An alternative approach to nonlinear filtering |
| |
Authors: | Izidor Gertner |
| |
Affiliation: | Technion—Israel Institute of Technology, Haifa, Israel |
| |
Abstract: | The structure of a nonlinear filter with observation process having continuous and discontinuous components is considered. The approach is based on the so-called “Bayes” formula for conditional expectations. “Fubini” type theorems for stochastic integrals are given and used to obtain the representations of an optimal estimate and of the conditional likelihood ratio. A linear unnormalized filtering equation for controlled system process is derived. |
| |
Keywords: | Discontinuous processes differentiation formula absolute continuity unnormalized recursive filter normalized recursive filter |
本文献已被 ScienceDirect 等数据库收录! |