In this paper, it is presented an algorithm to construct a cumulant matrix that has a well-separated extremal eigenvalue. The corresponding eigenvector is well-conditioned and could be used to develop robust algorithms for blind source extraction. Simulations demonstrate the effectiveness of the proposed approach. © Springer-Verlag 2004.
CITATION STYLE
Martín-Clemente, R., Hornillo-Mellado, S., Puntonet, C. G., & Acha, J. I. (2004). New eigensystem-based method for blind source separation. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3195, 33–40. https://doi.org/10.1007/978-3-540-30110-3_5
Mendeley helps you to discover research relevant for your work.