Journal of Applied Mathematics and Stochastic Analysis, 11:3 (1998), 289-300. FILTERING WITH A LIMITER (IMPROVED PERFORMANCE)

We consider a filtering problem for a Gaussian diffusion process observed via discrete-time samples corrupted by a non-Gaussian white noise. Com-bining the Goggin’s result [2] on weak convergence for conditional expecta-tion with diffusion approximation when a sampling step goes to zero we construct an asymptotic optimal filter. Our filter uses centered observa-tions passed through a limiter. Being asymptotically equivalent to a simi-lar filter without centering, it yields a better filtering accuracy in a pre-limit case