Analytical Investigation of Cognitive Radio Spectrum Sensing Performance in the Presence of Nigerian GSM Primary Users

Cognitive radio technology enables secondary spectrum users to opportunistically access licensed spectrum bands during periods when primary users are inactive, with spectrum sensing accuracy being the critical determinant of interference avoidance effectiveness. This study analytically investigates the spectrum sensing performance of cognitive radio receivers in the presence of primary GSM user signals representative of the operational characteristics of Nigerian mobile networks. An analytical and simulation study was conducted, combining mathematical analysis of detection probability and false alarm probability expressions for energy detection, cyclostationary feature detection, and matched filter detection techniques with Monte Carlo simulations parameterized using measured GSM signal characteristics from 900 MHz band occupancy measurements conducted in Kano and Jos. Detection performance was evaluated across a range of signal-to-noise ratios representing practical sensing scenarios. Cyclostationary feature detection achieved the highest detection probability at 0.95 for SNRs above negative 10 dB, outperforming energy detection, which required SNR above negative 5 dB for equivalent performance. Cooperative sensing with five cognitive radio nodes reduced the required SNR for 0.95 detection probability by 6.2 dB compared to single-node sensing, demonstrating the significant benefit of distributed sensing architectures. The study also evaluates the impact of GSM control channel periodicity on false alarm rates, finding that energy detection false alarm rates increase by 18% in the presence of beacon signals. Results provide parametric guidance for cognitive radio system design targeting GSM band spectrum reuse in the Nigerian regulatory context. Keywords: cognitive radio, spectrum sensing, GSM primary users, energy detection, Nigeria.