Sensor Selection and Power Allocation via Maximizing Bayesian Fisher Information for Distributed Vector Estimation

TL;DR

This paper addresses sensor selection and power allocation in wireless sensor networks to optimize distributed Gaussian vector estimation by maximizing Bayesian Fisher Information, proposing algorithms that outperform uniform power distribution.

Contribution

It introduces a novel formulation for sensor selection and power allocation based on Bayesian Fisher Information maximization, with three algorithms to solve the problem.

Findings

Proposed algorithms outperform uniform power allocation.

Maximizing Bayesian Fisher Information improves estimation accuracy.

Algorithms are effective under power constraints.

Abstract

In this paper we study the problem of distributed estimation of a Gaussian vector with linear observation model in a wireless sensor network (WSN) consisting of K sensors that transmit their modulated quantized observations over orthogonal erroneous wireless channels (subject to fading and noise) to a fusion center, which estimates the unknown vector. Due to limited network transmit power, only a subset of sensors can be active at each task period. Here, we formulate the problem of sensor selection and transmit power allocation that maximizes the trace of Bayesian Fisher Information Matrix (FIM) under network transmit power constraint, and propose three algorithms to solve it. Simulation results demonstrate the superiority of these algorithms compared to the algorithm that uniformly allocates power among all sensors.