Publication Date


Document Type


First Advisor

Fonseca, Benedito

Degree Name

M.S. (Master of Science)

Legacy Department

Department of Electrical Engineering


The detection, localization, and tracking of environmental and physical conditions can be accomplished using wireless sensor networks (WSNs). Recent advancements in sensors, processors, and wireless communications have improved the quality and acquisition speed of data in WSNs. However, the data gathered by a WSN is inherently random due to component and environmental variations. Thus, statistical signal processing algorithms are needed to analyze the random data in a robust way. Though many algorithms for the analysis of random data are established and available, they are problem-specific and must be adapted to the application. This thesis provides an analysis of established localization algorithms for wireless sensor networks using single-bit received signal strength (RSS) data. The algorithms are evaluated via Monte Carlo simulation using root mean square error performance metric with respect to design parameter variations. Three scenarios are considered including a single source scenario with known parameters, a single source scenario with unknown parameters, and a multi-source scenario where an additional source is present to add interference to sensor measurements. Two algorithms are selected: the Varshney algorithm, based on the most popular maximum likelihood estimator and the Michaelides algorithm, an empirical variant where prospective locations are scored based on observations of nearby sensors. The algorithms are evaluated in [7] and [8] under different conditions. Here, each is evaluated under identical conditions and assumptions. It is shown that the algorithms perform similarly under ideal conditions, and the empirical Michaelides algorithm can outperform the Varshney algorithm under high interference scenarios.


49 pages




Northern Illinois University

Rights Statement

In Copyright

Rights Statement 2

NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.

Media Type