异步LC-DS-CDMA信号的盲解扩

针对低信噪比下异步长码直接序列码分多址(direct sequence-code division multiple access, DS-CDMA)信号伪码序列难以估计的问题,提出一种基于嵌套迭代最小二乘投影和库搜索算法相结合的盲解扩方法。该方法先将接收到的异步长码DS-CDMA信号建模成含有缺失数据的短码DS-CDMA信号,并使用重叠窗对信号进行分段构造观测数据矩阵。然后,使用最大似然理论对其进行数学建模,并用嵌套迭代最小二乘投影算法实现各用户伪码序列片段的粗提取。最后,将提取出来的伪码序列在有限长的生成多项式库中进行匹配搜索,从而得到该序列的生成多项式。仿真实验及理论分析验证了该方法对异步长码DS-CDMA信号伪码序列估计的有效性。

Blind dispreading of asynchronous LC-DS-CDMA signals

For the problem of the blind estimation of asynchronous long code direct sequence-code division multiple access (DS-CDMA) signal pseudo-code sequences under low signal to noise ratio, a blind dispreading method based on nested iterative least squares projection and the library search algorithm is proposed. Firstly, the asynchronous long code DS-CDMA signal is virtualized into a short code DS-CDMA signal containing missing data, and on this basis, the signal is segmented to construct the observation data matrix with overlapping window. Secondly, the mathematical analysis is carried out based on the theory of maximum likelihood estimation, and the spread-spectrum sequence is estimated by the nested iterative least squares projection. Finally, the spread-spectrum sequence is searched in a limited generator polynomial library to estimate the primitive polynomial. Theoretical analysis and simulation results verify the effectiveness of the proposed method.