Display Full Text | Download Full Text
608.43 kB PDF file

Degree

Master of Science, Ohio State University, Electrical Engineering, 2002.

Abstract

Unattended sensor networks are becoming increasingly valuable for many military and commercial applications. A number of sensors are distributed in a region of interest. These sensors have the ability to sense and record energy, process data, and communicate with a central information processor. The information gained from signal processing is often used for detecting, tracking, and identifying objects of interest. In certain circumstances, location and orientation information regarding these sensors is unknown after being placed in the scene. The problem considered in this thesis is how to locate and orient these sensors. We present methods for solving this problem using calibration source signals. Sources are distributed in the same region of interest, with their locations and signal emission times being unknown. Each sensor has the ability to generate direction-of-arrival (DOA) and time-of-arrival (TOA) estimates from the source signals. The goal is to estimate the locations and orientations of all sensors using these TOA and DOA measurements. We develop necessary conditions for solving the self-calibration problem and provide a maximum likelihood solution and corresponding location error estimate. A lower bound on calibration accuracy via the Cramer-Rao Bound is found. We also consider the problem of locating and orienting a network of unattended sensors using nominal location information in the form of a prior probability distribution function. We develop a Bayes approach to the calibration problem and compute accuracy bounds on the calibration procedure. A maximum a posteriori estimation algorithm is shown to achieve the accuracy bound. Results using both synthetic data and field measurements are presented.

Keywords

SENSOR; Sensor and Source; SELF-CALIBRATION; DOA; SENSOR NETWORKS; TOA

Advisor

Randolph L. Moses

Pages

74p.

Document number: osu1023465547
Permalink: http://rave.ohiolink.edu/etdc/view?acc_num=osu1023465547

★

This ETD has been downloaded 1407 times (through March 2013)

This ETD was one of the top ten most frequent downloads of May 2005 and June 2005.