Adaptive echo cancellation method of time-frequency memory sub-band proportion and feedback method

Abstract

The invention discloses an adaptive echo cancellation method of a time-frequency memory sub-band proportion, and belongs to the technical field of adaptive electroacoustic echo cancellation. In view of the echo cancellation problem, the invention provides the adaptive echo cancellation method of thetime-frequency memory sub-band proportion. On one hand, in view of the instable properties of input echo signals, a sub-band structure is adopted to perform decorrelation on the input echo signals on the frequency domain, and an AP algorithm is adopted to perform pre-whitening from the time domain so as to acquire better echo cancellation performance; and on the other hand, a scale factor carrying time-frequency memory is used for resisting against the sparsity of an echo channel to acquire a higher convergence rate.

Description

technical field [0001] The invention specifically relates to a time-frequency memory sub-band ratio adaptive echo cancellation method, and belongs to the technical field of electroacoustic adaptive echo cancellation. Background technique [0002] During a hands-free conference call or a mobile phone call, due to the call environment, you can often hear your own delayed voice from the speaker, and this delayed voice is called an echo. In order to eliminate this kind of echo and improve the call quality, an acoustic echo canceller (AEC) is usually installed inside the acoustic system to generate an acoustic signal with the same amplitude as the echo but opposite in phase, and the received signal is compared with this at the sound receiving end. The signals of the same amplitude and anti-phase are summed to eliminate echo interference. [0003] The core problem of echo cancellation is the tracking of the echo channel and the estimation of the echo. Ideally, when the echo estim...

AI Technical Summary

Problems solved by technology

[0005] Reference 1: In the Chinese Invention Patent Application No. 201510028006.4, titled "An Improved Convex Combination Decorrelation Proportional Adaptive Echo Cancellation Method", it uses a convex combination structure of filters for the sparsity of the echo channel, combined with The decorrelation adaptive algorithm and the proportional adaptive algorithm effectively alleviate the influence of channel sparsity on the algorithm convergence rate, and obtain a faster convergence rate, low steady-state error and better anti-interference performance, but due to the need for two sets of The filter constitutes a convex combination structure, which consumes a lot of resources and computational complexity. The characteristics of the algorithm are more sensitive to the mixing parameters of the convex combination, which increases the difficulty of hardware implementation.

[0006] Reference 2: "An improved multiband-structured subband adaptive filter algorithm" (Chinese name: An improved multiband-structured subband adaptive filter algorithm Yang, F., Wu, M., Ji, P., Yang, J .,IEEE Signal Process.Lett.,2012,19,(10),pp.647-650) Aiming at the non-stationary characteristics of the echo signal, the sub-band segmentation method is used to divide the signal spectrum into multiple sub-bands, so that each Each sub-band signal has the characteristics of white noise, which reduces its non-stationarity. In addition, the affine projection algorithm (Affine projection, AP) is used to update the filter tap weight coefficients of each sub-band, so that the entire update structure has a time-frequency two-dimensional characteristics, effectively improving the performance of echo suppression, but it does not consider the sparse characteristics of the echo channel, and the case of a large number of sub-bands will bring huge computational complexity

[0007] Reference 3: "An efficient proportional affine projection algorithm forecho cancellation" (Chinese name: an effective echo cancellation affine projection algorithm Paleologu, C., Ciochina, S., Benesty, J.IEEE Signal Process. Lett., 2010 ,17,(2),pp.165–168) In view of the sparse characteristics of the echo signal and the non-stationarity of the echo, the filter tap weight coefficient ratio update and the whitening effect of the AP algorithm in the time domain are used to improve the convergence of the algorithm rate and echo cancellation performance, but this method still has the problem of slow convergence rate because it does not consider the frequency domain characteristics of non-stationary signals

Images