Title :
A novel kurtosis driven variable step-size adaptive algorithm
Author :
Pazaitis, Dimitrios I. ; Constantinides, A.G.
Author_Institution :
IMEC, Leuven, Belgium
fDate :
3/1/1999 12:00:00 AM
Abstract :
A new variable step-size LMS filter is introduced. The time-varying step-size sequence is adjusted, utilizing the kurtosis of the estimation error, therefore reducing performance degradation due to the existence of strong noise. The convergence properties of the algorithm are analyzed, and an adaptive kurtosis estimator that takes into account noise statistics and optimally adapts itself is also presented. Simulation results confirm the algorithm´s improved performance and flexibility
Keywords :
Gaussian noise; adaptive filters; adaptive signal processing; convergence of numerical methods; filtering theory; higher order statistics; least mean squares methods; Gaussian noise; LMS filter; adaptive kurtosis estimator; adaptive signal processing; convergence properties; estimation error kurtosis; fourth-order cumulant; noise statistics; performance; simulation results; time-varying step-size sequence; variable step-size adaptive algorithm; Adaptive algorithm; Adaptive filters; Adaptive signal processing; Equations; Filtering algorithms; Least squares methods; Numerical stability; Signal processing; Signal processing algorithms; Transversal filters;
Journal_Title :
Signal Processing, IEEE Transactions on
