|
| << vorige | volgende >> |
Tijdschrift beschrijving
Alle jaargangen van het bijbehorende tijdschrift
Alle afleveringen van het bijbehorende jaargang
Alle artikelen van de bijbehorende aflevering
Details van artikel 2 van 7 gevonden artikelen
| Titel: |
A new numerical technique for solving the nonlinear filtering problem |
| Auteur: |
Radaideh, S. M. Al-Rousan, M. |
| Verschenen in: |
International journal of electronics |
| Paginering: | Jaargang 89 (2002) nr. 7 pagina's 561-573 |
| Jaar: |
2002-07 |
| Inhoud: |
This paper describes the development of computable and accurate approximate expressions for the conditional law of a diffusion process observed in continuous time which satisfy the Zakai equation and solve the nonlinear filtering problem. The proposed numerical algorithm makes it possible to shift offline the most time-consuming operations. In comparison with the extended Kalman filter (EKF) it is more robust in the case of high signal-to-noise ratio (i.e. low measurement noise intensity). The estimation accuracy of the proposed algorithm increases when the process noise intensity increases. It is readily implementable on a digital computer and gives a sample pathwise solution. Using the Galerkin technique, the solution of the Zakai equation is approximated by means of a sequence of nonstandard basis functions given by a parametrized family of trigonometric Fourier series. For illustration, some numerical examples are solved using the proposed algorithm and the EKF. |
| Uitgever: |
Taylor & Francis |
| Bronbestand: |
Elektronische Wetenschappelijke Tijdschriften |
Details van artikel 2 van 7 gevonden artikelen
| << vorige | volgende >> |