Artificial Intelligence Based Cooperative Spectrum Sensing Algorithm for Cognitive Radio Networks
Mohamed Mourad Mabrook, Hussein A. Khalil, Aziza I. Hussein · Procedia Computer Science · 2019
Cognitive Radio (CR) technology is regarded as a key network technology used to manage the limitation of the available spectrum in wireless communication networks. Spectrum Sensing (SS) is the core process in CR engine based on detecting the free channels and sharing it among other users. In wideband spectrum, many algorithms are proposed to sense the available free channels. Cooperative sensing is mainly considered as an effective solution of signal fading and shadowing problems in CR networks. From another side, CR networks can utilize Artificial Intelligence (AI) techniques for dynamically sensing and decision making processes. In this paper, a blind adaptive spectrum sensing algorithm based over centralized cooperative sensing platform is proposed. Then, an Adaptive Neuro-Fuzzy Interference System (ANFIS) technique is applied in decision-making process to achieve the optimum and accurate decisions. The simulation process and the output results showed that, the proposed technique outperforms both a standalone sensing techniques and other cooperative sensing with conventional decision making algorithms regarding to the probability of false alarms, probability of detection and probability of missed detection especially in low Signal to Noise Ratio (SNR). The results obtained by the proposed algorithm based on ANFIS decision making technique outperformed the other conventional decision making techniques.