The Impact of Channel Type on Spectrum Sensing

In this letter, we investigate the impact of channel correlation and line-of-sight (LOS) on the performance of spectrum sensing, focusing on energy detection (ED) with estimated noise power and the sphericity test (ST). We derive the expected value and variance of the test statistic for ED with estimated noise power, showing that detection probability decreases with channel correlation but increases with LOS propagation. For the ST, we show that detection improves with increased LOS, but the effects of correlation are complex. Depending on the system parameters correlation can improve or degrade detection performance. The analysis is corroborated by means of simulations for correlated Rayleigh, Ricean and ray-based channel models. Finally, we leverage these results to investigate the performance of the recently proposed mixture detector (MD).