Research On Efficient Cooperative Spectrum Sensing

The contradiction between performance and cost has now became more and more severe during the evolution of wireless communication technology. The radio spectrum scarcity and the inefficiency of stationary spectrum allocation strategy is one of the critical problems which requiring a more effective utilization of spectrum resources. Cognitive radio technology is a brand new solution for the spectrum utilization efficiency problem, whose main idea is to pursue and utilize the"spectrum hole"within the authorized spectrum, in order to opportunistically access to the authorized spectrum without disturbing the authorized users. So cognitive radio is a key solution for spectrum utilization inefficiency problem.However, the performance-cost contradiction exists even within the cognitive radio technology. Spectrum sensing is one important topic in cognitive radio research, and cooperative spectrum sensing is a novel sensing method to achieve high accuracy and reliability. But during the cooperative spectrum sensing process many unauthorized users are involved so the high sensing performance is at the cost of enormous resource consuming. The main contribution of this thesis is to research effective and reliable cooperative spectrum sensing strategy which achieves acceptable accuracy at a minimum resource cost so that the resource usage efficiency can be maximized. Besides, the strategy has resistance to some illegal competition for the spectrum.An ideal circumstance has been researched in chapter 3, i.e. all the unauthorized users have the same local sensing accuracy(including false alarm probability and miss detection probability). In order to develop efficient cooperative sensing strategy in such case, this chapter has driven the main approaches for cooperative efficiency promotion and the framework of efficient cooperative sensing system, on the basis of which an improved Chair-Varshney decision fusion rule has been developed. Given the target sensing accuracy, the improved Chair-Varshney rule can be used to achieve the optimal number of sensing users, so a random choose can be adopt to get the sensing users. The simulation results have shown that only a small part of the total unauthorized users are needed in the sensing process to get the target sensing accuracy, so that the resource cost has been reduced and the efficiency of the cooperative sensing system has been improved.In chapter 4 a common circumstance has been researched, i.e. the sensing accuracy of the unauthorized users are not identical, besides there are malicious users who transmit false sensing data intentionally. In this case it is infeasible to achieve the optimal number of sensing users though mathematical analysis, but this chapter utilizes the estimation of the history judgment accuracy of each sensing user, and adopts an iterative search to choose the appropriate unauthorized users in cooperative sensing in attempt to improve the system efficiency. The simulation results have shown that this cooperative sensing strategy has a quick convergence and indeed minimizes the number of sensing users and system resource cost at a given sensing accuracy. Moreover, this strategy has an resistance to the malicious users who intent to inject false data into the cooperative sensing system.