Title :
On local convergence of a class of blind separation algorithms
Author :
Lindgren, Ulf ; Wigren, Torbjöm ; Broman, Holger
Author_Institution :
Dept. of Appl. Electron, Chalmers Univ. of Technol., Goteborg, Sweden
fDate :
12/1/1995 12:00:00 AM
Abstract :
A class of recursive stochastic gradient algorithms for blind separation of dynamically mixed independent source signals are analyzed. The studied methods utilize correlations and high-order moments in order to enforce statistical independence of the separated signals. The local convergence properties of the schemes are investigated, and it is demonstrated that local convergence is tied to positive realness of certain mixing transfer functions
Keywords :
convergence of numerical methods; correlation methods; higher order statistics; recursive estimation; signal processing; stochastic processes; transfer functions; blind separation algorithms; correlations; dynamically mixed independent source signals; high-order moments; local convergence properties; mixing transfer functions; recursive stochastic gradient algorithms; separated signals; statistical independence; Algorithm design and analysis; Convergence; Crosstalk; Microphones; Nonlinear filters; Signal analysis; Signal processing; Signal processing algorithms; Stochastic processes; Transfer functions;
Journal_Title :
Signal Processing, IEEE Transactions on
