Adaptive spectrum sensing of wireless microphones with noise uncertainty

Many spectrum sensing techniques have been proposed in the literature to enable cognitive radio technology. However, their reliability when primary users have very low signal-to-noise ratio (SNR) in the presence of noise uncertainty remains a challenging problem. This paper focuses on detecting wireless microphone signals in the presence of noise uncertainty. Power Spectrum Density (PSD)-based sensing has been proposed in the literature as the best sensing algorithm for wireless microphones. However, when there is noise uncertainty, PSD-based sensing performance is severely degraded. To solve this problem, eignevalues-based blind sensing, which does not need noise information, have been proposed. In this paper, we present a new adaptive spectrum sensing algorithm that outperforms both PSD-based sensing and the eigenvalues-based sensing in the presence of noise uncertainty. The algorithm combines the decisions of the two algorithms, and then, adapts the decision threshold required for the PSD-based sensing in an iterative way. Simulation results show that the proposed spectrum sensing algorithm outperforms the PSD-based sensing in the presence of 1 dB noise uncertainty by more than 2 dBs. At the same level of noise uncertainty, our algorithm outperforms the eigenvalue-based sensing by 1.2 dBs.