基于改进的C-均值聚类算法的动态频谱接入策略  

Dynamic spectrum access strategy based on C-means clustering

作　　者：滕志军 曲福娟 谢露莹 刘琰 马欣 朱婉宁 

机构地区：[1]东北电力大学信息工程学院,吉林吉林132012 [2]国网吉林供电公司信息通信分公司,吉林吉林132011

出　　处：《中国电子科学研究院学报》2017年第6期649-653,667,共6页Journal of China Academy of Electronics and Information Technology

基　　金：国家自然科学基金项目(51277023)

摘　　要：为尽可能满足认知用户对多业务的需求,不断提高频谱资源的利用效率,提出一种基于C-均值聚类算法的动态频谱接入技术。通过建立可接入信道的多目标矩阵模型,利用多目标聚类的方法完成对信道频谱池的准确划分,不仅能满足认知用户的不同业务需求,还有效提高了频谱利用效率。仿真结果表明,多目标聚类方法能更准确、更快速的依据信道性能实现频谱池准确划分,提高认知用户的频谱接入效率。In order to meet cognitive user demand for multi-service knowledge as possible,and constantly improve the efficiency of spectrum resources,a dynamic spectrum access technology based on C-means clustering algorithm is proposed. By building a multi-objective model of the access channels and using the method of multi-objective clustering complete the accurate division of the channel spectrum pool. It can not only meet the different business needs of cognitive users,but also effectively improve the efficiency of spectrum utilization. The simulation results show,multi-objective clustering method can more accurately and faster based on the channel performance to achieve accurate division of the spectrum pool to improve the cognitive user＇s spectrum access efficiency.

关 键 词：认知无线电 多目标矩阵模型 频谱池 C-均值聚类算法 动态频谱 

分 类 号：TN92[电子电信—通信与信息系统]