Approximate Local Utility Design for Potential Game Approach to Cooperative Sensor Network Planning

Abstract

This paper addresses information-based sensing point selection from a set of candidate sensing locations, which determines a set of measurement points maximizing the mutual information between the sensor measurements and the variables of interest. A potential game approach has been applied to implementing distributed decision making for cooperative sensor planning. When a sensor network consists of a large number of sensing agents, the local utility function for a sensing agent requires a lot of computation, because the local utility function depends on the other agents decisions while each sensing agent is inherently faced with limitations in both its communication and computational capabilities. Accordingly, a local utility function for each agent should be approximated to accommodate limitations in information gathering and processing. We propose an approximation method of a local utility function using only a part of the decisions of other agents. The part of the decisions that each agent considers is called the neighboring set for the agent. The error induced by the approximation is also analyzed, and to keep the error small we propose a neighbor selection algorithm that choose the neighbor set for each agent in a greedy way. The selection algorithm is based on the information structure of measurement variables taken by the agents. We illustrate the approximation method and the neighbor selection algorithm through a numerical simulation on simplified weather forecasting.