Efficiency-Accuracy Trade-off for Spectrum Sensing in Cognitive Network

Abstract

Spectrum sensing is the first and foremost step in cognitive radio technology, where sensing efficiency and accuracy cannot be simultaneously optimized. Tradeoff between the two metrics is represented by sensing duration and sensing period, which are among the most important parameters in spectrum sensing. Aiming at maximize overall available spectrum opportunity, an efficiency-accuracy tradeoff is proposed with interference to licensed user under specified threshold. Based on Neyman-Pearson criterion, joint optimal mathematical model is established and closed-form expressions are derived. Optimal sensing parameters can be obtained and simulation results verified our proposed scheme.