Journal of the Institute of Convergence Signal Processing (융합신호처리학회논문지)

The Korea Institute of Convergence Signal Processing (한국융합신호처리학회)
권호 표지

Variable Step LMS Algorithm using Fibonacci Sequence

피보나치 수열을 활용한 가변스텝 LMS 알고리즘

  • Woo, Hong-Chae (School of Computer and Communication Engineering, Daegu University)
  • 우홍체 (대구대학교 정보통신공학부)
  • Received : 2018.05.25
  • Accepted : 2018.06.25
  • Published : 2018.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

Adaptive signal processing is quite important in various signal and communication environments. In adaptive signal processing methods since the least mean square(LMS) algorithm is simple and robust, it is used everywhere. As the step is varied in the variable step(VS) LMS algorithm, the fast convergence speed and the small excess mean square error can be obtained. Various variable step LMS algorithms are researched for better performances. But in some of variable step LMS algorithms the computational complexity is quite large for better performances. The fixed step LMS algorithm with a low computational complexity merit and the variable step LMS algorithm with a fast convergence merit are combined in the proposed sporadic step algorithm. As the step is sporadically updated, the performances of the variable step LMS algorithm can be maintained in the low update rate using Fibonacci sequence. The performances of the proposed variable step LMS algorithm are proved in the adaptive equalizer.

다양한 신호처리 및 통신환경에서 적응신호처리는 매우 중요하다. 적응신호처리 방식 중에서 least mean square(LMS) 알고리즘은 단순하면서도 강인하기 때문에 널리 사용되고 있다. 가변스텝 LMS 알고리즘은 스텝을 가변하므로 빠른 수렴속도와 작은 초과자승오차를 얻을 수 있는 방식이다. 성능향상을 위하여 다양한 가변스텝 LMS 알고리즘이 연구되어 왔다. 하지만 성능향상을 위하여 가변스텝 LMS 알고리즘의 계산 복잡도는 일부 방식에서는 크게 높아지게 되었다. 계산 복잡도가 낮은 고정스텝 LMS 알고리즘과 빠른 수렴속도의 가변스텝 LMS 알고리즘의 장점을 같이 가질 수 있는 간헐적 스텝 갱신 알고리즘을 제안한다. 간헐적으로 스텝 갱신을 할 때 피보나치 수열을 사용하여 스텝 갱신 횟수를 상당히 낮추면서도 가변스텝 LMS 알고리즘의 성능을 유지할 수 있었다. 적응 등화기에 제안한 가변스텝 LMS 알고리즘을 적용하여 그 성능을 확인하였다.

Keywords

References

  1. D. Bismor, K. Czyz, and Z. Ogonowski, "Review and comparison of variable step-size LMS algorithm," International Journal of Acoustics and Vibration, vol. 21. no.1, pp. 24-39, 2016.
  2. R. H. Kwong, "A variable step size LMS algorithm," IEEE Trans. On Signal Proc., vol. 40., no. 7, pp. 1633-1642, 1992. https://doi.org/10.1109/78.143435
  3. V. J. Mathews, and Z. Xie, "A stochastic gradient adaptive filter with gradient adaptive step size," IEEE Trans. On Signal Proc., vol. 41, no. 6, pp. 2075-2087, 1993. https://doi.org/10.1109/78.218137
  4. T. Aboulnasr and K. Mayyas, "A robust variable step size LMS-type algorithm: analysis and simulations," IEEE Trans. On Signal Proc., vol. 45, no. 3, pp. 631-639, 1997. https://doi.org/10.1109/78.558478
  5. D. I. Pazaitis and A. G. Constantinides, "A novel kurtosis driven variable step-size adaptive algorithm," IEEE Trans. On Signal Proc., vol. 47, no. 3, pp. 864-872, 1999. https://doi.org/10.1109/78.747793
  6. T. I. Haweel, "simple variable step size LMS adaptive algorithm," International Journal of Circuit Theory and Applications, vol. 32, pp. 523-536, 2004. https://doi.org/10.1002/cta.294
  7. Hong Chae Woo, "Deterministic function variable step size LMS algorithm," Journal of the Korea Institute of Convergence Signal Processing, vol. 2, no. 2, pp. 128-132, 2011.
  8. S. Haykin, "A adaptive filter theory," 4th Ed., New York: Prentice Hall, 2002.