Title :
Derivation of a stable recursive form of FIR filter/FIR smoother
Author :
Kim, Jae-Hun ; Lyou, Joon
Author_Institution :
Agency for Defense Dev., Taejeon, South Korea
Abstract :
Kwon et al. (1989) derived a recursive adjoint form of FIR filter and smoother for discrete-time state-space models. Even though it is algebraically identical to the FIR filter and the FIR smoother, unfortunately it is pointed out to be numerically unstable. This makes the recursive estimator (RFIR) seldomly used in real applications. So we have devised a stable recursive estimator (SRFIR) which can supplement the RFIR so that it could be stabilized, and show that the SRFIR has unbiased and minimum variance error
Keywords :
FIR filters; recursive estimation; smoothing methods; stability; FIR filter; FIR smoother; discrete-time state-space models; minimum variance error; stable recursive estimator; unbiased error; Covariance matrix; Equations; Filtering; Finite impulse response filter; Frequency; Nonlinear filters; Recursive estimation; Robust stability; Signal design; Signal processing;
Conference_Titel :
SICE 2001. Proceedings of the 40th SICE Annual Conference. International Session Papers
Conference_Location :
Nagoya
Print_ISBN :
0-7803-7306-5
DOI :
10.1109/SICE.2001.977846
