Adaptive load-balanced protocol for clustered wireless sensor networks

Other Title(s)

بروتوكول متكيف لتوازن الحمل في شبكات الاستشعار اللاسلكية العنقودية

Dissertant

Salih, Tariq Ahmad

Thesis advisor

al-Samawi, Venus Wazir
al-Mubayyidin, Wisam

Comitee Members

al-Smadi, Adnan M.

University

Al albayt University

Faculty

Prince Hussein Bin Abdullah Faculty for Information Technology

Department

Department of Computer Science

University Country

Jordan

Degree

Master

Degree Date

2010

English Abstract

Recently, a lot of attention is paid for wireless sensor networks due to their several applications such as environment monitoring for both military and civil applications.

Gathering Sensor nodes into groups is called clusters.

Clustering allows data aggregation and reduces congestion and energy consumption.

Due to the limited data transmission range and battery power of the sensor nodes, it is important to develop protocols that minimize energy consumption to increase network lifetime.

Since the sensor nodes are deployed for real time, emergency, and rescue environments, there is a need to minimize data delay and maximize throughput. The traffic load balancing among cluster-heads (CHs) is not considered in the most of the published clustering protocols.

Increasing the network life time by reducing the energy consumption is the most consideration of these protocols than quality of service like throughput, and End-To-End delay.

This is due to the battery capabilities that provides sensor with power.

Some of the proposed protocols use re-clustering strategies that increase overhead messages.

As a result the network life time will decrease, so a network without traffic load balancing technique leads to exhaust the energy of cluster-heads, this means decreasing the network life time.

Cluster-heads that have high traffic load will consume their energy very soon, thus an approach is needed to assign a set of nodes to cluster-heads to distribute the sensor nodes traffic load among these cluster-heads. In this research, we study the impact of heterogeneity of nodes in terms of their traffic generation in wireless sensor networks that are hierarchically clustered.

We propose an adaptive load balanced protocol that called TEACH (Traffic Efficiency using Adaptive Clustering Hierarchy), to increase the network life time by fairly distributing sensor nodes among cluster-heads in wireless sensor networks.

For a fair distribution of the sensor nodes, the base station assigns the nodes to the proper cluster head upon their traffic load.

In other words, in cluster head election process, a group of sensors that have low traffic will be assigned to cluster head to which large number of those low traffic sensors will belong.

A certain cluster head will be assigned a smaller number of sensors as the traffic load of this set of sensors increases.

Clustering sensor nodes around cluster-heads regarding their traffic load allow the sensor nodes to save power and balancing the load among cluster-heads.

Also if cluster head dies or its energy lower than predefined threshold, the cluster head role will be rotated.

By assigning the sensor nodes among cluster-heads, the traffic load between cluster-heads can be balanced, which leads to minimize energy consumption of the cluster-heads, End-To-End delay, and maximizing network throughput. Clustering the sensor nodes in a network around several cluster-heads depending on their traffic is the main objective of this work.

The suggested protocol acts to achieve traffic load balancing, power saving, maximizing throughput, and reduce delay, whereby the traffic load contributed by all sensor nodes are different. The experimental results of TEACH are compared with LEACH-C protocol (Low Energy Adaptive Clustering Hierarchy-Centralized) (Heinzelman et al, 2002).

LEACH-C protocol depends on centralized approach that choosing random cluster-heads to distribute sensor nodes around them. Simulation results have shown that TEACH protocol always give better results than LEACH-C protocol in term of: • Average network throughput: - 66% when the packet size is 500 byte. - 71.9% when the packet size is 1000 byte. - 103% when the traffic is duplicated to triple amount. • Average network End-To-End delay: - 64.7% when the packet size is 500 byte. - 64.4% when the packet size is 1000 byte. - 94.2% when the traffic is duplicated to triple amount. • Prolonging network life time when the number of nodes increases and it is comparable to LEACH-C: - 4.8% when the packet size is 500 byte. - 2.7% when the packet size is 1000 byte. - (-34.2%) when the traffic is duplicated to triple amount. From these results we conclude that distributing the traffic load among cluster-heads led to balance the load of these cluster-heads, which increase throughput and minimize End-To-End delay.

In addition to that, rotating the role of cluster-heads and use data aggregation before send data to the base station reduce the energy consumption, and these lead to prolong network life time.

Main Subjects

Information Technology and Computer Science

No. of Pages

95

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-311510