Wideband Spectrum Sensing Based on Sub-Nyquist Sampling

In this paper, we consider the problem of locating multiple active spectrum subbands in a wide range of frequency bands. A major challenge associated with such wideband spectrum sensing is that it is either infeasible or too expensive to perform Nyquist sampling on the wideband signal. In this paper, we propose a sensing scheme based on a sub-Nyquist sampling method called multicoset sampling, which is similar to the polyphase implementation of the Nyquist sampling, but requires less A/D converters. In contrast to the traditional sub-Nquist approaches where the wideband signal is first reconstructed from the sub-Nyquist samples, we develop a method that directly estimates the power spectrum of the wideband signal of interest using the sub-Nyquist samples, by exploiting its statistical properties. We also characterize the statistical distribution of the proposed power spectrum estimator, based on which we obtain a constant-false-alarm energy detector for the frequency bins. Simulation results are provided to demonstrate the effectiveness of the proposed multiband spectrum sensing method based on sub-Nyquist sampling.