Throughput and Spectrum Sensing Trade-Off by Incorporating Self-interference Suppression for Full Duplex Cognitive Radio

Abstract

As the growth of wireless applications increasing very rapidly, the requirement of radio frequency spectrum also increases. But due to limited frequency spectrum, it has become overcrowded. In order to optimize spectrum utilization, cognitive radio has emerged, which allows secondary users to coexist in licensed users spectrum band and when found primary user in RF channel, secondary user required to evacuate that channel for a specific time interval and introduces significant latency for primary user. In traditional half duplex systems, secondary user first sense the channel and then transmission will happen. In this paper, full duplex communication with self interference suppression is proposed to optimize detection probability of primary user and secondary user throughput. In the presence of self-interference, waveform based spectrum sensing technique for WLAN channel usage is adopted to allow secondary user to detect primary user with high accuracy. Our objective is to decrease latency and improve throughput subject to varying self-interference of secondary user. Results of the proposed method implemented using Matlab shows reduced latency by a factor of 2.1 in comparison with full duplex slotted method and 14 in comparison with half duplex method by considering zero SIS and non-zero SIS.