Title :
Two rapidly convergent algorithms for signal separation
Author :
Fahmy, M.F. ; El-Raheem, G.M.A. ; El-Sallam, A.A.
Author_Institution :
Dept. of Electr. Eng., Assiut Univ., Egypt
Abstract :
This paper presents two methods for signal separation. In these methods, the parameters of the decoupling system are adaptively adjusted to minimize either the output cross correlation or the output cross cumulant functions, respectively. Alternatively, these parameters can also be determined as the least squares solution of a set of equations describing either the cross correlation or cross cumulant functions when evaluated for different lags. Both methods are characterized by their rapid convergence property. Illustrative examples are also given
Keywords :
correlation theory; higher order statistics; least squares approximations; signal resolution; decoupling system; least squares solution; output cross correlation; output cross cumulant functions; rapidly convergent algorithms; signal separation; Deconvolution; Decorrelation; Differential equations; Eigenvalues and eigenfunctions; Finite impulse response filter; Higher order statistics; Least squares methods; Maximum likelihood detection; Source separation; Vectors;
Conference_Titel :
Circuits and Systems, 1999. ISCAS '99. Proceedings of the 1999 IEEE International Symposium on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-5471-0
DOI :
10.1109/ISCAS.1999.778773
