Distributed Detection of Correlated Signal Using Markov Chain Monte Carlo (Poster)

Abstract

The distributed detection problem with consideration of correlated sensor observations is an NP-hard problem. In this paper, a heuristic Markov Chain Monte Carlo algorithm, which consists of methods of slice sampling and simulated annealing, is investigated to solve this problem. Based on the criterion of minimizing the probability of error, sub-optimal solutions including fusion rules and sensor decisions are acquired. The performance of this algorithm is studied with analysis of experimental results.