Restraint Method Research for Coupling Random Error Based on High Dimensional Data Set Multiscale Analysis

摘要

For coupling random error, a new method based on multiscale analysis on high dimensional data set is advanced in this paper. It is extends traditional wavelet to high dimensional data set and does multiscale analysis, so that the information in the multi-measure data can be used better. The step of the restraint algorithm is given. Finally, simulated ship attitude data is used to verify the new method. The results show that the three angles in ship attitude are coupling and the method proposed in this paper is valid, which can get a better restraining result than traditional wavelet method.