Ensemble Particle Filter with Posterior Gaussian Resampling
X. Xiong, I. M. Navon
An ensemble particle lter(EnPF) was recently developed as a fully nonlinear filter of Bayesian conditional probability estimation, along with the well known ensemble Kalman filter(EnKF). A Gaussian resampling method is proposed to generate the posterior analysis ensemble in an effective and efficient way. The Lorenz model is used to test the proposed method. With the posterior Gaussian resampling method the EnPF can approximate more accurately the Bayesian analysis. Moreover, it is applicable to systems with typical multimodal behavior, provided that certain prior knowledge is available about the general structure of posterior probability distribution. A simple scenario is considered to illustrate this point based on the Lorenz model attractors. The present work demonstrates that the proposed EnPF possesses good stability and accuracy and is potentially applicable to large-scale data assimilation problems.