Sensor network control based on randomized and multi-agent approaches

Abstract

In this paper, a development of randomized and multiagent algorithms is presented. The examples and their advantages are discussed. Different combined algorithms which are applicable for the multi-sensor multitarget tracking problem are shown. These algorithms belong to the class of methods used in derivative-free optimization and has proven efficacy in the problems including significant non-statistical uncertainties. The new algorithm which is an Accelerated consensus-based SPSA algorithm is validated through simulation.The main feature of that algorithm, combining the SPSA techniques, iterative averaging (“Local Voting Protocol”) and Nesterov Acceleration Method, is the ability to solve distributed optimization problems in the presence of signals with fully uncertain distribution; the only assumption is the signal’s limitation.