A variable leaky LMS adaptive algorithm

• Published in: 2004 38th Asilomar Conference on Signals, Systems and Computers Vol. 1; pp. 125 - 128 Vol.1
• Main Authors: Kamenetsky, M., Widrow, B.
• Format: Conference Proceeding
• Language: English
• Published: Piscataway NJ IEEE 2004
• Subjects: Adaptive algorithm; Adaptive signal processing; Applied sciences; Cost function; Eigenvalues and eigenfunctions; Exact sciences and technology; Information, signal and communications theory; Iterative algorithms; Least squares approximation; Mathematical methods; Process control; Random processes; Signal processing algorithms; Telecommunications and information theory; Vectors; Performance evaluation; Adaptive algorithm; Simulation; Heuristic method; Greedy algorithm; Eigenvalue; Adaptive signal processing; Least mean squares methods
• ISBN: 0780386221; 9780780386228
• DOI: 10.1109/ACSSC.2004.1399103

The LMS algorithm has found wide application in many areas of adaptive signal processing and control. We introduce a variable leaky LMS algorithm, designed to overcome the slow convergence of standard LMS in cases of high input eigenvalue spread. The algorithm uses a greedy punish/reward heuristic together with a quantized leak adjustment function to vary the leak. Simulation results confirm that the new algorithm can significantly outperform standard LMS when the input eigenvalue spread is high.