Title :
Least squares order-recursive lattice smoothers
Author :
Yuan, Jenq-Tay ; Stuller, John A.
Author_Institution :
Dept. of Electron. Eng., Fu Jen Catholic Univ., Taipei, Taiwan
fDate :
5/1/1995 12:00:00 AM
Abstract :
Conventional least squares order-recursive lattice (LSORL) filters use present and past data values to estimate the present value of a signal. This paper introduces LSORL smoothers which use past, present and future data for that purpose. Except for an overall delay needed for physical realization, LSORL smoothers can substantially outperform LSORL filters while retaining all the advantages of an order-recursive structure
Keywords :
adaptive filters; lattice filters; least squares approximations; recursive filters; smoothing methods; adaptive LS lattice filters; delay; future data; least squares order-recursive lattice; order-recursive structure; past data; present data; recursive lattice filters; recursive lattice smoothers; Delay estimation; Estimation error; Finite impulse response filter; Kalman filters; Lattices; Least squares approximation; Least squares methods; Mean square error methods; Nonlinear filters; Transversal filters;
Journal_Title :
Signal Processing, IEEE Transactions on
