改良式噪音降低和迴音消除系統實作

An Implementation of Improving Noise Reduction and Acoustic Echo Cancellation

李志偉(Chih-Wei Lee)

指導教授 ： 汪柏

淡江大學/工學院/資訊工程學系碩士班/碩士(2005年)

https://doi.org/10.6846/TKU.2005.00519

摘要

一般應用於迴音消除上使用的Adaptive signal processing都需要一個期望訊號，根據期望訊號與來源訊號間不斷的誤差調整，來獲得所需訊號。不過在通訊環境中，為了消除環境週遭的噪音與迴音，如何取得有用的期望訊號，根據期望訊號來適時調整取得所需訊號，將增加系統的複雜程度。
本論文提出藉由噪音降低機制所產生的訊號透過特殊處理，使其具備迴音路徑，並根據此訊號作為迴音消除器的期望訊號，結合來源訊號做迴音消除動作，藉此簡化迴音消除器和噪音降低機制的系統架構，達到同時降低噪音與消除迴音的功能。
本論文將實地應用單一麥克風音源錄製在汽車行駛環境內的通訊樣本，而非模擬車內環境噪音所後製的樣本，並針對錄製的樣本，使用本論文改良的系統架構進行降低噪音和消除迴音動作，以驗證在本論文系統架構中，可以有效消除在車內環境通訊時，所產生的噪音和迴音，同時達到簡化系統架構的目的。

關鍵字

噪音降低機制 ； 迴音抑制器 ； 快速最小均方值演算法

並列摘要

In general, adaptive signal processing applied in echo cancellations all need a desired signal value, This desired signal is used along with the source signal as references to continuously calculate and tune up the needed signal. However, under a communications environment the effort of obtaining the signal by computing the result of the desired and source signal in order to eliminate ambient noise and echo would increase the system's complexity.
This thesis proposes a method to reduce noise and eliminate echo without introducing excessive complications to the infrastructure of the system. Our method involves using a particular process to find echo route relations with the signals produced by the noise reduction system, and using this data as the desired signal value combined with the source signal as inputs of the echo cancellation step.
The experiment's environmental setup involves using a single microphone as the sound input recording the sounds in a driving automobile as samples instead of capturing data from a simulated environment. We apply our improved method to the samples and perform the steps of reducing noise and eliminating echo. The result will show our method is capable of eliminating noise and echo produced in a car's interior and at the same time avoiding complexities added to our system's infrastructure.

並列關鍵字

Noise Reduction ； Echo Cancellation ； FLMS algorithm

參考文獻

[3] Marc Moonen and Ian Proudler, An Introduction to Adaptive Signal Processing, course notes available at ESAT website at K.U.Leuven.

[7] Earl R. Ferrara, “Fast Implementation of LMS Adaptive Filters,” IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-28, pp.474-475, Aug. 1980.

[8] S. F. Boll, “Suppression of Acoustic Noise in Speech Using Spectral Subtraction,” IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-27, pp.113-120, Apr. 1979.

[9] M. Dahl and I. Claesson, “Acoustic Noise and Echo Canceling with Microphone Array,” IEEE Transactions on Vehicular Technology, vol.48, No.5, pp.1518-1526, Sep. 1999.

[10] H. Buchner, J. Benesty and W. Kellermann, “An Extended Multidelay Filter: Fast Low-Delay Algorithms for Very High-Order Adaptive Systems” Proc. IEEE Int. Conf. on Acoustics, Speech, and Signal Processing (ICASSP), Hong Kong, China, Apr. 2003.