Multiantenna-Assisted Wideband Spectrum Sensing Based on Sub-Nyquist Sampling

We consider the problem of wideband spectrum sensing, where a multi-antenna secondary user (SU) attempts to identify the frequency locations of a number of single-antenna primary users (PUs)' transmissions that spread over a wide frequency band. To overcome the sampling rate bottleneck, we propose a new sub-Nyquist sampling architecture as well as a compressed power spectrum estimation method whose objective is to reconstruct the power spectrum of the transmitted signals from PUs. The proposed sub-Nyquist sampling scheme has a simple structure, and a power spectrum recovery analysis shows that the proposed method can achieve a significant sampling rate reduction, without imposing a sparse constraint on the frequency band under monitoring. Simulation results are presented to show the effectiveness of the proposed method.