基于小波变换的分布式信息一致滤波算法

Wavelet transform based distributed information consensus filter

研究一类基于小波变换的分布式信息一致滤波算法.首先,利用Haar小波变换建立目标状态及其观测在不同粗尺度下的系统模型;然后,基于该模型,在不同粗尺度上分别进行分布式信息一致滤波估计;最后,针对不同粗尺度估计,通过Haar小波逆变换重构最细尺度(初始尺度)目标状态的估计.仿真结果表明,所提出的算法可以有效提高分布式信息一致滤波算法的计算效率.

A class of wavelet transform based distributed information consensus filtering algorithm is studied. Firstly,the Haar wavelet transform is applied to establish the systems models of the target state and its observations at different coarser scales. Then, based on the above models, the consensus-based distributed information filtering is proceed at different coarser scales. Finally, the inverse Haar wavelet transform is applied to reconstruct the estimation of target state at the finest scale（initial scale） by using estimations at different coarser scales. Simulation results show that the proposed algorithm can effectively improve the computation efficiency of the consensus-based distributed information filtering algorithm.