Research On Cooperative Spectrum Detection Algorithm Based On Soft Data Fusion

With the rapid development of mobile Internet in recent years lead to wireless data flows is increasing exponentially and the bandwidth resources that can be authorized are less and less.As a dynamic spectrum access mechanism,cognitive radio has great potential advantages for improving the spectrum efficiency and frequency spectrum utilization of future wireless communication system.Therefore,it becomes one of the effective solutions for the spectrum access of the future wireless communication system.This paper focuses on the research hotspot of spectrum detection technology in cognitive radio,and makes a comprehensive research on user power control in cooperative spectrum detection,proposed three cooperative spectrum detection algorithm based on cognitive user power control.Meanwhile,In view of the shortcomings of the traditional clustering cooperative spectrum detection scheme,a cluster weighted cooperative spectrum detection optimization scheme is proposed.The main work and innovation of this paper include the following two aspects:1.The system modeling of traditional cooperative spectrum detection is analyzed,and focus on cognitive user power control aided cooperative spectrum detection algorithm.In order to make the detection scenario more realistic,the detection environment is modeled as a two-level channel,which consisting of a detection channel and a reporting channel.Analyzed and deduced the optimal report power of each user in the unknown and known joint channel statistics,and proposed three kinds of power control algorithm.At the same time,the entire power control process is coordinated by the fusion center,which can obtain a rough estimation of the channel statistical and obtain the optimal power control vector,effectively set the each user's reporting power in the observation data reporting phase and avoiding the huge system overhead caused by the user's individual power control.Compared with the existing equal power schemes,the proposed power control aided cooperative spectrum detection scheme can improve greatly the detection performance.2.a cooperative spectrum detection optimization scheme based on clustering weighting is proposed.Due to each user in the cluster is relatively concentrated and experiences similar channel characteristics,in order to reduce the time delay and system overhead for the user to report the spectrum measurements,and also reduce the error probability of the decision,the proposed clustering weighted cooperative spectrum detection optimization scheme obtained the optimal number of users and the decision threshold when they cooperate to report the spectrum measurements under the constraint of the minimum error probability.Moreover,In order to reduce the congestion and latency of multi-user reporting measurements caused by limited reporting channel bandwidth,the soft data fusion is used at the cluster head to make better use of the hidden primary user information in the spectrum observations,the proposed clustering-weighted cooperation scheme has a smaller error probability and smaller time delay.The above two research content focuses on cooperative spectrum detection in cognitive radio,which focused on the shortcomings of the current cooperative spectrum detection scheme.And both are in the data collection stage of cooperative spectrum detection,through the processing of this stage,we can finally get the accurate channel state decision in the fusion center.