Multiple Antenna Based Low Complexity Spectrum Sensing with Binary Phase Rotator Selection

This paper presents a computationally efficient energy detection based spectrum sensing with multiple receive antennas. In traditional spectrum sensing with multiple receive antennas (soft decision case), a statistic for the signal detection is computed at each receive antenna, and these statistic are combined for the signal detection. Therefore, the computational complexity of spectrum sensing increases as the number of receive antenna. In the presented technique, received signals are combined (added/subtracted) firstly, and the statistic for the signal detection is computed from the combined signal. Because only one statistic computation is required regardless of the number of receive antenna, the computational complexity of the spectrum sensing can be reduced. In order to prevent the cancellation of each received signal at the combining due to the phase uncertainty, the presented technique employs the binary phase rotator which can take +/- 1. By choosing an appropriate rotator for each received signal, we attempt to maintain the magnitude of the combined signals. Furthermore, we present the spectrum sensing schemes to obtain a suboptimum phase rotator. Some numerical examples are provided to valid the effectiveness of the presented technique, and these results show that the presented technique is effective for the signal detection of narrowband modulation signals.