Abstract

We study distributed estimation (DES) problem in power and bandwidth constrained wireless sensor networks (WSNs), where several sensors make noisy observations of an unknown, and transmit a locally processed version of their observations to a fusion center (FC) over wireless channels. The FC reconstructs the unknown via fusing the received data from sensors. We explore the following problems: (i) we derive Bayesian Fisher information matrix (FIM) for bandwidth-constrained DES of a Gaussian vector with linear observation model, where sensors transmit their digitally modulated quantized observations to the FC over power-constrained fading channels. We develop two transmit power allocation schemes from solving the maximization of trace and log-determinant of Bayesian FIM, subject to network transmit power constraint, and study the system performance using these schemes. (ii) Consider the DES of a Gaussian source in a hierarchical power-constrained WSN. Sensors within each cluster send their noisy measurements to a cluster head (CH). CHs fuse the received signals and transmit to the FC over orthogonal fading channels. To enable estimation of these fading channels at the FC, CHs send pilots to the FC, prior to data transmission. We derive the MSE corresponding to the LMMSE estimator at the FC, and explore the best power scheduling scheme among sensors and CHs, to minimize the MSE subject to network transmit power constraint. (iii) Assuming the DES of a Gaussian source with additive and multiplicative Gaussian observation noises, we derive different estimators such as minimum mean square error (MMSE), maximum-a-posteriori (MAP), and different lower bounds on MSE, such as Bayesian Cramer-Rao bound (BCRB), Weiss-Weinstein bound (WWB). We characterize the scenarios that multiplicative noise improves the DES performance (we call the phenomena as enhancement mode (EM) of multiplicative noise), when we assume the variance of multiplicative noise is known/unknown, and also when the observations are quantized/unquantized.

Notes

If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu

Graduation Date

2019

Semester

Summer

Advisor

Vosoughi, Azadeh

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering and Computer Science

Department

Electrical and Computer Engineering

Degree Program

Electrical Engineering

Format

application/pdf

Identifier

CFE0008101; DP0023240

URL

https://purls.library.ucf.edu/go/DP0023240

Language

English

Release Date

2-15-2021

Length of Campus-only Access

1 year

Access Status

Doctoral Dissertation (Open Access)

Download

Share

COinS