Performance analysis of blind spectrum sensing in cooperative environment

According to recent studies, most of the radio spectrum is being underutilized. Cognitive radio has been proposed as a solution to address the problem of spectrum underutilization. It allows the secondary users to occupy spectrum allocated to primary users if primary users do not use the allocated spectrum at the current time and location. To detect unused spectrum cognitive radio must perform spectrum sensing. Spectrum sensing is an important function to be performed by every cognitive radio. Either single or multiple users can take part in spectrum sensing process. Later one is called Cooperative spectrum sensing. Cooperative spectrum sensing improves sensing performance by including a number of cognitive users rather than a single one. In cooperative spectrum sensing, each cognitive radio performs individual spectrum sensing by any of the available methods and individual sensing results are combined according to certain rules. Usually, current algorithms require signal knowledge, noise knowledge or the both to perform spectrum sensing. This paper describes the idea of blind spectrum sensing based on QR decomposition and linear prediction and extends it to cooperative sensing. It does not require either signal information or noise information to perform spectrum sensing. Simulation results show that the proposed method outperforms conventional energy based cooperative spectrum sensing under noise uncertainity. This algorithm is compared with blind source seperation method and significant improvement in performance is observed.