Moving Average Estimator Least Mean Square Using Echo Cancellation Algorithm

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 215)


Eco cancellation algorithm should not only promptly adapt itself to changing environment but also minimize effects of a speech signal. However, since the color noise does not feature a consistent signal, it certainly has a significant influence on the speech signal. In this paper, the echo cancellation algorithm with a moving average LMS filter applied has been proposed. For the color noise cancellation method, an average estimator was measured by LMS adaptation filter techniques while a LMS filter step size was controlled. In addition, as it was designed to converge on a non-noise signal, the echo signal was cancelled which would, in return, lead it to the improvement of a performance. For the color noise environment, the echo cancellation Algorithm with the Average Estimator LMS filter used was applied and, a result to prove a convergence performance and stability to be improved by 10 dB comparing to the current method was gained.


Echo cancellation Moving average estimator Least mean square (LMS) filter Adaptive filter Noise cancellation 



This work was supported by the Gachon University research fund of 2012. (GCU-2012-R168).


  1. 1.
    Homer J, Mareels I (2004) LS detection guided NLMS estimation of sparse system. In: Proceedings of the IEEE 2004 international conference on acoustic, speech, and signal processing (ICASSP), MontrealGoogle Scholar
  2. 2.
    Michael H (2003) Acoustic echo cancellation digital signal processing. Bachelor of Engineering thesis, the school of electrical engineering, the University of QueenslandGoogle Scholar
  3. 3.
    Edward V (1999) Signal prewhitening schemes for detection-guided LMS estimates. Department of Electrical and Computer Engineering, University of Queensland, BrisbaneGoogle Scholar
  4. 4.
    Gay SL (2000) Acoustic signal processing for telecommunication. Kluwer Academic Publishers, DordrechtGoogle Scholar
  5. 5.
    Homer J (2000) Detection guided NLMS estimation of sparsely parameterized channels. IEEE Trans Circuits Syst II Analog Signal Process 47(12):1437–1442Google Scholar
  6. 6.
    Haykin S (2002) Adaptive filter theory. Prentice Hall, Upper Saddle RiverGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  1. 1.School of Interactive MediaGachon UniversitySeong-Nam-SiSouth Korea
  2. 2.School of Computer EngineeringKwangwoon UniversitySeoulSouth Korea

Personalised recommendations