利用ε贪婪的认知无线电网络频谱感知策略

Spectrum sensing policy in cognitive radio network based on ε-greedy

为提高认知无线电网络(CRN)的检测率和频谱利用率,提出一种基于ε-贪婪和D-S证据的协作频谱感知策略。在频谱探测阶段,利用ε-贪婪算法统计主用户(PU)的占用情况并估计次级用户(SU)的检测性能,利用具有固定分集度的伪随机跳频码,实现无线电频谱的快速扫描;在频谱利用阶段,通过D-S证据规则将感知作业问题转变为二进制整数规划问题;在检测性能约束下,最小化每个子带感知器的数目,获得最优的感知任务分配方案。实验结果表明,该策略具有很好的收敛性,能够避免主要与次要用户之间的频谱占用冲突,有效提高网络吞吐量。

To improve the detection rate and spectrum utilization of cognitive radio （CR） networks, a cooperative spectrum sensing policy based on ε-greedy and D-S evidence was proposed. In the spectrum detection stage, the ε-greedy algorithm was used to count the occupancy of primary users （PU） and estimate the detection performance of secondary users （SU） . The scan of the radio spectrum was realized using pseudo-random frequency hopping code with fixed diversity degree. In the stage of spectrum utilization, the problem was transformed into a binary integer programming problem using D-S evidence rule. Under the detection performance constraints, the number of subband sensor was minimized and the optimal sensing task allocation scheme was obtained. Experimental results show that the proposed method has good convergence, and it can avoid the spectrum occupy con flict between primary and secondary user, and improve the network throughput.