A Mixing Matrix Estimation Method for Underdetermined Blind Source Separation

Abstract

The invention relates to a mixing matrix estimation method for underdetermined blind source separation, which performs short-time Fourier transform on two received observation signals to obtain the short-time Fourier coefficients X1(t,f) of the two observation signals ) and X2(t,f) to form multiple scattered points and remove low-energy scattered points; calculate the ratio of the remaining scattered points, classify the scattered points according to the ratio, and obtain the M category with the largest number of scattered points, and calculate each Then convert the mean value into the slope angle of the straight line, and then convert it into a column vector, and obtain the rotation matrix T according to the column vectors with close directions, and perform rotation transformation to obtain a new two-way observation signal X′1( t,f) and X′2(t,f); for the two observation signals X′1(t,f) and X′2(t,f), classify the scattered points again according to the ratio of the two observation signals , get the S class with the largest number of scattered points, calculate the mean value of the ratio of scattered points in each class, and then get the similar slope angles, get the corresponding column vectors through these slope angles, and finally get the entire mixing matrix.

Description

technical field [0001] The invention relates to the technical field of blind signal processing, in particular to a mixing matrix estimation method for underdetermined blind source separation. Background technique [0002] How to extract each single information from the information received by the sensor is called blind source separation, and the corresponding solution is called blind source separation technology. With the continuous research of experts and scholars at home and abroad, blind source separation technology has gradually developed, and has been widely used in mechanical fault detection, communication signal processing, voice signal processing, image signal processing, biomedical engineering and other fields. According to the number of source signals and the number of observation signals, the problem of blind source separation can be divided into two types: underdetermined blind source separation and non-underdetermined blind source separation. If the number of s...

AI Technical Summary

Problems solved by technology

This method has low complexity and is easy to implement, but its performance is easily affected by the initial value, and the number of source signals needs to be given, but the actual number of source signals may be unknown

Zhang Ye proposed a potential function method in the article "Estimation of Source Number in Underdetermined Blind Separation Based on Laplace Potential Function" published in Signal Processing Journal Vol. 25, No. 11. When the mixing matrix is ​​estimated, but this method lacks a certain theoretical basis, the subjective experience is too strong, and it is only applicable to two-dimensional space

At the same time, no matter which method is used above, when the directions of any two column vectors of the mixing matrix are very close, that is, when the slopes of two straight lines in the straight line formed by the observed signal are very close and small due to the source signal, the estimation of the mixing matrix will be caused. produce a large error

Images