An explicit expression for the discrete-time fixed-lag smoothing equation
β Scribed by A. Klein
- Publisher
- Elsevier Science
- Year
- 1982
- Tongue
- English
- Weight
- 155 KB
- Volume
- 8
- Category
- Article
- ISSN
- 0377-0427
No coin nor oath required. For personal study only.
β¦ Synopsis
Kalman filtering is extensively used in estimation techniques. The problem is that one tries to estimate a state based on measures perturbated by noise. However in many applications some time lag can be allowed; this means that this estimate which is called smoothed can be made using a greater amount of information than in "on line" filtering and the estimation error can be reduced. In this paper the smoothing equation is investigated by writing the smoother as t+k ~lt/t+k)= --~t HsY(S) andlett0=-°° (k>0) s o t+k Where I;
Hszt-S is a non-causal transfer function. s= t o
We also notice that H s is time-dependent. In our study we focus on H s and its variation with k. (The results are illustrated with an example).
π SIMILAR VOLUMES
The modified Bryson-Frazier fixed interval smoothing algorithm [6], is an addendem to the Kalman filterβ’ This algorithm when applied to the problem of fixedlag smoothing is computationally more efficient than the algorithms recently reported in refs. [1][2][3]. Features of the algorithm are ease of
A family of new hybrid explicit four-step tenth algebraic order methods ## Ε½ . with phase lag of order 18 2 26 is developed for efficient computations of the Schrodinger Γ«quation. Based on these new methods, a new embedded variable-step method is obtained. Numerical results produced for the nume