Monotonic convergence of fixed-point algorithms for ICA

• Published in: IEEE transactions on neural networks Vol. 14; no. 4; pp. 943 - 949
• Main Authors: Regalia, P.A., Kofidis, E.
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.07.2003; Institute of Electrical and Electronics Engineers
• Subjects: Algorithms; Appeals; Background noise; Convergence; Engineering Sciences; Fixed points (mathematics); Image restoration; Independent component analysis; Information processing; Initial conditions; Mathematical models; Neural networks; Probability density function; Random variables; Sensors; Signal analysis; Signal and Image processing; Signal restoration; Vectors; Non-Gaussian signals; Fixed-point algorithms; Monotonic convergence; Independent component analysis (ICA)

We re-examine a fixed-point algorithm proposed by Hyvarinen for independent component analysis, wherein local convergence is proved subject to an ideal signal model using a square invertible mixing matrix. Here, we derive step-size bounds which ensure monotonic convergence to a local extremum for an...