基于特征值的动态数字信道化子带检测算法  

Dynamic digital channelized sub-band detection algorithm based on eigenvalue distribution

作　　者：李晓辉[1] 万宏杰 石明利 王先文 LI Xiaohui;WAN Hongjie;SHI Mingli;WANG Xianwen(School of Telecommunications Engineering,Xidian University,Xi’an 710071,China)

机构地区：[1]西安电子科技大学通信工程学院,陕西西安710071

出　　处：《系统工程与电子技术》2024年第5期1801-1809,共9页Systems Engineering and Electronics

摘　　要：动态数字信道化接收结构通过检测所有子带,判断信号是否存在,为综合滤波器组处理提供依据,因此子带检测在接收结构中起着关键作用。针对传统检测算法存在低信噪比(signal-to-noise ratio,SNR)下检测性能不高的问题,依据随机矩阵理论,提出了基于最大最小特征值之差与平均特征值之比的检测算法,利用平均特征值和最小特征值的极限分布规律来推导算法的检测门限。其次,根据所有子带数据获取的特征值信息对所提算法进行了优化。最后,在动态数字信道化接收结构中,分析不同因素下算法的性能,表明了所提算法能够克服低SNR的影响,子带检测的性能更好。The dynamic digital channelized receiving structure detects all sub-bands to determine if there is a signal.The detection results provide a basis for processing the integrated filter bank,so it plays a vital role in the receiving structure.Since the poor performance of the traditional detection algorithm under low signal-to-noise ratio(SNR),a novel detection algorithm based on the ratio of the difference between the maximum and minimum eigen values to the average eigenvalue is proposed according to the random matrix theory.The detection threshold of the algorithm is derived by using the limit distribution law of the average eigenvalue and the minimum eigenvalue.Then,the algorithm is optimized according to the eigenvalue information obtained from all sub-band data.Finally,the performance of the proposed algorithm under different factors is analyzed in the dynamic digital channelized receiving structure,which shows that the algorithm can overcome the effects of the low SNR and perform better in the sub-band detection process.

关 键 词：动态数字信道化接收 子带信号检测 随机矩阵 特征值检测 

分 类 号：TN911[电子电信—通信与信息系统]