Bias-compensated normalised LMS algorithm with noisy input

• Published in: Electronics letters Vol. 49; no. 8; pp. 538 - 539
• Main Authors: Kang, B, Yoo, J, Park, P
• Format: Journal Article
• Language: English
• Published: Stevenage The Institution of Engineering and Technology 11.04.2013; Institution of Engineering and Technology
• Subjects: adaptive filters; Applied sciences; bias compensated normalised LMS algorithm; Computer science; control theory; systems; condition checking constraint; Control theory. Systems; Exact sciences and technology; Information and communication; input noise; least mean square algorithm; least mean squares methods; Linear inference, regression; Mathematics; Modelling and identification; noisy input; Numerical analysis; Numerical analysis. Scientific computation; Numerical approximation; parameter estimation; Probability and statistics; Sciences and techniques of general use; statistical properties; Statistics; least mean squares methods; adaptive filters; input noise; noisy input; parameter estimation; bias compensated normalised LMS algorithm; statistical properties; condition checking constraint; least mean square algorithm; Parameter estimation; Statistical analysis; Probabilistic approach; Constraint; Least squares method; Bias; Cost function; Adaptive filter; System identification; Modeling
• ISSN: 0013-5194; 1350-911X; 1350-911X
• DOI: 10.1049/el.2013.0246

A new bias-compensated normalised least mean square (NLMS) algorithm for parameter estimation with a noisy input is proposed. The algorithm is obtained from an approximated cost function based on the statistical properties of the input noise and involves a condition checking constraint to decide whether the weight coefficient vector must be updated. Simulation results show that the proposed algorithm is more robust and accurate than the conventional method.