High-precision delay estimation method for high-order cumulant

Abstract

The invention discloses a high-precision delay estimation method for a high-order cumulant, belonging to the field of signal processing. The method comprises the following steps: performing cross-correlation processing and autocorrelation processing on two paths of received signals; performing high-order cumulant one-dimensional slice operation on the output of the autocorrelation processing and the cross-correlation processing; setting two corresponding variables in the high-order cumulant one-dimensional slice to zero and taking another corresponding variable as time to obtain two time domain signals; performing cross-correlation operation on the two time domain signals to obtain delay information; performing Hilbert transform on the output of the previous cross-correlation operation andtaking an absolute value; sharpening correlation peaks; and performing peak detection to obtain high-precision delay information. The method effectively suppresses various noises such as white Gaussian noise, correlated noise and non-Gaussian color noise by multi-level suppression on the noises, improves the signal-to-noise ratio of the received signals, and is very suitable for high-precision delay estimation of low signal-to-noise signals in the presence of various noises.

Description

technical field [0001] The invention belongs to the field of signal processing, and in particular relates to a high-precision time delay estimation method of a high-order cumulant. Background technique [0002] Delay estimation technology has always been an important research topic in the field of signal processing. Many time delay estimation techniques have been proposed, such as generalized correlation method, adaptive method, multiple signal classification method, expectation maximization algorithm, WRELAX method and signal parameter time delay estimation technology with rotation invariant technology, etc. These time delay estimation The technical performance is different, and they are suitable for different application requirements. [0003] High-precision time delay estimation technology can be realized by quadratic correlation method, but only using quadratic correlation technology can not effectively suppress non-Gaussian color noise, thereby reducing the accuracy of...

AI Technical Summary

Problems solved by technology

[0003] High-precision time delay estimation technology can be realized by quadratic correlation method, but only using quadratic correlation technology can not effectively suppress non-Gaussian color noise, thereby reducing the accuracy of time delay estimation; due to the existence of correlation noise and some non-Gaussian noise The classic generalized correlation method is not suitable for dealing with it; although the high-order cumulants can fully describe the properties of the probability characteristics of non-Gaussian random signals, and are very suitable for the processing of non-stationary random signals, the calculation is relatively large. In practical applications, the data cannot be processed in real time, so it is limited to theoretical research and cannot be really applied to engineering practice; although the adaptive delay estimation technology converts the delay estimation problem into the estimation problem of the filter coefficient, it can be obtained by filtering However, when the filter parameters are set unreasonably, the adaptive delay estimation algorithm has the problem of slow convergence or even no convergence. When the amount of detection data is small, the adaptive delay estimation technology large error

There are many defects in the existing time delay estimation methods, so that the accuracy of the time delay estimation of the received signal is not high

Images