一种同频混合信号伪随机序列盲估计方法探究.docVIP

一种同频混合信号伪随机序列盲估计方法探究 　　摘 要： 通常采用独立分量分析法（ICA）采集同频混合信号时存在盲分离随机性问题，不能分离出同频混合信号伪随机序列，无法对信号进行准确检测。为解决该问题，提出融合独立分量分析法以及Massye算法的同频混合信号伪随机序列盲估计方法。先采集同频混合信号，再通过PCA方法对同频混合信号进行白化预处理，对同频混合信号的协方差矩阵的特征值进行分解，确保信号间相互独立，为后续ICA方法进行数据分割提供基础。采用基于峰度的固定点ICA算法对白化处理后的同频混合数据进行划分，融合ICA和Massye算法，对同频混合信号的伪随机序列进行盲估计。实验结果说明，该方法可以获取准确的同频混合信号伪随机序列，具有较强的信号分离性能 关键词： 同频混合信号； 伪随机序列； 盲估计； Massye算法 中图分类号： TN911?34 文献标识码： A 文章编号： 1004?373X（2017）09?0010?04 Abstract： The independent component analysis （ICA） method is usually used to collect the co?frequency mixed signal， but it has the problem of blind separation randomness， and it can′t separate the pseudorandom sequence of the co?frequency mixed signal， nor can it detect the signal accurately. In order to solve the above problems， a co?frequency mixed signal′s pseudorandom sequence blind estimation method based on ICA method and Massye algorithm is proposed. The co?frequency mixed signal is acquired， and performed with the whitening pretreatment by means of PCA method. The eigenvalue of the covariance matrix of the co?frequency mixed signal is decomposed to ensure the mutual independence among the signals， and provide the basis for data segmentation with the follow?up ICA method. The fixed?point ICA algorithm based on kurtosis is used to divide the co?frequency mixed data after whitening treatment. The ICA algorithm and Massye algorithm are fused to perform the blind estimation to the pseudorandom sequence of the co?frequency mixed signal. The experimental results show that the proposed method can obtain the accurate pseudorandom sequence of the co?frequency mixed signal， and has strong signal separation performance. Keywords： co?frequency mixed signal； pseudorandom sequence； blind estimation； Massye algorithm 0 引 言 扩频通信是当前通信领域中的重要通信手段，其中的直扩频系统可将待发送信息通过伪随机序列扩展到更宽的频带中[1]。干扰信号和伪随机序列间相互独立，极大地降低了信号频带中有干扰信号的功率，可增强总体扩频通信系统的抗噪性。直扩频技术广泛应用在通信抗干扰、导航以及保密通信等领域[2]。因此，对直扩频系统中伪随机序列的检测方法进行研究，实现信息的准确侦察，具有重要的应用价值 1 融合独立分量分析法和Massye算法的伪随 机序列盲估计 本文