Degree Type

Thesis

Date of Award

2016

Degree Name

Master of Science

Department

Computer Science

Major

Computer Science

First Advisor

Wensheng Zhang

Abstract

The wireless sensor network (WSN), once deployed, is usually expected to operate for months even years. In the WSN, each individual sensor node is typically equipped with small batteries which offer only a limited amount of energy. Hence, numerous energy-efficiency schemes have been developed for the WSN, aiming to help the WSN to achieve long lifetime under the stringent constraint of energy supply. These energy-efficiency schemes can reduce energy consumption, but they cannot address the issue of unbalanced energy consumption and energy replenishment in the WSN. In practice, different sensor nodes could have different initial energy supplies and/or play different roles in the network; consequently, some nodes may deplete their energy faster than the others. In such cases, the lifetime of the network becomes determined by the energy-bottleneck nodes who deplete their energy the fastest. To address this issue, the ideas of energy-balancing and lifetime-balancing have been proposed and energy-balancing and lifetime-balancing schemes have been developed for the MAC and routing layers of the WSN. However, there have been very few efforts on studying how to apply the idea to multiple layers of the WSN simultaneously and the effectiveness of such an integrated solution. To fill this gap, this thesis presents an integrated design, which includes lifetime-balancing schemes for the application, routing and MAC layers. Specifically, in the application layer where the sensing duty should be assigned to sensor nodes, we present a scheme in which leader nodes coordinate the sensing duty schedule among the sensor nodes based on their estimated nodal lifetime. In the routing layer, the routing metric is calculated based on both nodal lifetime and required end-to-end delay. In the MAC layer, we adopts the LB-MAC protocol, which balances nodal lifetime by adjusting the MAC parameters. Our design has been simulated in Network Simulator 2 (NS2) and compared with other solutions which do not apply lifetime-balancing idea or only applying the idea in some but not all of the three layers. The evaluation results have shown that our integrated design prolongs the network lifetime more effectively. Also, the integrated design achieves better performance in terms of data delivery ratio, end-to-end delay and wasted energy.

Copyright Owner

Rui Bai

Language

en

File Format

application/pdf

File Size

61 pages

Share

COinS