Control theoretic approach to tracking radar: First step towards cognition

In Haykin (2006) [8], the idea of Cognitive Radar was described for the first time. Four essential points were emphasized in that seminal paper: Bayesian filtering in the receiver, dynamic programming in the transmitter, memory, and global feedback to facilitate computational intelligence. This paper provides a first step towards designing a cognitive radar for tracking applications by presenting a fore-active tracking radar; a radar that utilizes its previous measurements and actions to optimize its transmitted waveform (Haykin, 2011 [11]). In our design, the emphasis is being placed on the cubature Kalman filter to approximate the Bayesian filter in the receiver, approximate dynamic programming for transmit-waveform selection in the transmitter, and global feedback embodying the transmitter, the radar environment, and the receiver all under one overall feedback loop. Simulation results, based on the tracking of an object falling in space, are presented, which substantiate practical validity of the superior performance of a fore-active tracking radar over a traditional active radar with fixed waveform.