Fast spectrum search and detection for BPSK signals in a cognitive network

In order to achieve a dynamic spectrum management as in cognitive radio, the first task is to sense the spectrum holes. This paper is trying to make a contribution in the challenge of spectrum sensing for cognitive radios. We consider a case in which the local sensors involved in the spectrum sensing are aware of the primary signal's digital modulation. As a case of study the primary user's signal is assumed to have a binary phase-shift keying (BPSK) modulation, we therefore concentrate only on the presence detection of a BPSK signal. This work presents an innovative solution for presence detection of BPSK signals, in additive white Gaussian noise. We state the problem as a binary hypothesis test considering the noise variance, complex signal amplitude, and the symbol code sequence unknown. A new and simple search method for solving the maximization problem involved in deriving the generalized likelihood ratio (GLR) test is proposed. We evaluate the performance of the detector based on our simple search scheme and compare it with that of the exact GLR obtained via complete search and also with the previous works. The simulation results show that our proposed detector performs just as good as the exact GLR detector, with much less computational complexity.