Undersampled signal high-resolution reconstruction method based on dictionary learning and sparse representation

Abstract

The invention relates to an undersampled signal high-resolution reconstruction method based on dictionary learning and sparse representation, solves the problems that high-frequency components in undersampled signals are lost and the signal acquisition process is influenced by the external environment, and mainly comprises two stages of training and test reconstruction. According to the method, dictionary training is carried out on a large amount of training data mainly through a sparse representation model, dictionary pairs containing high-resolution and low-resolution signal feature information are established, and the dictionary pairs and sparse coding are used for completing the signal reconstruction process on low-resolution signals. The preprocessing method is to obtain the feature information of the low-resolution signal to the maximum extent, optimize the reliability of dictionary training and perform discrete wavelet transform on the low-resolution signal to serve as the signal. The method provided by the invention is easy to implement, short in running time and high in reconstruction efficiency, and by utilizing the similarity of sparse representation, not only is the influence of a low-sampling-rate signal on the reconstruction effect reduced, but also the lost frequency component in the undersampled signal is successfully estimated.

Description

technical field [0001] The invention belongs to the technical field of signal processing, and in particular relates to high-resolution reconstruction of under-sampled signals, in particular to a high-resolution reconstruction method of under-sampled signals based on dictionary learning and sparse representation. It is used in applications such as industrial robot status monitoring and fault diagnosis. Background technique [0002] Motor is an important rotating mechanical device, and it is more and more widely used in the production system of modern society. Due to the complex working environment, large load, and narrow installation space of the motor, its failure rate is high. As the main driving equipment of industrial production equipment, once a failure occurs, it will cause great economic losses. Therefore, the research on motor fault diagnosis technology has important practical significance. The motor fault diagnosis method based on the vibration signal is the most c...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a high-resolution reconstruction method for undersampled signals based on dictionary learning and sparse representation. This invention solves the problem of signal reconstruction for undersampled current signals obtained from motor drivers in the prior art. technical issues of the method

Images