A detection method of closely spaced frequency components in non-stationary signals

Abstract

A detection method for closely spaced frequency components of a non-stationary signal, said method comprising the following steps: performing EMD decomposition on the non-stationary signal to be measured, and then obtaining the time spectrum and marginal spectrum of the non-stationary signal to be measured by Hilbert transform; The value of each frequency component is obtained by processing the spectrogram data, and the signals of different frequency components are extracted by filtering method; through AMD decomposition, it is judged whether each frequency component contains multiple frequency values ​​that have not been separated; if the signal has frequency mixing If there is an overlap phenomenon, AMD decomposition is performed on the frequency components to separate signals with similar frequencies; the separated signals are then operated in the order of the above steps until a single frequency signal is separated to ensure that the decomposed signals are all single frequency components. The invention solves the problem that the Hilbert-Huang transform cannot effectively separate two closely spaced frequency component signals, ensures the correct decomposition of effective data of the signal, and improves the resolution precision of the signal.

Description

Technical field [0001] The invention relates to the technical field of signal processing, in particular to a detection method for closely spaced frequency components of non-stationary signals. Background technique [0002] In the field of signal decomposition and processing, the Hilbert-Huang Transformation (HHT) algorithm is proposed to make up for the shortcomings of traditional time-frequency analysis methods. HHT has strong adaptability to the processing of non-stationary signals. . Its core idea is to decompose the time series into EMD first, and then apply the Hilbert transform signal processing method to each component. HHT can analyze linear stationary signals as well as nonlinear non-stationary signals. [0003] Although the Hilbert-Huang transform method can effectively process non-stationary signals, not all signals can be effectively decomposed. When processing signals with similar frequencies, the function will be greatly weakened. The effective decomposition of Hil...

AI Technical Summary

Problems solved by technology

This method overcomes the problem that HHT cannot decompose signals with closely spaced frequency components, but this method needs to determine each frequency component in the signal

[0005] Since the Fourier transform is not suitable for non-stationary signals, although the Hilbert-Huang transform can effectively deal with non-stationary signals, it cannot identify closely spaced frequency components; while the AMD method overcomes the fact that HHT cannot decompose closely spaced frequencies However, AMD is based on Fourier transform, which makes it difficult to detect closely spaced frequency components of non-stationary signals.

Images