A Powerful Numerical Technique Solving Zakai Equation for Nonlinear Filtering

In this paper we have developed a simple but powerful numerical method forthe approximation of the unnormalized conditional (probability) density offiltered diffusion process which satisfies Zakai equation and solves thenonlinear filtering problem. Using Galerkin technique the solution of Zakaiequation is approximated by means of a sequence of nonstandard basisfunctions given by a parameterized family of Gaussian densities. The methodis then illustrated by some examples.

[1]  Zeev Schuss,et al.  Theory and Applications of Stochastic Differential Equations , 1980 .

[2]  W. Willman Edgeworth expansions in state perturbation estimation , 1981 .

[3]  J. Douglas Faires,et al.  Numerical Analysis , 1981 .

[4]  M. Zakai On the optimal filtering of diffusion processes , 1969 .

[5]  N. U. Ahmed,et al.  Elements of Finite-dimensional Systems and Control Theory , 1988 .

[6]  N. U. Ahmed,et al.  Parameter identification for partially observed diffusions , 1992 .

[7]  Alain Bensoussan,et al.  Maximum principle and dynamic programming approaches of the optimal control of partially observed diffusions , 1983 .

[8]  Kok Lay Teo,et al.  Optimal control of distributed parameter systems , 1981 .

[9]  S. Marcus,et al.  On Lie algebras and finite dimensional filtering , 1982 .

[10]  E. Pardouxt,et al.  Stochastic partial differential equations and filtering of diffusion processes , 1980 .

[11]  H. Kushner Nonlinear filtering: The exact dynamical equations satisfied by the conditional mode , 1967, IEEE Transactions on Automatic Control.

[12]  E. Wong,et al.  On the Convergence of Ordinary Integrals to Stochastic Integrals , 1965 .

[13]  V. Benes Exact finite-dimensional filters for certain diffusions with nonlinear drift , 1981 .

[14]  A. Friedman Partial Differential Equations of Parabolic Type , 1983 .

[15]  R. Bucy Nonlinear filtering theory , 1965 .

[16]  R. Elliott,et al.  Approximations to solutions of the zakai filtering equation , 1989 .