A New Multi-Scale Estimation Scheme for Dynamic System
Zhou Funa, Tianhao Tang, Wen Chenglin · 2006
Hybrid wavelet-Kalman filter method is an efficient multi-scale estimation method for dynamic system. Researches on multi-scale data fusion have become a hot topic in data fusion field. However, limited by the constraint that signal to implement wavelet transform must have the length of 2q, multi-scale estimation problem with non-2nsampled observation data still hasn't been efficiently solved, which is an obstacle of multi-scale data fusion. Based on present multi-scale method, we develop a new multi-scale estimation scheme aiming to design the stacked observation model for non-2nsampled observation by analyzing the possible observation structure of non-2nsampled sensor.