一种基于模糊Q学习算法的认知无线电频谱分配策略  

A Spectrum Allocation Strategy Based on Fuzzy Q Learning Algorithm for Cognitive Radio

作　　者：徐勇[1] 果鑫[1] 刘丰年[1] 文鸿[1] 张文平[1] 李长云[1] 

机构地区：[1]湖南工业大学计算机与通信学院,湖南株洲412007

出　　处：《湖南工业大学学报》2013年第2期74-78,88,共6页Journal of Hunan University of Technology

基　　金：湖南省研究生科研创新基金资助项目(CX2011B393);湖南省自然科学基金资助项目(11JJ3002)

摘　　要：认知无线电是一种智能推理学习的通信系统,为了实现认知无线电频谱分配智能学习过程,引入模糊Q学习方法。认知用户通过在线Q学习来调节模糊推理系统,得到最优的频谱分配模糊规则,实现自适应的频谱分配方案。最后将模糊Q频谱分配算法与非智能学习算法(模糊频谱分配算法以及随机分配算法)进行比较,仿真结果证明了该方案能在一定程度上提高系统带宽收益,同时降低系统的冲突率。Cognitive radio is a kind of intelligent reasoning learning communication system, and fuzzy Q learning method is introduced to realize the cognitive radio spectrum allocation intelligent learning process. Cognitive users adjust the fuzzy inference system through online Q learning, obtain the optimal fuzzy rules of spectrum allocation and realize the adaptive spectrum allocation scheme. The fuzzy Q spectrum allocation algorithm is compared with non intelligent learning algorithm （fuzzy spectrum allocation algorithm and random distribution algorithm）. And the simulation results show that this algorithm improves the system bandwidth income and reduces the system conflict rate.

关 键 词：认知无线电 自适应 模糊Q 频谱分配 带宽收益 冲突率 

分 类 号：TN915.9[电子电信—通信与信息系统]