Recursive filtering for two-dimensional random fields (Corresp.)
暂无分享,去创建一个
A class of recursive filtering problems for random fields with a two-dimensional parameter is considered. After a brief introduction of two-parameter stochastic calculus, a class of Markovian random fields generated by stochastic integral equations is defined and considered. It is then shown that the problem of estimating such a Markovian field in additive white Gaussian noise can be reduced to a recursive formalism. If the random field is itself Gaussian, the recursive formalism reduces to a finite set of stochastic integral equations involving the conditional mean and covariance.
[1] P. Levy. A Special Problem of Brownian Motion, and a General Theory of Gaussian Random Functions , 1956 .
[2] Eugene Wong. A likelihood ratio formula for two-dimensional random fields , 1974, IEEE Trans. Inf. Theory.
[3] J. P. McKean. Brownian Motion with a Several-Dimensional Time , 1963 .
[4] M. Zakai. On the optimal filtering of diffusion processes , 1969 .
[5] Moshe Zakai,et al. Martingales and stochastic integrals for processes with a multi-dimensional parameter , 1974 .