Deconvolution power spectrum estimation method

Abstract

The invention provides a deconvolution power spectrum estimation method. The method comprises the steps that (1) a data sample is preprocessed, and power spectrum estimation is performed on the preprocessed data sample; (2) a deconvolution algorithm is utilized to perform deconvolution operation on a power spectrum of a power spectrum identical-window function of the data sample; and (3) appropriate parameters are selected for deconvolution operation, the number of iterations is selected, and then an estimated value of the power spectrum with a true signal is obtained through iteration convergence. The method overcomes the defects of spectrum leakage and low frequency resolution during power spectrum estimation due to a limited length of a data sample. Through the method, the frequency resolution of power spectrum estimation can be remarkably increased under the condition of a small data length, the influence of a sidelobe brought by a limited data length can be effectively suppressed while high-resolution power spectrum estimation is realized, therefore, extra signal processing gain is obtained, and the method has important significance for weak signal detection in a strong interference background.

Description

Technical field [0001] The invention relates to a signal processing method, specifically a signal spectrum estimation method. Background technique [0002] Spectrum analysis techniques include spectrum analysis and power spectrum estimation, and are the most commonly used means for signal analysis in the frequency domain. The power spectrum of a signal is about the energy distribution of the signal at all frequencies that make up the signal. As the basic theory of signal spectrum analysis, at first, Fourier transform was defined as a kind of global transformation carried out in the infinite time domain and frequency domain, decomposing the time domain signal into sinusoidal signals of different frequencies. In order to deal with discrete sequences, discrete-time Fourier transform and discrete Fourier transform were developed. Fourier transform can not only be applied to power spectrum estimation, but also can be extended to the field of signal time-frequency characteristic anal...

AI Technical Summary

Problems solved by technology

[0005] The above methods and other high-frequency resolution signal spectrum estimation methods proposed by scholars can indeed improve the resolution of the signal spectrum estimation results, but the signal sampling length is still the main factor affecting the processing performance of these high-resolution spectrum estimation methods one

And other factors, such as the signal-to-noise ratio of the signal, the model matching degree of the model method, etc., will have an impact on the resolution and accuracy of the spectral estimation results.

Images