Title :
Application of ICA in harmonic retrieval in chaotic noise
Author :
Peng, Lu ; Yue, Li ; Chao, Li
Author_Institution :
Inst. of Commun. Eng., Jilin Univ., Changchun, China
fDate :
31 Aug.-4 Sept. 2004
Abstract :
In this paper, we apply the independent component analysis (ICA) method to the retrieval of harmonic signal from chaotic noise. Our method utilizes the characteristic of ICA that can separate the mixed signals, which are statistic independent, although the chaotic noise was generated by some deterministic system in contrast to the previous methods that constrain the time or amplitude scale of signals, our method can separate the signals from the chaotic noise without such constraint and the knowledge of the deterministic system that generate the chaotic noise. The simulation study has shown that ICA method can extract very weak signal under some well known chaotic background.
Keywords :
chaos; harmonics; independent component analysis; signal denoising; source separation; ICA; chaotic noise; harmonic retrieval; independent component analysis; signal amplitude scale; signal extraction; signal separation; statistic independent; Background noise; Chaos; Electrocardiography; Electroencephalography; Independent component analysis; Noise generators; Noise reduction; Signal generators; Statistics; Working environment noise;
Conference_Titel :
Signal Processing, 2004. Proceedings. ICSP '04. 2004 7th International Conference on
Print_ISBN :
0-7803-8406-7
DOI :
10.1109/ICOSP.2004.1442301