Convergence Results on methods for solving non-linear filtering problems

Thursday, February 21, 2013 - 10:30am - 12:00pm
Lind 305
Xiaoying (Maggie) Han (Auburn University)
Optimal filtering problems are important in signal processing and related fields. Most of the time the optimal filter does not admit a closed-form expression, and various numerical methods are developed to solve the filtering problem. Among these methods, particle filter is a widely used tool for numerical prediction of complex systems when observation data are available. In this talk I will first give a brief introduction on particle filtering and present error analysis on the numerical formulation of particle filter. Then I will introduce an implicit filtering method, along with its convergence results.