition to energy
efficiency, the parameters (delay, packet loss ratio, throughput and coverage) have
now become issues of primary concern. Such performance metrics are usually
referred to as the Quality of Service (QoS) in communication systems. Therefore, to
have efficient use of a sensor node’s energy, and the ability to transmit the imaging
and multimedia data in a timely manner, requires both a QoS based and energy
efficient routing protocol. In this research work, a QoS based energy efficient routing
protocol for WSN is proposed. To achieve QoS based energy efficient routing, three
protocols are proposed, namely the QoS based Energy Efficient Clustering (QoSEC)
for a WSN, the QoS based Energy Efficient Sleep/Wake Scheduling (QoSES) for a
WSN, and the QoS based Energy Efficient Mobile Sink (QoSEM) based Routing for a
Clustered WSN.
Firstly, in the QoSEC, to achieve energy efficiency and to prolong
network/coverage lifetime, some nodes with additional energy resources, termed as
super-nodes, in addition to normal capability nodes, are deployed. Multi-hierarchy
clustering is done by having super-nodes (acting as a local sink) at the top tier, cluster
head (normal node) at the middle tier, and cluster member (normal node) at the lowest
tier in the hierarchy. Clustering within normal sensor nodes is done by optimizing the
network/coverage lifetime through a cluster-head-selection algorithm and a
sleep/wake scheduling algorithm. QoSEC resolves the hot spot problem and prolongs
network/coverage lifetime.
Secondly, the QoSES addressed the delay-minimization problem in sleep/wake
scheduling for event-driven sensor networks for delay-sensitive applications. For this
purpose, QoSES assigns different sleep/wake intervals (longer wake interval) to
potential overloaded nodes, according to their varied traffic load requirement defined
a) by node position in the network, b) by node topological importance, and c) by
handling burst traffic in the proximity of the event occurrence node. Using these
heuristics, QoSES minimizes the congestion at nodes having heavy traffic loads and
ultimately reduces end-to-end delay while maximizing the throughput.
Lastly, the QoSEM addresses hot spot problem, delay minimization, and QoS
assurance. To address hot-spot problem, mobile sink is used, that move in the network
to gather data by virtue of which nodes near to the mobile sink changes with each
movement, consequently hot spot problem is minimized. To achieve delay
minimization, static sink is used in addition to the mobile sink. Delay sensitive data is
forwarded to the static sink, while the delay tolerant data is sent through the mobile
sink. For QoS assurance, incoming traffic is divided into different traffic classes and
each traffic class is assigned different priority based on their QoS requirement
(bandwidth, delay) determine by its message type and content. Furthermore, to
minimize delay in mobile sink data gathering, the mobile sink is moved throughout
the network based on the priority messages at the nodes. Using these heuristics,
QoSEM incur less end-to-end delay, is energy efficient, as well as being able to
ensure QoS.
Simulations are carried out to evaluate the performance of the proposed protocols
of QoSEC, QoSES and QoSEM, by comparing their performance with the established
contemporary protocols. Simulation results have demonstrated that when compared
with contemporary protocols, each of the proposed protocol significantly prolong the
network and coverage lifetime, as well as improve the other QoS routing parameters,
such as delay, packet loss ratio, and throughput.

Item Type: Thesis (PhD.)
Depositing User: Users 5 not found.
Date Deposited: 05 Jun 2012 10:23
Last Modified: 25 Jan 2017 09:42
URI: http://utpedia.utp.edu.my/id/eprint/3029

Actions (login required)

View Item
View Item