Sinusoidal frequency-modulated signal parameter estimation method

Abstract

The invention belongs to the technical field of communication, and particularly relates to a sinusoidal frequency-modulated signal parameter estimation method. The sinusoidal frequency-modulated signal parameter estimation method comprises the following steps: firstly, demodulating a received signal according to an initial parameter estimation value; for reducing noise of a frequency-modulated signal, replacing the real part of the demodulated signal with a constant, the constant being the mean value of the real part of the demodulated signal, then obtaining the phase of the received signal bythe initial parameter estimation value and the phase estimation of the new demodulated signal; and finally obtaining the final parameter estimation value according to the estimated phase of the received signal. In the second part, it is considered that the modulation index is larger than pi / 2, and a parameter estimation method based on iteration is provided to obtain an initial parameter estimation value, and then the precision of the initial estimation value is improved through a method based on PNR. In the third part, the accuracy of the parameter estimates from the first part and the second part is compared and a set of parameters with higher accuracy are selected as final parameter estimates.

Description

technical field [0001] The invention belongs to the technical field of communication, and in particular relates to a parameter estimation method of a sinusoidal frequency modulation signal. Background technique [0002] Sinusoidal Frequency-Modulated (SFM) signal is a general model of micro-Doppler effect. This effect is associated with rotating or vibrating parts of radar targets. Therefore, parameter estimation of SFM signals is widely used in communication, radar, sonar and other fields. The parameters of the SFM signal include modulation frequency, modulation phase and modulation index. Through the Jacobi-Anger expansion, the SFM signal can be decomposed into an infinite number of sinusoidal signals with the modulation frequency as the interval. When the modulation index is less than 2, the SFM can be approximated as a DC signal and a real sinusoidal signal with the modulation frequency as the frequency. [0003] Aiming at the parameter estimation of sinusoidal FM sig...

AI Technical Summary

Problems solved by technology

[0004] Because the QML method needs to perform STFT transformation at different scales, its computational complexity is high, and when the modulation frequency is large, low-pass filtering will reduce the estimation performance; the SVD-MLE method needs to perform SVD decomposition on matrices of different sizes. High computational complexity; APWV-Hough method requires Hough transform, which has high computational complexity; EDSFMT method requires 3-dimensional search, which has high computational complexity

Images