Joint DOA and frequency estimation with sub-Nyquist sampling for more sources than sensors

摘要

Multiple antenna systems with sub-Nyquist sampling technique have been applied to jointly estimate frequency and direction-of-arrival of narrowband far-field signals. Unlike the current some techniques, where only one signal can be handled in some certain subband, the authors propose a novel technique to deal with the case of multiple signals. The authors first build a new time-space union model for this case and analyse the identification of the model, and then provide the maximum number of signals which can be classified. The authors also prove that the Cramér-Rao Bound (CRB) of the model is lower than the CRB based on Nyquist sampling. Simulation results verify the identification capacity and show that the proposed algorithm has a superior estimation performance close to CRB especially in the case of the minimum redundancy array case.