Efficient linear time-varying system identification using chirp waveforms

Linear, time-varying (LTV) systems are operators composed of time shifts, frequency shifts, and complex amplitude scalings that act on continuous finite-energy waveforms. This paper builds upon a novel, resource-efficient method previously proposed by the authors for identifying the parametric description of such systems from the sampled response to linear frequency modulated (LFM) waveforms. If the LTV operator is probed with a sufficiently diverse set of LFM pulses, more LFM pulses than reflectors, then the system can be identified with high accuracy. The accuracy is shown to be proportional to the uncertainty in the estimated frequencies and confirmed with numerical experiments.