Target Existence Based Resource Allocation

Abstract

Assume an ESA (electronically steered antenna) radar. In between the surveillance mode operations (where the radar completes one scan of the surveillance area), this radar has spare time for (at least one) additional mode which could be optimized for various special requirements. The optimization is often translated to the choice of the next track to update. In this paper we propose a resource allocation scheme designed to minimize the time to confirmation of tracks following new targets, as well as to minimize the time to terminate tracks whose targets are lost or have disappeared from the view. The resource allocation cost is a function of the probability of target existence of individual tracks. Simulation results show that resource allocation based on this cost significantly increases the number of targets that can be reliably tracked in the sense of them being quickly followed by confirmed tracks upon arrival, and of their tracks being quickly terminated when the targets themselves disappear. The penalty of this approach is an increase in the root mean square target trajectory estimation errors of confirmed true tracks, as they are not updated as often.