Semi-blind locally optimum detection for spectrum sensing in cognitive radio

Spectrum sensing in cognitive radio becomes a challenging task when the signals received at the secondary users´ transmitters exhibit low power. Locally optimum detectors (LOD) are therefore desirable thanks to their optimality in the low SNR regime. Here, we assume that the primary user transmits a training sequence, and propose a semi-blind LOD (SBLOD). In the case of BPSK signals, the test statistic of the proposed SBLOD is shown to be a weighted sum of the matched filter output, the energy and pseudo-energy. For higher size constellations, the SBLOD reduces to a linear combination of the matched filter and the energy detector. Although combining the matched filter and energy detector is a classical approach, our study provides a systematic and (locally) optimal way of combining these detectors. Simulations results show the merits of the proposed detector.