Proceedings of the Institute of Acoustics Novel Digital Techniques for Echo Cancellation Applied to Speech Signals

Figure 6 shows the desired signal (i.e. signal without echo) superimposed on to the Kalman Filter output. As it is clear from this figure, the output of the Kalman filter is not completely matching to the desired non echo signal and therefore the output of the Kalman filter is acting as an input to the LMS adaptive algorithm; result of which is presented in Figure 7.

As well as the graphical illustration of the hybrid filter outputs, numerical comparison of this filter outputs with that of desired has also been considered and presented in Table 1.

Graphical and numerical illustration of the LMS adaptive filter output signal and the desired signal demonstrates that the LMS adaptive filter and the desired signal display similar profile, proving that the echo has been successfully removed.

In this research, all the existing algorithms of the adaptive filter have been studied, mathematically derived and implemented; out of which, the LMS algorithm was chosen due to the computational simplicity of this adaptive algorithm.

In this paper, a novel approach of using the Kalman Filter and the LMS adaptive algorithm to build a hybrid filter, in an attempt to reduce the effect of echoes over the channel, has been demonstrated. As expected, this merger produced a very good response in terms of echo cancellation. The success of this combination was confirmed by both graphical and numerical analysis as well as the output sound.

The above procedures involved acquisition and processing of both simulated and real data sets (recorded voices). Having considered the examined algorithm and filter, and results obtained, it could be said that this paper has met the specified goals and could act as a stepping stone towards further research in speech enhancement for the communication based applications.

1. N. Pavlidou, A. J. Han Vinck, J. Yazdani, and B. Honary. Power Line Communications: State of the Art and Future Trends. IEEE Communications Magazine. 41(4) pp. 34-40. (April 2003)

2. A. Spanias and M. E. Diesher, “Speech Enhancement”, Arizona State University. (1997)

3. S. V. Vaseghi, Multimedia Signal Processing Theory and Application in Speech, Music and Communication, John Wiley & Sons, Ltd. (2007)

4. S. Raghavendran, Implementation of an Acoustic Echo Canceller Using Matlab®, University of South Florida, p. 12. (2003)

5. S. Haykin, Adaptive Filter Theory. 1st Edition, Prentice-Hall Inc., pp. 1-31. (1986)

6. R.E. Kalman, A New Approach to Linear Filtering and Prediction Problems, Transaction of the ASME - Journal of Basic Engineering. (March 1960)

7. G. Welch, and G. Bishop, An Introduction to the Kalman Filter, University of North Carolina. (July 2006)

8. Peter S. Maybeck, Stochastic Models, Estimation, and Control, Volume 1, Academic Press, Inc. (1979)

9. Sorenson, H. W. 1970. “Least-Squares estimation: from Gauss to Kalman,” IEEE Spectrum, vol. 7, pp. 63-68. (July 1970)

10. A. Gelb. Applied Optimal Estimation, MIT Press, Cambridge. (1974)

11. Grewal, S. Mohinder, and P. Andrews Angus, Kalman Filtering Theory and Practice. Upper Saddle River, NJ USA, Prentice Hall. (1993)

12. Lewis, Richard. Optimal Estimation with an Introduction to Stochastic Control Theory, John Wiley & Sons, Inc. (1986)

13. R. G. Brown and P. Y. C. Hwang., Introduction to Random Signals and Applied Kalman Filtering, Second Edition, John Wiley & Sons, Inc. (1992)

14. O.L.R. Jacobs. Introduction to Control Theory, 2nd Edition. Oxford University Press. (1993)

15. S. Haykin., Adaptive Filter Theory. 2nd Edition, Prentice-Hall Inc., New Jersey. (1991)

16. A.D. Poularikas and Z.M. Ramadan, Adaptive Filtering Premier with MATLAB®, Taylor & Francis Group LLC. (2006)

17. M. Hutson, Acoustic echo cancellation using digital signal processing, The University of Queensland. (2003)