Regularization methods for intercepted radar signals

A transmitted, unknown radar signal is observed through more than one path in additive noise. The aim is to recover the waveform of the intercepted signal and to simultaneously estimate the DOA. We propose an approach exploiting the parsimonious time-frequency representation of the signal by applying regularization methods with sparsity inducing norms. An important issue is the scalability of the algorithm since high-dimensional models shall be used for radar signals. Simulations for modulated signals illustrate the good performance of the method even for low signal-to-noise ratios and a gain of 20 dB for the DOA estimation compared to a standard method.