一种基于时频相关性的快速频谱感知算法  

A Fast Spectrum Sensing Algorithm based on Time-Frequency Correlation

作　　者：林波[1] 丛键[1] 

机构地区：[1]中国电子科技集团公司第三十研究所,四川成都610041

出　　处：《通信技术》2015年第11期1261-1264,共4页Communications Technology

摘　　要：频谱感知技术是认知无线电的核心和基础,而自从认知无线电技术被提出以来就一直受到极大的关注。通过对无线频谱占用情况的实时采集处理,对频谱占用情况在时域和频域上的相关性进行分析。利用这种时频域的相关性提出一种快速频谱感知算法,通过频繁模式挖掘提取有效模式,使得在只需要对部分频域和时域的频谱进行检测,就可以完成对整个频谱的占用情况的认知。最后通过实时采集数据进行实验分析,表明该算法确实能达到较好的性能。Spectrum sensing technology is the core and foundation of cognitive radio, and cognitive radio always attracts great attention since its presence. Through the real-time acquisition and processing of wire- less spectrum occupation, the correlation of spectrum utilization both in time domain and frequency domain is analyzed. Based on this time-frequency domain correlation, a fast spectrum sensing algorithm is pro- posed. Through frequent pattern mining and extracting effective model, and only by detecting the frequency spectrum of partial frequency-domain and time-domain, this algorithm could realize cognition of the whole channel state. Finally, experimental analysis based on real-time data collection indicate that this algorithm could achieve fairly zood oerformance.

关 键 词：频谱感知 认知无线电 信道预测 频繁模式挖掘 

分 类 号：TN918[电子电信—通信与信息系统]