Noise suppression method and device based on Gaussian distribution weighting

Abstract

The invention relates to a noise suppression method and a noise suppression device based on Gaussian distribution weighting. The noise suppression method comprises the steps of determining the local radius of a filter; determining a window length of the filter according to the local radius of the filter, and determining discrete signals in the window of the filter; determining corresponding mean values and variances according to the discrete signals in the window of the filter; determining a Gaussian function of the discrete signals in the window of the filter by using the mean values and the variances; determining the corresponding Gaussian function value of each discrete signal in the window of the filter by using the Gaussian function, and performing summation of the obtained Gaussian function values; determining the corresponding weighted value of each discrete signal in the window of the filter by using the corresponding Gaussian function value of each discrete signal in the window of the filter and the sum of the Gaussian function values; and performing noise suppression treatment on the discrete signals at the center of the window by using the weighted values by the filter.

Description

technical field [0001] The invention relates to the technical field of noise processing, in particular to a noise suppression method and device based on Gaussian distribution weighting. Background technique [0002] Signal filtering technology is the core research topic in the field of signal processing. Signal filtering methods are divided into linear filtering and nonlinear filtering. In the early digital signal and image signal processing and research, linear filtering technology is the main means of suppressing noise, mainly because the linear filtering method has an appropriate mathematical expression and is easy to design and implement. When linear filtering techniques are applied to non-superimposed noise signals, the results are often unsatisfactory. Signals are inevitably subject to varying degrees of noise interference during the acquisition and transmission process, and sometimes even shock signals are generated. At this point, the signal needs to be properly p...

AI Technical Summary

Problems solved by technology

Aiming at the noise interference or even impact interference that occurs during the signal acquisition process in the field of hydroelectric unit condition monitoring, there has not been a good solution for a long time, resulting in a series of problems such as the failure of the vibration protection system of the hydroelectric unit to be fully put into use.

Images