Improved phase difference frequency estimation method

Abstract

The invention discloses an improved phase difference frequency estimation method. The method comprises: (2N-1) samples are inputted, windowless full-phase fast fourier transform is carried out on the samples to obtain a frequency spectrum Y(k), and N points are selected from the samples to carry out windowless full-phase fast fourier transform to obtain a frequency spectrum X(k); searching is carried out on the X(k) and the Y(k) to obtain a spectral peak location, spectral peak phases phiX(k<*>) and phiY(k<*>), a difference value delta phi~, phase adjustment is carried out to obtain delta phi, and dividing is carried out by tau=(N-1) / 2 to obtain a frequency offset estimation value delta^1; |delta^1| is larger than or equal to xi, a sequence s(n)=exp(-j2pidelta^1n / N), time-domain modulation is carried out on the samples to obtain a frequency offset estimation value delta^<*>1, the delta^<*>1 is added to a delta^2 to obtain a frequency offset estimation value delta^; the |delta^1| is larger than or equal to 1 and is less than xi and the frequency offset estimation value delta^ is equal to delta^1; and the delta^ is substituted into a formula: f^0=(k<*>+delta^) / N to calculate a frequency offset estimation value. With introduction of frequency shift compensation measures, a solution measure to phase difference integer ambiguity is put forward, so that practicability is enhanced.

Description

technical field [0001] The invention relates to the field of digital signal processing, and more specifically relates to an improved phase difference frequency estimation method. Background technique [0002] Frequency estimation is an important part of digital signal processing, especially the frequency estimation of signals containing Gaussian white noise has always been a classic subject of signal processing. Frequency estimation has important research value not only in theory but also in practical applications (such as many Puller effect detection [1] , array direction of arrival estimation [2] , Speed ​​measurement in vibration analysis [3] etc. can be transformed into a frequency estimation problem). Since FFT (Fast Fourier Transform, Fast Fourier Transform) has the advantage of high computing efficiency, frequency estimation based on FFT has been a long-standing research topic in the academic circle. [0003] At present, frequency estimation mainly includes interp...

AI Technical Summary

Problems solved by technology

However, there is still room for improvement in the estimation accuracy of the classic apFFT / FFT phase difference method, and its RMSE curve still has a certain distance from the Cramer-Raine limit; in addition, in order to enhance the value of practical engineering applications, the estimator should also have Adaptability to Frequency Estimation of Multi-frequency Signals

Images