Title :
System identification based on bounded error constraints
Author :
Tsakalis, Kostas ; Deisher, Michael ; Spanias, Andreas
Author_Institution :
Dept. of Electr. Eng., Arizona State Univ., Tempe, AZ, USA
fDate :
12/1/1995 12:00:00 AM
Abstract :
Examines the performance of adaptive algorithms such as gradient, dead-zone, and bounding ellipsoid in the presence of bounded disturbances. The authors begin with a simple bursting example and continue by describing the algorithms and their properties, including analytical results on their asymptotic performance. Finally, a new algorithm is proposed, providing both MSE and maximum steady-state error performance guarantees
Keywords :
FIR filters; adaptive estimation; digital filters; error analysis; parameter estimation; signal processing; MSE; adaptive algorithms; asymptotic performance; bounded disturbances; bounded error constraints; bounding ellipsoid; bursting example; dead-zone algorithm; gradient algorithm; maximum steady-state error performance guarantees; system identification; Acoustic measurements; Acoustic noise; Adaptive systems; Additives; Echo cancellers; Noise cancellation; Noise measurement; Parameter estimation; Signal processing algorithms; System identification;
Journal_Title :
Signal Processing, IEEE Transactions on