Waveform design and scheduling for agile sensors for target tracking

Waveform agile sensors are capable of modifying transmitted waveforms to adapt to a particular scenario of interest. In this paper, we investigate the problem of optimizing the pulse length and frequency sweep rate of transmitted sonar waveforms to track a target moving in two dimensions using two sensors. We use simultaneous perturbation stochastic approximation (SPSA) in conjunction with a particle filter tracker to optimize the waveform parameters to minimize the expected squared tracking error at each time (greedy minimization). Simulation results show that waveforms with optimized parameters provide better performance than those that use fixed parameters.