【作者】 胡学友；

【导师】 高隽；

【作者基本信息】 合肥工业大学 ， 信号与信息处理， 2003， 硕士

【摘要】 盲信号分离技术是信号处理中用于阵列处理和数据分析的一种新的技术，在源信号和混合过程未知情况下，从混合信号中分离出源信号。盲信号分离技术可广泛应用于无线通讯、医学信号处理、图像处理以及雷达信号处理等多个应用领域，已成为信号处理领域的一个研究热点。本论文对盲信号分离技术理论及其应用问题进行了研究，主要工作如下： (1)搜集、整理、总结了国内外盲分离方面最新的研究成果和进展，介绍盲分离基本概念和理论，并着重讨论了几种比较常用的盲分离学习算法及算法特点。 (2)在最小互信息盲分离算法的基础上，提出了基于递归网络形式的改进型算法，从理论上给予了证明，并通过计算机仿真模拟证实了该方法的可行性。 (3)论文对盲信号分离技术在模式识别以及信号处理方面的应用研究进行了探讨。交通标志识别的实验及该识别系统的实现表明该方法在模式识别中可以取得很好的效果。基于盲分离的数字水印研究实验验证了该方法在信号处理方面的有效性。更多还原

【Abstract】 Blind signal separation is a new technique of signal processing aimed at retrieving a system’s unknown information from its output only, which can be used in array processing and data analysis. There are many potential application, especially in wireless communication, medical signal processing, image processing and radar signal processing. In recent years, blind signal separation has been an attractive trend in the area of academia. In this thesis, the theory and applications of blind source separation has been studied. The thesis consists of following parts:1. The paper reviews systematically the present research situation of blind source separation in the world. The basic principles and concepts of blind source separation and some algorithms are introduced.2. A minimum mutual information algorithm based recurrent neural networks are given. The main principle, derivation has been presented. Computation simulation experiments are also given to illustrate the performances of the algorithm.3. The traffic sign recognition and digital watermarking based blind source separation technique has been demonstrated on this paper. It indicated that blind source separation have a good result on the realization of this recognition system and its application to digital water marking. The experiment on research of digital watermarking based blind source separation also validated this method’s application on the signal-process domain.更多还原