Multiscale distributed fusion estimation in multi-sensor single model dynamic systems

Multiscale methods in signal and image processing have experienced a surge of activity in recent years, inspired primarily the emerging theory of multiscale representation of signals and wavelet transforms. One of the lines of investigation that has been sparked these developments is statistical signal processing based on wavelet transforms and multiscale re pre- sentstion. By use of the multiscale analysis, this paper combines the model-based dynamic system analysis method with the multiscale transforrnation method based on the statistical characteristics. Two new algorithms of multiscale fusion estimation based on multi-sensor dynamic system and single sensor dynamic system are proposed and the effectiveness of multiscale distributed fusion algorithm is illustrated.