Title :
A simplified informax approach for blind signal separation
Author :
Xi, Jiangtao ; Chicharo, Joe F. ; Reilly, James P.
Author_Institution :
Sch. of Electr., Comput. & Telecommun. Eng., Wollongong Univ., NSW, Australia
Abstract :
This paper proposes a simplified training algorithm for blind separation of speech signals based on the principle of entropy maximization. Compared to the algorithms Amari and Cichocki (see Proc. of IEEE, vol.86, no.10, p.2026-48, 1998), the proposed approach leads to lower computational burden while yielding comparable performance
Keywords :
computational complexity; maximum entropy methods; speech processing; blind signal separation; computational burden; entropy maximization; information maximization; performance; simplified informax approach; simplified training algorithm; speech signals separation; Arithmetic; Australia; Blind source separation; Computer networks; Entropy; Jacobian matrices; Noise measurement; Signal processing algorithms; Speech; Telecommunication computing;
Conference_Titel :
Signal Processing and Its Applications, 1999. ISSPA '99. Proceedings of the Fifth International Symposium on
Conference_Location :
Brisbane, Qld.
Print_ISBN :
1-86435-451-8
DOI :
10.1109/ISSPA.1999.818108
