基于K-out-of-N融合准则的认知无线电网络优化

在认知无线电网络中，单个用户的感知很容易受到环境的影响，导致对目标频谱状态检测的误判。为提高系统频谱感知的准确性，引入了协作频谱感知机制。协作感知有效地降低了多径和阴影效应的影响，提高了系统检测精度。为了使协作频谱感知发挥出最优的性能，本文主要研究协作频谱感知中的数据融合方式，对K-out-of-N准则进行优化，考虑在漏检概率约束下，通过最小化错误概率从而求得K-out-of-N准则中的最优K值，降低协作遍历搜索K值所耗时间与能量，并在不同信道下进行仿真对比，验证优化后K-out-of-N准则的有效性。仿真结果表明，优化后的K-out-of-N准则可有效改善系统感知性能。

OPTIMIZATION BASED ON K-OUT-OF-N FUSION CRITERIA IN COGNITIVE WIRELESS NETWORKS

In cognitive radio networks, the perception of a single user is easily affected by the environment, resulting in misjudgment of the target spectrum state detection. In order to improve the accuracy of spectrum sensing, a cooperative spectrum sensing mechanism is introduced. Cooperative sensing can effectively reduce the influence of multipath and shadow effect and improve the detection accuracy of the system. In order to give full play to the optimal performance of cooperative spectrum sensing, the data fusion mode in cooperative spectrum sensing was studied in this paper, the K-out-of-n criterion was optimized, and the optimal K value in the K-out-of-n criterion was obtained by minimizing the error probability under the constraint of missed detection probability, so as to reduce the time and energy consumed by cooperative traversal search for K value. Simulation comparison was performed in different channels to verify the effectiveness of the optimized K-out-of-N criterion. Simulation results showed that the optimized K-out-of-N criterion could effectively improve the system perception performance.