Date of Award

9-2007

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Department of Electrical and Computer Engineering

First Advisor

Rusty O. Baldwin, PhD

Abstract

Due to the distributed nature of information collection in wireless sensor networks and the inherent limitations of the component devices, the ability to store, locate, and retrieve data and services with minimum energy expenditure is a critical network function. Additionally, effective search protocols must scale efficiently and consume a minimum of network energy and memory reserves. A novel search protocol, the Trajectory-based Selective Broadcast Query protocol, is proposed. An analytical model of the protocol is derived, and an optimization model is formulated. Based on the results of analysis and simulation, the protocol is shown to reduce the expected total network energy expenditure by 45.5 percent to 75 percent compared to current methods. This research also derives an enhanced analytical node model of random walk search protocols for networks with limited-lifetime resources and time-constrained queries. An optimization program is developed to minimize the expected total energy expenditure while simultaneously ensuring the proportion of failed queries does not exceed a specified threshold. Finally, the ability of the analytical node model to predict the performance of random walk search protocols in large-population networks is established through extensive simulation experiments. It is shown that the model provides a reliable estimate of optimum search algorithm parameters.

AFIT Designator

AFIT-DS-ENG-07-19

DTIC Accession Number

ADA472292

Download

Share

COinS