Title :
A penalty function method for stable IIR adaptive filtering
Author_Institution :
Nanjing Marine Radar Inst., Jiangsu, China
Abstract :
A penalty function method is introduced to stabilize the parallel IIR (infinite impulse response) adaptive filters. In order to obtain an optimal step size for the adaptation, the input data are processed in a block processing form. A recursive formula is derived for computing the optimal step size, and a conjugate gradient direction is used to obtain a faster convergence. A simplified online algorithm is obtained which is a modification of the recursive least-mean-square algorithm
Keywords :
adaptive filters; conjugate gradient methods; convergence; digital filters; filtering and prediction theory; parallel processing; stability; block processing; conjugate gradient direction; convergence; infinite impulse response; least-mean-square algorithm; online algorithm; optimal step size; penalty function method; recursive LMS algorithm; recursive formula; stable IIR adaptive filtering; Acceleration; Adaptive algorithm; Adaptive filters; Constraint optimization; Convergence; Finite impulse response filter; IIR filters; Monitoring; Radar; Stability;
Conference_Titel :
Circuits and Systems, 1991., IEEE International Sympoisum on
Print_ISBN :
0-7803-0050-5
DOI :
10.1109/ISCAS.1991.176363