An EMD-Based Double Threshold Detector for Spectrum Sensing in Cognitive Radio Networks

In this paper, an adaptive multi-channel energy detector is presented for spectrum sensing applications. This method exploits the Empirical Mode Decomposition (EMD) and Cell-Averaging Constant False Alarm Rate (CA-CFAR) in an effort to maximize the probability of detection for a given probability of false alarm. First, the oversampled baseband signal is compared to its corresponding EMD noise-only model. If the band is occupied, then an EMD-CA technique is used to estimate the noise power for a given false alarm probability. Then, from the estimated noise power, double thresholds based on two given false alarm rates are calculated to detect and localize the occupied channels of band of interest. The proposed approach is able to work blindly and it is independent of the noise power. Simulations for different sampling and false alarm rates are used to validate the performance of the proposed detector. The results revealed the robustness of the proposed technique to the noise uncertainty and the capability to sense and localize multiple channels simultaneously.