Sub-Nyquist Sampling-Based Wideband Spectrum Pre-Sensing via Branch-to-Maximum Energy Ratio.

Sub-Nyquist sampling-based wideband spectrum sensing has attracted considerable attention due to its potential of significantly reducing the sampling rate. However, most existing methods assume the presence of primary users, focus solely on support recovery, and disregard the possibility that the concerned spectrum band may be free, leading to high false alarm rate and computational complexity. This paper proposes a novel pre-sensing algorithm called the branch-maximum energy ratio (BMER) detection algorithm, which uses sub-Nyquist samples drawn from the modulated wideband converter to calculate the ratio of branch-maximum energy. We provide a theoretical development to calculate the false alarm probability and detection probability of the proposed BMER algorithm. Simulation results indicate that the proposed BMER algorithm is robust against noise uncertainty and significantly improves the sensing performance with low computational complexity.