Title :
New method of designing adaptive nonlinear filters
Author :
Rao, K. Deergha ; Reddy, D.C.
Author_Institution :
Dept. of Electron. & Commun. Eng., MVSR Coll. of Eng., Osmania Univ., Hyderabad, India
fDate :
10/1/1991 12:00:00 AM
Abstract :
A new method of designing nonlinear filters is proposed. It is shown to be general enough to be applicable to the design of higher order Volterra filters. LMS and fast least squares adaptive algorithms for updating the linear quadratic and cubic weights of a third order nonlinear filter are suggested. Simulation studies are carried out in respect of a third order Volterra filter using the proposed method of design and suggested updating algorithms. The fast least squares implementation is shown to converge to the unknown system parameters considerably faster than the LMS
Keywords :
adaptive filters; digital filters; least squares approximations; LMS algorithm; Volterra filters; adaptive nonlinear filters; cubic weights; design; fast least squares adaptive algorithms; quadratic weights; simulation; system parameters; third order nonlinear filter; weights updating;
Journal_Title :
Radar and Signal Processing, IEE Proceedings F
