Blind Separation of Digital Signal Sources in Noise Circumstance
During blind separation, noise exists and effects the work. This paper presents novel techniques for blind separation of instantaneously mixed digital sources in noise circumstance, which is based on characteristics of digital signals. The blind separation and denoising algorithms include two steps. First, one of adaptive blind separation algorithms in existence is used to separate sources, but there still exists noise in the separating signals, and then, the second step is adopted to denoise according to the characteristics of digital signals. In the last simulations, the good performance is illustrated and the algorithm is very excellent.
KeywordsDigital Signal Independent Component Analysis Blind Source Separation Separation Matrix Denoising Algorithm
Unable to display preview. Download preview PDF.
- 2.Stone, J.: Blind source separation using temporal predictability. Neural Computation (13), 1559–1574 (2001)Google Scholar
- 6.Amari, S., Cichocki, A., Yang, H.: A new learning algorithm for blind signal separation. In: Touretzky, D.S., Hasselmo, M.E. (eds.) Advances in Neural Information Processing Systems, vol. 8, pp. 757–763 (1996)Google Scholar
- 8.Pham, D.T.: Mutual information approach to blind separation of stationary sources. IEEE Trans. Inform. Theory 48, 1–12 (2002)Google Scholar
- 11.Pajunen, P.: Blind separation of binary sources with less sensors than sources. In: Processings of 1997 International Conference on neural Networks, Houston, Texas, USA (1997)Google Scholar
- 12.Belouchrani, A., Cichocki, A.: An adaptive algorithm for the separation of finite alphabet signals. In: Proceedings of workshop on signal processing and applications 2000, Brisbane, Australia (December 2000)Google Scholar
- 13.Li, Y., Cichocki, A., Zhang, L.: Blind separation and extraction of binary sources. Communication and Computer Sciences E86-A(3), 580–590 (2003)Google Scholar
- 14.Li, Y., Zhang, L.: New algorithm for blind enhancement of digital images. Control Theory and Applications 20(4), 525–528 (2003)Google Scholar