Carter May

Abstract

Over the past few years, advances in electrical engineering have allowed electronic devices to shrink in both size and cost. It has become possible to incorporate environmental sensors into a single device with a microprocessor and memory to interpret the data and wireless transceivers to communicate the data. These "sensor nodes" have become small and cheap enough that they can be distributed in very large numbers into the area to be monitored and can be considered disposable. Once deployed, these sensor nodes should be able to self-organize themselves into a usable network. These "wireless sensor networks," or WSNs, differ from other ad hoc networks mainly in the way that they are used. For example, in ad hoc networks of personal computers, messages are addressed from one PC to another. If a message cannot be routed, the network has failed. In WSNs, data about the environment is requested by the "data sink." If any or multiple sensor nodes can return an informative response to this request, the network has succeeded. A network that is viewed in terms of the data it can deliver as opposed to the individual devices that make it up has been termed a "data-centric" network [26]. The individual sensor nodes may fail to respond to a query, or even die, as long as the final result is valid. The network is only considered useless when no usable data can be delivered.

In this thesis, we focus on two aspects. The first is data aggregation with accurate timing control. In order to maintain a certain degree of service quality and a reasonable system lifetime, energy needs to be optimized at every stage of system operation. Because wireless communication consumes a major amount of the limited battery power for these sensor nodes, we propose to limit the amount of data transmitted by combining redundant and complimentary data as much as possible in order to transmit smaller and fewer messages. By using mathematical models and computer simulations, we will show that our aggregation-focused protocol does, indeed, extend system lifetime. Our secondary focus is a study of cross-layer design. We argue that the extremely specialized use of WSNs should convince us not to adhere to the traditional OSI networking model. Through our experiments, we will show that significant energy savings are possible when a custom "cross-layer" communication model is used.

Library of Congress Subject Headings

Wireless communication systems--Mathematical models; Wireless communication sysetms--Computer simulation; Computer networks--Mathematical models; Computer networks--Computer simulation; Electric network topology

Publication Date

2005

Document Type

Thesis

Student Type

Graduate

Degree Name

Computer Engineering (MS)

Department, Program, or Center

Computer Engineering (KGCOE)

Advisor

Fei Hu

Advisor/Committee Member

Marcin Lukowiak

Advisor/Committee Member

Shanchieh Jay Yang

Comments

Physical copy available from RIT's Wallace Library at TK5103.2 .M39 2005

Campus

RIT – Main Campus

Download

Share

COinS