Feature detection based computationally efficient spectrum sensing by test statistics sharing among multiple receive antennas

This paper presents a feature detection based spectrum sensing technique by some test statistics sharing among multiple receive antennas. In conventional feature (signal cyclostationarity) detection based spectrum sensing techniques with multiple antenna systems, computational complexity is not low, and it increases in proportion to the number of receive antennas. In this paper, test statistics are shared among multiple receive antennas for improving computational efficiency. The results are compared with results obtained using some conventional techniques and they show that 1) the presented technique can obtain favorable results even though the presented technique has a low computational complexity in comparison with the conventional techniques; 2) the performances and computational complexity of the presented technique are better and lower than that of conventional techniques in proportion to the number of receive antennas; 3) when noise variances at each of the RF chains are different, degradation of performance can be suppressed by adjusting computational complexity.