Sensor selection for optimal target tracking in sensor networks

Abstract

To realise the capability of the network centric warfare (NCW) systems concept, the coordination between various sensor platforms (e.g., naval units, combat aircrafts, helicopters or unmanned aerial systems) will have to be increased and cross-platform sensor management (SM) will be applied. The sensor selection process, as part of SM, serves to find the appropriate sensor for doing an observation. With a properly working selection process, sensor deployment can be optimised for the entire sensor network. This paper presents a sensor selection algorithm (SSA) for a radar target tracking scenario. It is based on the expected performance, computed with the modified Riccati equation. The particle filtering technique is used for target tracking. The SSA takes limited detection probability into account and various expected performance criteria. Scenarios for two non-moving co-located radar systems and three simple target trajectories are simulated to validate the sensor selection strategy and demonstrate the tracking algorithm performance.