以獨(dú)立分量分析為主要對(duì)象, 描述了盲信號(hào)源分離技術(shù)的基本模型，介紹了盲分離的主要方法和數(shù)學(xué)原理, 分析了盲信號(hào)源的可辨識(shí)性。提出基于神經(jīng)網(wǎng)絡(luò)無監(jiān)督學(xué)習(xí)的盲分離方法，并改進(jìn)了分離效果評(píng)判指標(biāo)。在生物信息處理的背景下將人工神經(jīng)網(wǎng)絡(luò)和信息理論相結(jié)合，解決了盲信號(hào)源分離，自適應(yīng)地求得分離矩陣，且可以同時(shí)分離具有正峭度和負(fù)峭度的信號(hào)源，對(duì)盲信號(hào)源分離的研究有極大的促進(jìn)作用。
關(guān) 鍵 詞 盲信號(hào)源分離; 神經(jīng)網(wǎng)絡(luò); 無監(jiān)督學(xué)習(xí); 獨(dú)立分量分析板

This paper focuses on the independent component analysis presenting a review on the basic model, the main method, the mathematical principle of blind source separation，analyzing the possibility of separation. The paper puts forward the method that based on the neural network unsupervised learning, also, improves the index on separation effects. Under the biology information processing background, combining artificial neural network with information theory to resolve this kind of problem can get the separation matrix adaptively by itself. It can separate mixtures which have both positive and negative kurtosis, and promotes the research of blind source separation greatly.