A Neighboring Point Correlation Mean Mean Noise Reduction Method for FFT Data

Abstract

The invention discloses a noise reduction method for the correlation mean value of adjacent points implemented on FFT data. A reference value dRef is set, and the FFT signal data to be processed is compared with the reference value dRef. For the data of dRef, take the data of the adjacent set number to obtain the average value, and replace the data value lower than the reference value dRef with the average value; reduce the amplitude of the distortion data below the reference value, and reduce the noise at the same time The band makes the signal below the reference value close to the center of the noise. The method of averaging adjacent data can reduce the amplitude of the distorted data below the reference value, and at the same time reduce the noise band, so that the signal below the reference value converges to the center of the noise, which can suppress the noise and reduce the energy of the interference signal to a certain extent.

Description

technical field [0001] The invention relates to the field of digital signal processing, in particular to an adjacent point correlation mean noise reduction method for FFT data. Background technique [0002] In the field of high-speed acquisition signal analysis and processing, after the high-speed sampling data is transformed and processed by FFT, due to the difficulty of hardware design and real-time data analysis and processing under high-speed sampling conditions, it is often caused by large bandwidth problems, signal leakage problems, signal aliasing problems, band Internal self-excitation problems, etc., make the effect of FFT unsatisfactory. Some unwanted signals or small signals that are not concerned will be mixed on the signal noise floor. It is difficult to improve hardware and optimize programmable logic timing at high speeds. Solve and affect the data processing effect. [0003] Usually, when the hardware environment is not ideal, especially in high-speed applic...

AI Technical Summary

Problems solved by technology

[0002] In the field of high-speed acquisition signal analysis and processing, after the high-speed sampling data is transformed and processed by FFT, due to the difficulty of hardware design and real-time data analysis and processing under high-speed sampling conditions, it is often caused by large bandwidth problems, signal leakage problems, signal aliasing problems, band Internal self-excitation problems, etc., make the effect of FFT unsatisfactory. Some unwanted signals or small signals that are not concerned will be mixed on the signal noise floor. It is difficult to improve hardware and optimize programmable logic timing at high speeds. Solve, affect the effect of data processing

[0003] Usually, when the hardware environment is not ideal, especially in high-speed applications, it is generally difficult to completely improve the signal effect through hardware optimization.

At the same time, the improvement of the hardware environment needs to increase labor costs, time expansion, and hardware resource consumption to a certain extent, and in many cases, due to different environments, the results of one debugging do not meet the general requirements of multiple environments.

Images