Energy-Efficient Spectrum Sensing Using Cyclostationarity

This paper introduces a reduced complexity approach to cyclostationarity detection of primary user signals in cognitive radio systems. The decision statistic is formulated by exploiting the variations of portions the sensed signal´s spectral correlation density function estimate, due to either the presence or absence of the primary user signal. The presented detector has low computational complexity and has decision error performance which is superior to energy detection; these qualities make it suitable for energy-efficient cognitive radio terminals operating with limited power sources. The performance of this detection method is obtained by computer simulation and the computational complexity is evaluated.