The Research Of The Optimization Of Cooperative Spectrum Sensing Based On Double-threshold

The rapid expansion of wireless communication technology has provided a more powerful and convenient means for people’s production and life. However, the wireless spectrum resource as the other valuable natural ones is becoming more and more precious. Research on cognitive radio technology has been extensive attention, taking into account of limited spectrum resource in recent years. As the core technology of cognitive radio, spectrum detection offers perception users a way to access the licensed spectrum bands in an opportunistic way. By means of sensing spectrum holes cognitive radio systems can improve spectrum utilization. Meanwhile, cooperative spectrum sensing mitigates the effects of shadowing and multipath fading, when only one cognitive radio is used, and enhances the ability of the system to oppose disturbance.To achieve the research aim of optimization strategy for collaborative spectrum sensing, the main researches of this thesis are around the characteristics and structure of radio spectrum detection, the mechanism of double-threshold energy detection and the fusion methods of collaborative network. The main work of this thesis has achieved as follows:1) This thesis has analyzed the characteristics and application conditions of common spectrum detection model. Energy detector is selected as basic methods of detection for this thesis. Focusing on performance of multi-user cooperative perception system using different fusion rules, induced the global error probability as the main performance indicators. The optimal fusion threshold of the k-out-of-N rule is obtained by theoretical derivation, which establishes the base of the double threshold collaborative sensing system optimization in order to minimize the system error probability.2) Combined with the derived expression, this thesis improved the traditional k-out-of-N rule fixed voting threshold application in fusion center. Perception system performance optimization is respectively accomplished by dynamic adjustment of voting threshold in order to minimize the system error probability. This effectively solves that traditional fixed-voting threshold does not adapt to the changes in the external environment, which satisfies high efficiency, anti-interference, scalable performance demand of double threshold perception system.3) This thesis improved partition algorithm by the combination of EGC fusion algorithm and variable-threshold k-out-of-N rule, and established the mathematical model. The operation efficiency and results display the effect of perception in collaboration system is improved and the global error probability is degraded by simulation in AWGN and Rayleigh channel, which provides guarantee for system communication under the fading environment.