基于最小生成树的分簇协作频谱感知改进算法

An Improved Clustering Cooperative Spectrum Sensing Algorithm Based on Minimum Spanning Tree

为减少分簇过程中的时延,基于最小生成树的单向比较优势提出簇首快速推举方法,并提出改进的分簇协作频谱感知算法,分析了算法的时间复杂度。算法首先基于最小划分对所有次用户节点进行分簇,簇内节点根据设置的评价条件进行性能比较,推举簇首。由簇首进行本地簇内频谱检测,并上传检测结果,最后融合中心在簇首间实现协作的频谱检测。在瑞利信道条件下,仿真显示在大信噪比时,融合中心应用AND规则,系统具有较小的虚警率,所提算法检测性能优;小信噪比时,应用OR规则能扩展系统的有效检测区间,所提算法在满足系统要求的前提下检测性能较差,但簇内信道效率提高了n-1倍。

For decreasing the delay of clustering process,a cluster head election scheme is proposed based on the single-direct comparison characteristic of minimum spanning tree. The time complexity of proposed clustering coomperative spectrum sensing algorithm is analyzed based on the fast cluster head election scheme. Here the secondary users are divided into different clusters based on minimum clique partition. A cluster head will be elected to detect the spectrum hole in the cluster according to preset metrics and trans-mit the sensing result to the fusion center. The global decision will be made by the fusion center according to results from all cluster heads. Simulation results show that the proposed algorithm can achieve the high detecion probability and low false alarm probability by employing the AND rule under high signal-to-noise ratio（ SNR） . Moreover,the algorithm can expand the available dectection area and enhance the bandwidth efficiency by employing the OR rule under low SNR.