Research On Cooperative Spectrum Sensing Technology In Cognitive Radio

Cognitive radio (CR) can dynamically use those temporarily idle spectrum resource and improve the spectrum utilization significantly by sharing the spectrum with unlicensed users without causing the interference to the licensed users. How to reliably and rapidly detect the signals from licensed users is the principal point for CR. Spectrum sensing for that is one of the key technologies in CR. Cooperative spectrum sensing is a research hotspot of the spectrum sensing technologies. The dissertation has researched on cooperative spectrum sensing technologies in CR system and the main contents are as follows:1. The existing spectrum sensing algorithms are summarized as theoretic foundation. Spectrum sensing algorithms for single user and cooperative spectrum sensing are described and simulation results are analyzed.2. An optimal number of secondary user selection algorithm for minimizing the error detection probability is proposed. The optimal number of secondary user to minimize the modified error detection probability based on energy detection under different thresholds is deduced and cooperative spectrum sensing is performed according to the optimal number of secondary user. Compared to the traditional K/N rule, the error detection probability is degraded. Aimed at the noise uncertainty problem, a cooperative spectrum sensing algorithm with double threshold under noise uncertainty is proposed by combining the double threshold achieved according to noise uncertainty with equal gain combination, which decreases the impact of noise uncertainty to some extent. Compared to the existing algorithm the performance is better under low signal-to-noise.A cooperative spectrum sensing algorithm based on selective clustering under unperfect control channel is proposed. For the situation that the detection result sent by the cluster head may experience different fading through the control channel, some of clusters according with the condition are chosen to participate to the final decision by fuse center based on the credit regulation. Simulation results show that the detection performance is improved and also the system spending is reduced compared to the existing algorithm based on clustering.3. The algorithm about optimization of parameters for cooperative spectrum sensing mechanism to maximize the sensing efficiency is proposed. For single licensed channel, the optimal object and limitation condition are defined and the process of optimization and selection of the parameters is deduced. Simulation results show that there existing the optimal sensing parameters which maximize the spectrum sensing efficiency while satisfying the interference constraint. Then a sensing channel sequence selection method to maximize the total spectrum sensing efficiency is proposed for multi-channel. Simulation results prove the effectiveness of the proposed method.4. The realization of the selective clustering algorithm on the TMS320C6455 chip is studied. A series of measures have been taken to optimize the coding with concerns of coding and memory optimization. The test results and performance analysis are given.