强噪声背景下基于子空间的盲信号提取
Blind Signal Extraction Based on Subspace over High Noise Source Background
-
摘要: 低信噪比下的去噪一直是一个难题，最近Emir等人提出了Independent Component Analysis(ICA)去噪方法，该方法在光学功能成像中得到了成功应用。但研究发现在极低信噪比下，由于观测数据的样本协方差矩阵具有奇异性，这使得ICA去噪算法中的白化处理步骤无法进行。为解决这一问题，本文利用子空间的概念，在ICA去噪方法的基础上提出了一种新的基于子空间的ICA(ICA based on signal Subspace, SICA)去噪方法。仿真表明该方法能在极低信噪比下有效去噪，同时与传统的滤波去噪相比， SICA去噪方法在去噪的同时还能够成功地将频域重叠的信号正确分离。Abstract: It is a difficult problem to denoise in the low SNR, recently, Emir et al present a novel ICA denoising method, this method has been successfully applied to the function optical imaging. But in the very low SNR circumstance, because of the covariance matrix of the observed signals being singularity, the ICA denoising method can not be used. In order to resolve this problem, a new SICA denoising method based on the signal subspace is presented in this paper. The simulations show that compared to the ICA denoising method and the traditional filtering denoising methods, the method can not only get rid of the noise, but can successfully separation the signals.
-
Emir E, Akgul B, Akin A, et al.. Wavelet denoising vs ICA denoising for functional optical imaging[A]. Proceedings of the 1st International IEEE EMBS Conference on Neural Engineering[C]. Capril Island, Italy, 2003: 384-387.[2]Hyvinen A, Karhunen J, Oja E. Independent Component Analysis[M]. New York, Wiley, 2001: Chapter 6-8.[3]Bell A, Sejnowski T. An information-maximization approach to blind separation and blind deconvolution[J].Neural Computation.1995, 7(6):1129-1159[4]Hyvinen A, Oja E. A fast fixed-point algorithm for independent component analysis[J].Neural Computation.1997, 9(7):1483-[5]Zibulevsky M, Pearlmutter A. Blind source separation by sparse decomposition[J].Neural Computation.2001, 13(4):863-882 -
计量
- 文章访问数: 2598
- HTML全文浏览量: 107
- PDF下载量: 1287
- 被引次数: 0
下载:
