Using Combinative Proposal Distribution in Particle Filtering Algorithm

Particle filtering has gained special attention in various fields due to its capacity to handle nonlinear/non-Gaussian dynamic problems. One crucial issue in particle filtering is the selection of the proposal distribution that generates particles. In this paper, the particle filtering algorithm is improved by using more than one proposal distributions. Firstly, divide-conquer strategy is taken, and the particles used are divided into several parts. Then, they are drawn from different proposal distributions. Users can adjust the number of the particles drawn from specific proposal distributions according to their special requirements. We provide simulation results that show the efficiency and performance.