Abstract: Reconstruction of high sample-rate industry process data is important for multi-rate control and identification problem. In this paper, a hierarchical error-compensation algorithm is presented to treat data reconstruction issues based on wavelet multiresolution analysis theory. The low sample-rate data contaminat-ed by noises is first filtered in the time-frequency domain, then the proposed algorithm is used to reconstruct it to a new high sample-rate signal. Reconstruction errors are given and analysis results are justified. This algorithm has the advantages of noise-free, high reconstruction accuracy and explicit physical background. Simulation examples are given to illustrate the proposed algorithm.