Echo Cancellation (779800), страница 4
Текст из файла (страница 4)
The main advantages of a sub-band echo canceller are a reduction in filterlength and a gain in the speed of convergence as explained below:(a) Reduction in filter length. Assuming that the impulse response ofeach sub-band filter has the same duration as the impulse response ofthe full band FIR filter, the length of the FIR filter for each downsampled sub-band is 1/R of the full band filter.(b) Reduction in computational complexity. The computationalcomplexity of an LMS-type adaptive filter depends directly on the412Echo CancellationRoom wallsSubband analyserBP(1) BP(2)...Loud speakerBP(N-1)...y k (m)BP(N-1)...BP(2)+...BP(1)–BP(2)Synthesiseryˆ kf ( m)Micro phoneSubband analyserAdaptationalgorithmBP(1)BP(N-1)Acoustic feedbacksynthesizery(m) = x(m) + yf (m)Figure 14.11 Configuration of a sub-band acoustic echo cancellation system.product of the filter length and the sampling rate.
As for each subband, the number of samples per second and the filter lengthdecrease with 1/R, it follows that the computational complexity ofeach sub-band filter is 1/R2 of that of the full band filter. Hence theoverall gain in computational complexity of a sub-band system isR2/N of the full band system.(c) Speed of convergence.
The speed of convergence depends on boththe filter length and the eigenvalue spread of the signal. The speed ofconvergence increases with the decrease in the length of the FIRfilter for each sub-band. A more important factor affecting theconvergence of adaptive filter is the eigenvalue spread of theautocorrelation matrix of the input signal. As the spectrum of asignal becomes flatter, the spread of its eigenvalues decreases, andthe speed of convergence of the adaptive filter increases.
In general,the signal within each sub-band is expected to have a flatterspectrum than the full band signal. This aids the speed ofconvergence. However, it must be noted that the attenuation of subband filters at the edges of the spectrum of each band creates somevery small eigenvalues.Summary41314.7 SummaryTelephone line echo and acoustic feedback echo affect the functioning oftelecommunication and teleconferencing systems. In general, line echocancellation, is a relatively less complex problem than acoustic echocancellation because acoustic cancellers need to model the more complexenvironment of the space of a room.We began this chapter with a study of the telephone line echoes arisingfrom the mismatch at the 2/4-wire hybrid bridge.
In Section 14.2, line echosuppression and adaptive line echo cancellation were considered. Foradaptation of an echo canceller, the LMS or the RLS adaptation methodscan be used. The RLS methods provides a faster convergence rate and betteroverall performance at the cost of higher computational complexity.In Section 14.3, we considered the acoustic coupling between aloudspeaker and a microphone system.
Acoustic feedback echo can result inhowling, and can disrupt the performance of teleconference, hands-freetelephones, and hearing aid systems. The main problems in implementationof acoustic echo cancellation systems are the requirement for a large filter tomodel the relatively long echo, and the adaptation problems associated withthe eigenvalue spread of the signal. The sub-band echo canceller introducedin Section 14.4 alleviates these problems.BibliographyALLEN J., BERKLEY D.
and BLAURET J. (1977) Multi-Microphone SignalProcessing Technique to Remove Room Reverberation from SpeechSignals. J. Acoust. Soc. Am., 62, 4.ARMBRUSTER W. (1992) Wideband Acoustic Echo Canceller with TwoFilter Structure. Proc. Eusipco-92, 3, pp. 1611–1617.CARTER G. (1987) Coherence and Time Delay Estimation. Proc. IEEE.
75,2, pp. 236–55.FLANAGAN J.L. et al. (1991) Autodirective Microphone systems. Acoustica73, pp.58–71.FLANAGAN J.L. et al. (1985) Computer-Steered Microphone Arrays forSound Transduction in Large Rooms. J. Acoust. Soc. Amer., 78, pp.1508–1518.GAO X.Y. and SNELGROVE W.M. (1991) Adaptive Linearisation of aLoudspeaker, IEEE. Proc. Int. Conf. Acoustics, Speech and SignalProcessing, ICASSP–91, 3, pp. 3589-3592.414Echo CancellationGILLOIRE A. and VETTERLI M. (1994) Adaptive Filtering in Sub-bands withCritical Sampling: Analysis, Experiments and Applications to AcousticEcho Cancellation, IEEE. Trans. Signal Processing, 40, pp.
320–28.GRITTON C.W. and LIN D.W. (1984) Echo Cancellation Algorithms, IEEEASSP Mag., 1, 2, pp. 30–37.GUSTAFSSON S. and MARTIN R. (1997) Combined Acoustic Echo Controland Noise Reduction for Mobile Communications, Proc. EuroSpeech97, pp. 1403–1406.HANSLER E. (1992) The Hands-Free Telephone Problem An AnnotatedBibliography. Signal Processing, 27, pp. 259–71.HART J.E., NAYLOR P.A. and TANRIKULU O. (1993) Polyphase All-pass IIRStructures for Subband Acoustic Echo Cancellation.
EuroSpeech-93, 3,pp. 1813-1816.HUA YE and BO-XIA WU (1991) A New Double-Talk Detection AlgorithmBased on the Orthogonality Theorem. IEEE Trans onCommunications, 39, 11, pp. 1542-1545, Nov.KELLERMANN W. (1988) Analysis and Design of Multirate Systems forCancellation of Acoustical Echoes. IEEE. Proc. Int. Conf. Acoustics,Speech and Signal Processing, ICASSP-88, pp. 2570-73.KNAPPE M.E. (1992) Acoustic Echo Cancellation: Performance andStructures.
M. Eng. Thesis, Carleton University, Ottawa, Canada.MARTIN R. and ALTENHONER J. (1995) Coupled Adaptive Filters forAcoustic Echo Control and Noise Reduction. IEEE. Proc. Int. Conf.Acoustics, Speech and Signal Processing, ICASSP-95, 5, pp. 3043-46.McCASLIN S. R., HEMKUMAR N. and REDHEENDRAN B. (1997) Double-TalkDetector for Echo Canceller. US patent No. 5631900, May 20.OSLEN H.F.
(1964) Acoustical Engineering, Toronto, D. Van Nostrand Inc.SCHROEDER M.R. (1964) Improvement of Acoustic-Feedback Stability byFrequency Shifting. J. Acoust. Soc. Amer., 36, pp. 1718-1724.SILVERMAN H.F. et al. (1992) A Two-Stage Algorithm for DeterminingTalker Location from Linear Microphone Array Data. ComputerSpeech and Language, 6, pp. 129-52.SONDHI M.M. and BERKLEY D.A. (1980) Silencing Echoes on the TelephoneNetwork. Proc.
IEEE, 68, pp. 948-63.SONDHI M.M. and MORGAN D.R. (1991) Acoustic Echo Cancellation forStereophonic Teleconferencing. IEEE Workshop on Applications ofSignal Processing to Audio And Acoustics.SONDHI M.M. (1967) An Adaptive Echo Canceller. Bell Syst. tech. J., 46,pp. 497-511.Bibliography415TANRIKULU O., etal. (1995) Finite-Precision Design and Implementation ofAll-Pass Polyphase Networks for Echo Cancellation in subbands.IEEE.
Proc. Int. Conf. Acoustics, Speech and Signal Processing,ICASSP-95, 5, pp. 3039-42.VAIDYANATHAN P.P. (1993) Multirate Systems and Filter Banks. PrenticeHall.WIDROW B., McCOOL J.M., LARIMORE M.G. and JOHNSON C.R. (1976)Stationary and Nonstationary Learning Characteristics of the LMSAdaptive Filters. Proceedings of the IEEE, 64, 8, pp. 1151-62.ZELINSKI R.
(1988) A Microphone Array with Adaptive Post-Filtering forNoise Reduction in Reverberant Rooms. IEEE. Proc. Int. Conf.Acoustics, Speech and Signal Processing, ICASSP-88, pp. 2578-81..
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
