Genetic algorithm for Echo cancelling

Subject Areas : International Journal of Smart Electrical Engineering

Alireza rezaee ¹

1 - Electrical Engineering Department, Hashtgerd Branch, Islamic Azad University, Hashtgerd, Iran

Received: 2015-04-13 Accepted : 2015-04-13 Published : 2012-06-01

Keywords: Genetic Algorithm, echo cancellation, heuristic crossover, microbial crossover,

Abstract :

In this paper, echo cancellation is done using genetic algorithm (GA). The genetic algorithm is implemented by two kinds of crossovers; heuristic and microbial. A new procedure is proposed to estimate the coefficients of adaptive filters used in echo cancellation with combination of the GA with Least-Mean-Square (LMS) method. The results are compared for various values of LMS step size and different types of crossovers which are all satisfactory. Reverse SNR is used as the fitness function. It can estimate an echo path with definite length of impulse response with an adaptive filter with desired length. Results show that the proposed combined GA-LMS method operates more satisfactory than simple GA in terms of the number of generations needed to achieve a particular amount of echo cancellation. Different tests show that GAs running with heuristic crossover converge faster than GAs with microbial crossover. Results are also compared with LMS algorithm. Although LMS is faster, but its solutions are less precise and it diverges in some cases. But our proposed method always converges.

References:

[1] V. Krishna, J. Rayala and B. Slade, “Algorithmic and implementation aspects of echo cancellation in packet voice networks”, Proc. Of Signals, Systems and Computers, vol. 2, pp. 1252-1257, Nov. 2002.

[2] MATLAB version 7, Help documents, 2004

[3] D. M. Etter, M. J. Hicks, and K. H. Cho, “Recursive adaptive filter design using an adaptive genetic algorithm”, Proc. of Acoustics, Speech, and Signal Processing, pp. 635-638, May 1982.

[4] A. Neubauer, “Non-linear adaptive filters based on genetic algorithms with applications to digital signal processing”, Proc. of Evolutionary Computation, vol.2, pp. 527-532, Nov. 1995.

[5] S. Haykin, “Adaptive filter theory”, third edition, New Jersey, 1996

[6] International Telecommunication Union, “ITU-T G.168: Digital network echo cancellers”, ITU 2004.