Distributed Sparsity-Aware Sensor Selection

Abstract

The selection of the minimum number of sensors within a network to satisfy a certain estimation performance metric is an interesting problem with a plethora of applications. The problem becomes even more interesting in a distributed configuration when each sensor has to decide itself whether it should contribute to the estimation or not. In this paper, we explore the sparsity embedded within the problem and propose a sparsity-aware sensor selection paradigm for both uncorrelated and correlated noise experienced at different sensors. We also present reasonably low-complexity and elegant distributed algorithms in order to solve the centralized problems with convergence guarantees within a bounded error. Furthermore, we analytically quantify the complexity of the distributed algorithms compared to centralized ones. Our simulation results corroborate our claims and illustrate a promising performance for the proposed centralized and distributed algorithms.