利用FastICA和三阶相关函数的多天线辅助NPLC-DS-CDMA扩频码盲估计  

Blind Estimation of Spreading Codes for Multi-antenna NPLC-DS-CDMA Signals Based on FastICA and Third Order Correlation Function

作　　者：尹辉 赵知劲[1] YIN Hui ,ZHAO Zhi-jin(School of Communication Engineering, Hangzhou Dianzi University, Hangzhou, Zhejiang 310018, Chin)

机构地区：[1]杭州电子科技大学通信工程学院,浙江杭州310018

出　　处：《信号处理》2018年第3期289-295,共7页Journal of Signal Processing

基　　金：国家自然科学基金(61571172)

摘　　要：针对非周期长码直扩CDMA信号(NPLC-DS-CDMA)扩频码盲估计问题,首先利用多天线接收信号并构建成矩阵形式;再对接收信号矩阵进行奇异值分解,利用Fast ICA方法估计每个用户直扩信号矩阵;然后将每个用户的直扩信号移位相乘去除信息码并计算其三阶相关函数(TCF),通过比较坐标点及其共轭系坐标点上的TCF估计值和理论值的距离排除大量伪峰值点,得到正确的TCF峰值点;最后利用矩阵斜消法得到扩频码本原多项式。仿真结果验证了本文算法对NPLC-DS-CDMA扩频码盲估计的有效性。Aiming at the problem of the spreading codes blind estimation of the non-periodic long code direct-spreading CD- MA signals （NPLC-DS-CDMA） , firstly, the signals were received by the multi-antennas and the signals were constructed to matrix. Secondly, the singular value decomposition was performed on the received signal matrix, and the direct sequence spread spectrum signal matrix of each user was estimated by FastICA algorithm from the extended singular value vector. Thirdly, the direct sequence spread spectrum signal of each user was shifted and multiplied in order to remove the information codes, and then the shift sequences were obtained. Fourthly, the third order correlation function （TCF） of the sequence was calculated, and the correct TCF peaks were obtained by comparing the distance between the TCF＇ s estimate and the theoretical value at the coordinate point and its conjugate coordinate points. Finally, the primitive polynomial of the spreading code was obtained by using the matrix oblique elimination method. The simulation results verify the effectiveness of the proposed algorithm for the blind estimation of NPLC-DS-CDMA spread spectrum code.

关 键 词：非周期长码扩频码分多址 扩频码估计 奇异值分解 FASTICA 三阶相关函数 

分 类 号：TN911[电子电信—通信与信息系统]