A collaborative spectrum detection based on MTM with low complexity

The identification of the primary user is one of the most important issues in the spectrum sensing process for the cognitive radio networks. In the paper, a spectrum detecting algorithm is proposed based on the Neyman-Pearson criterion. The optimum threshold formula is obtained under multi-sensor environment. The optimal required number of the sensors is determined which is adapted to the signal to noise ratio (SNR) in every frequency bin. Simulation results indicate that under constraints of the false alarm probability P_FA=0.1, signal to noise ratio SNR=0 dB, number of tapes K=4, only four sensors are required to achieve the best detection performance under the Neyman-Pearson criterion. Secondly, because the improved algorithm does not require singular value decomposition (SVD) for denoising, the proposed algorithm also reduces the complexity of the system compared with traditional algorithm based on multitaper method with singular value decomposition (MTM-SVD).