DOI QR코드

DOI QR Code

Decision Feedback Algorithms using Recursive Estimation of Error Distribution Distance

오차분포거리의 반복적 계산에 의한 결정궤환 알고리듬

  • Kim, Namyong (School of Electronic, Info. & Comm. Engineering, kangwon National University)
  • 김남용 (강원대학교 전자정보통신공학부)
  • Received : 2015.02.11
  • Accepted : 2015.05.07
  • Published : 2015.05.31

Abstract

As a criterion of information theoretic learning, the Euclidean distance (ED) of two error probability distribution functions (minimum ED of error, MEDE) has been adopted in nonlinear (decision feedback, DF) supervised equalizer algorithms and has shown significantly improved performance in severe channel distortion and impulsive noise environments. However, the MEDE-DF algorithm has the problem of heavy computational complexity. In this paper, the recursive ED for MEDE-DF algorithm is derived first, and then the feed-forward and feedback section gradients for weight update are estimated recursively. To prove the effectiveness of the recursive gradient estimation for the MEDE-DF algorithm, the number of multiplications are compared and MSE performance in impulsive noise and underwater communication environments is compared through computer simulation. The ratio of the number of multiplications between the proposed DF and the conventional MEDE-DF algorithm is revealed to be $2(9N+4):2(3N^2+3N)$ for the sample size N with the same MSE learning performance in the impulsive noise and underwater channel environment.

정보이론적 학습의 한 성능기준인 두 오차확률분포간 유클리드거리(MEDE)는 비선형 (결정 궤환, DF) 등화 알고리듬에 채택되었고 심각한 채널 왜곡과 충격성 잡음이 있는 환경에서 탁월한 성능을 보였다. 그러나 이 MEDE-DF 알고리듬은 과중한 계산 복잡성이라는 문제를 지니고 있다. 이 논문에서는 MEDE-DF 알고리듬을 위한 반복적 ED를 먼저 유도하고 그 다음 전후방 영역에 대해 가중치 기울기를 반복적으로 추정하는 식을 유도하였다. MEDE-DF 알고리듬의 반복적 기울기 추정방식의 효과를 입증하기위해 곱셈 계산량을 비교하였고 충격성 잡음과 수중 통신 환경에서 모의 실험한 MSE 성능 결과를 비교하였다. 제안한 DF 방식과 기존의 MEDE-DF 알고리듬의 곱셈 계산량 비는 샘플사이즈 N 에 대해 $2(9N+4):2(3N^2+3N)$로 나타나면서도 충격성 잡음과 수중통신 채널환경에서 동일한 MSE 학습 성능을 유지하였다.

Keywords

References

  1. S. Haykin, Adaptive Filter Theory, Prentice Hall, Upper Saddle River, 4th edition, 2001.
  2. B. Widrow, Adaptive Signal processing, Prentice - Hall, New Jersey, 1985.
  3. J. Proakis, Digital Communications, McGraw - Hill, 2nd edition, 1989.
  4. J. Principe, D. Xu and J. Fisher, Information Theoretic Learning in: S. Haykin, Unsupervised Adaptive Filtering, Wiley, (New York, USA), pp. 265-319, 2000.
  5. K. Jeong, J. Xu, D. Erdogmus, and J. Principe, "A new classifier based on information theoretic learning with unlabeled data," Neural Networks, Vol. 18, pp. 719-726, 2005. DOI: http://dx.doi.org/10.1016/j.neunet.2005.06.018
  6. N. Kim, K. Jung, and L. Yang, "Maximization of zero-error probability for adaptive channel equalization," JCN, Vol. 12. pp. 459-465, Oct. 2010. DOI: http://dx.doi.org/10.1109/jcn.2010.6388491
  7. N. Kim and Y. Hwang, "Communication equalizer algorithms with decision feedback based on error probability," Journal of the Korea Academia -Industrial cooperation society, Vol. 12, pp. 2390-2395, May, 2011. DOI: http://dx.doi.org/10.5762/KAIS.2011.12.5.2390
  8. S. Kim, S. Kim, C. Yun, Y. Lim, "Performance analysis of receiver for underwater acoustic communications using acquisition data in shallow water," The Journal of the Acoustical Society of Korea, Vol. 29, pp. 303-313, June 2010.
  9. E. Parzen, "On the estimation of a probability density function and the mode," Ann. Math. Stat. Vol. 33, p.1065, 1962. DOI: http://dx.doi.org/10.1214/aoms/1177704472
  10. Han X and C. C, "Nonlinear filter based decision feedback equalizer for optical communication systems," Opt Express, pp. 8712-8719, Apr. 2014.