Research On Intelligent Algorithm Of Collaborative Spectrum Detection Under Noise Uncertainty

The contradiction between the low utilization of wireless spectrum resources and the demand for wireless spectrum has been widely used as a technology that can effectively solve the problem of spectrum shortage at this stage.As a key technology in cognitive radio,collaborative spectrum detection can effectively improve the detection performance of the cognitive system.In a real environment,there are various kinds of interference,and how cognitive nodes detect collaborative spectrum under such a noisy environment becomes a problem.In this thesis,the cooperative spectrum detection algorithms based on evolutionary game theory and cooperative spectrum detection algorithm based on multi-bit are applied to the environment with noisy noise so as to improve the system detection performance.The research contents and innovations of this thesis are divided into two parts as follows:(1)It is studied that under the condition of noise uncertainty,cognitive users will be more inclined to reduce the detection consumption to improve their own throughput.Applying evolutionary game theory to collaborative spectrum detection under noise uncertainty,cognitive users can dynamically choose whether to participate in collaborative spectrum detection and obtain an evolutionary stable strategy(ESS)through continuous iterative learning.All cognitive users as a whole to participate in the game by the evolutionary game theory algorithm to get the average cognitive users involved in the collaboration throughput and its average throughput compared with all users.If it is not equal,iteratively iteratively until equal,so as to obtain the final evolutionary stability strategy.(2)The multi-bit collaborative spectrum detection algorithm based on credibility is applied to collaborative spectrum detection under noise uncertainty to set a credible level for each cognitive node.After each collaborative spectrum detection,an update of credibility is performed.When the weighted average value of the decision result of the cognitive node and the weighted average of the judgment results of all the cognitive nodes is large,the cognitive node is reduced.Credibility,and vice versa.When the credibility of a cognitive node is reduced to a certain threshold value,it is determined that the node will no longer be trusted,and the weighted weight of the result of the node becomes 0 in the cooperative fusion decision center,effectively improve the system's detection performance.