A geometric feature-aided game theoretic approach to sensor management

Given the single-point node failure limitation of centralized sensor management schemes, decentralized sensor management (DSM) techniques are increasingly important for a simultaneous tracking and identification system. DSM options are realizable with the support of modern systems through increased bandwidth, wireless communication, and enhanced power, but require novel strategies to address constraints imposed by a large number of users. Since game theory offers competitive models for distributed allocations of surveillance resources and provides mechanisms to handle the uncertainty of a surveillance area, we propose an agent-based negotiable game theoretic sensor management (ANGSm) approach. By incorporating subgame Nash equilibrium into negotiation, all agent's needs are considered. With the DSM scheme, sensor assignment occurs locally, and with no central node, DSM reduces the risk of whole-system failures. To make the ANGSm approach more efficient and practical, a geometric feature, the range between a target and a sensor, is also incorporated in the bargaining strategy during sensor assignment. Simulation results of the geometric feature-aided game theoretic (GFGT) approach demonstrate the applicability of the proposed approach.