EE 571

Course Name: 
Bayesian Signal Processing
Course Credit: 
Course Description: 
Bayesian theory and Bayesian estimation. Deterministic, probabilistic and sequential inference techniques. Batch and sequential Bayesian estimation. Sampling methods and simulation-based Bayesian methods. State-space models for Bayesian processing. Classical approach to Bayesian estimation, linear optimal filters: Kalman filters and extended Kalman filters. Unscented transformation, unscented Kalman filter and Gaussian sum filter-based Bayesian estimation. Particle filters, importance sampling, selection of importance function, resampling. Particle filter-based Bayesian estimation. Bayesian joint state/parameter estimation. Cramer-Rao bounds for particle filters.