论文部分内容阅读
针对可变扩频长度多速率直接序列码分多址(direct sequence/code division multiple access,DS/CDMA)信号扩频序列盲估计问题,提出了一种基于盲源分离的并行估计方法,该方法首先利用二次功率谱的方法估计出信号不同速率的扩频码周期,然后利用多速率DS/CDMA信号周期特性将信号依据周期并行地建模为主用户和干扰用户(即将同一周期下的用户视为主用户,其余的用户和噪声一律视为独立的单个干扰用户)的模型,最后针对同步和异步两种情况,分别以单倍周期和两倍周期长度对数据进行分割,生成数据矩阵,并结合稳健的矩阵联合近似对角化(JADE)算法对扩频序列进行并行的盲估计.计算机仿真结果表明方法在较低信噪比下有效.
Aiming at the blind estimation problem of DS / CDMA sequence spreading in DSSS CDMA system, a parallel estimation method based on blind source separation is proposed. This method Firstly, the second power spectrum method is used to estimate the spreading code periods of different rates and then the signal is modeled as the primary user and the interfering user according to the periodicity of the multi-rate DS / CDMA signal (ie the users in the same cycle As the main user, the rest of the users and the noise will be regarded as an independent single interference users) model. Finally, for both synchronous and asynchronous cases, the data is divided into single cycle and double cycle length to generate the data matrix, Combined with the robust JADE algorithm, a parallel blind estimation of the spreading sequence is carried out.Computer simulation results show that the proposed method is effective at lower signal-to-noise ratio.