Robust transcale decentralised estimation fusion for multisensor systems based on wavelet packet decomposition

In this study, an effective wavelet packet decomposition (WPD)-based robust transcale decentralised estimation fusion algorithm is proposed for multisensor systems. The standard state-space models of wavelet packet coefficients at the coarsest scale decomposition layer are established by Haar WPD. Based on the above models, the decentralised H_∞ filtering fusion formulae is employed to obtain the fusion estimates at the coarsest scale decomposition layer, and these estimates are then used to reconstruct the estimates at the finest scale by an inverse WPD transform. At last, the feasibility analysis of the proposed algorithm is given. Two simulation examples are presented to illustrate the effectiveness of the proposed algorithm.