Generalized Energy-based Spectrum Sensing: Active Threshold Correction Under Noise Uncertainty

Abstract

In this paper, we propose an active threshold correction (ATC) for the generalized energy detector (GED) under spectrum sensing to alleviate the effects of noise variance uncertainty (NVU) and fading. First, we derive the probability of false alarm and the probability of detection for GED, with an arbitrary exponent, assuming that the test statistic follows a gamma distribution and under Nakagami-m fading. Next, we present the effect of NVU on the performance of GED. Later, we derive an analytic expression for the corresponding ATC under NVU and Nakagami-m fading. Through numerical method, we show that by employing the ATC, the performance of GED improves in the presence of NVU and fading. Moreover, in the considered setup for GED, we show that the GED outperforms the energy detector.