Cognitive Radio Cooperation Spectrum Sensing Based On Elite Opposition-Based Group Search Optimization

With the development of the Internet era, wireless communication has played an important role in the daily work and life, the wireless spectrum resource that wireless communication relies on is very limited. A large number of reports have indicated that the use of authorized spectrum exist a lot of waste, the spectrum utilization rate is low. How to improve spectrum efficiency under the original static spectrum allocation system, which has become the main direction of the radio researches naturally. In order to solve this problem, cognitive radio technology arises at the historic moment, spectrum sensing technology which is as one of the key technologies has been greatly concerned. Spectrum sensing technology can be divided into Single node perception and collaboration perception generally. Collaborative perception technology can solve some problems existing in Single node perception technology, it becomes a research hotspot.The paper makes a brief overview on the research background, significance and research status quo of cognitive radio firstly, and it also makes a related introduction on the detecting principle and characteristics of the common single point of spectrum sensing and cooperative spectrum sensing technology. This paper analyzes the energy test and centralized collaborative perception methods specially, and it made a detailed description of the data fusion algorithm in collaborative perception. After that, using the energy test method and linear fusion algorithm to build a mathematical model of collaborative spectrum sensing based on the collaborative spectrum sensing framework, all of that turn the spectrum perception problem into the use of intelligent algorithm to calculate the optimal value of the function.In order to improve the convergence speed and convergence precision of the algorithm in calculating the optimal value problems. The paper, on the basis of the existing group search algorithm, proposed to introduce the elite opposition learning into group search algorithm, and compared with the traditional group search algorithm and particle swarm optimization on the contrast test of11classical test functions. The final test results show that the proposed algorithm has a good performance on convergence precision and convergence speed, especially in the case of high-dimensional. Then, the proposed algorithm has been applied to the problem of collaborative spectrum sensing, when the dimension of the fitness function respectively in6,18and30, compared with the traditional group search algorithm and particle swarm optimization on the convergence performance, the simulation results verify the superior performance of the new algorithm again.