低信噪比下卷积交织器识别  

Recognition of convolutional interleaver at low SNR

作　　者：吴昭军 张立民 钟兆根 WU Zhaojun;ZHANG Limin;ZHONG Zhaogen(Institute of Information Fusion,Naval Aviation University,Yantai 264001,China;The Southwest Institute of Electronics and Telecommunications,Chengdu 610041,China;School of Basis Aviation,Naval Aviation University,Yantai 264001,China)

机构地区：[1]海军航空大学信息融合研究所,山东烟台264001 [2]西南电子电信研究所,四川成都610041 [3]海军航空大学航空基础学院,山东烟台264001

出　　处：《系统工程与电子技术》2021年第2期546-554,共9页Systems Engineering and Electronics

基　　金：国家自然基金重大研究计划(91538201);泰山学者工程专项经费(Ts201511020)资助课题。

摘　　要：针对现有卷积交织器识别算法,在低信噪比下存在误判概率高、识别效率低等缺陷,首先分析了构建出的数据矩阵统计特性,给出了同步码以及随机数据位置上的概率密度分布函数,基于最小错误判决准则,设定了同步码检测门限,同时基于三倍标准差准则,设定出更为稳健的交织周期识别门限;其次,分析出了数据矩阵中每一行与每一列累积量的对应关系,提出了一种快速交织周期遍历方法,使得矩阵构建次数大大减少;最后定义了聚合度概念,仅通过二重循环遍历即可完成交织深度与交织宽度的快速识别。仿真结果表明,该算法能够在低信噪比下实现卷积交织器参数的有效识别,同时相比于现有的方法,识别性能提升了1 dB到2 dB,且计算效率得到了明显的提高。In order to overcome the shortcomings of the existing algorithms for recognition of convolutional interleaver,such as high error probability and low recognition efficiency at low signal to noise ratio(SNR),the statistical characteristics of the constructed matrix are analyzed firstly.The functions of probability density for synchronous codes and the random bits are given,and the detection threshold of the synchronous codes based on the minimum error decision criterion is set.Secondly,a more robust threshold for recognition of interleaver period based on the three times standard deviation criterion is set.Then,the relationship between each row and column in the matrix is analyzed,and a fast method for traversing interleaver period is proposed,which could greatly reduce the times of constructing the matrix.Finally,the concept of polymerization degree is defined.Only through the double traversal,the recognition of the depth and width of interleaving could be completed,which greatly improves the efficiency.The simulation results show that the algorithm can effectively recognize the parameters of the convolutional interleaver at low SNR,the performance of the algorithm is improved by 1 dB to 2 dB compared with the existing methods,and the computation efficiency is improved significantly.

关 键 词：认知无线电 卷积交织器 同步码 聚合度 盲识别 

分 类 号：TN911.7[电子电信—通信与信息系统]