The Journal of Korean Institute of Communications and Information Sciences (한국통신학회논문지)

The Korean Institute of Commucations and Information Sciences (한국통신학회)
권호 표지
DOI QR코드

DOI QR Code

Step Size Normalization for Maximum Cross-Correntropy Algorithms

최대 상호코렌트로피 알고리듬을 위한 스텝사이즈 정규화

  • Received : 2016.06.12
  • Accepted : 2016.08.16
  • Published : 2016.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

The maximum cross-correntropy (MCC) algorithm with a set of random symbols keeps its optimum weights undisturbed from impulsive noise unlike MSE-based algorithms and its main factor has been known to be the input magnitude controller (IMC) that adjusts the input intensity according to error power. In this paper, a normalization of the step size of the MCC algorithm by the power of IMC output is proposed. The IMC output power is tracked recursively through a single-pole low-pass filter. In the simulation under impulsive noise with two different multipath channels, the steady state MSE and convergence speed of the proposed algorithm is found to be enhanced by about 1 dB and 500 samples, respectively, compared to the conventional MCC algorithm.

무작위 발생된 심볼 집합과 최대 상호 코렌트로피 (maximum cross-correntropy) 로 설계된 MCC 알고리듬은 최소자승평균 (MSE) 기반 알고리듬과 달리, 충격성 잡음 하에서 최적 가중치가 동요 없이 안정을 유지하며 그 요인이 오차 전력에 따라 입력의 세기를 조절하는 입력 크기 조정기 (input magnitude controller, IMC)에 있음이 밝혀졌다. 이 논문에서는 스텝사이즈를 정규화한 알고리듬 (normalized MCC, NMCC)를 제안하였으며 여기서 IMC 통과된 신호 전력은 1-pole 저역 통과 필터로 반복적 추정한다. 두 가지 다중경로 채널 모델과 충격성 잡음 환경에서 시행된 시뮬레이션 결과, 정규화된 NMCC알고리듬은 MCC알고리듬에 비해 정상상태 MSE에서 1 dB 정도의 성능 향상을, 수렴 속도에서도 500 샘플 정도 빠른 성능을 나타냈다.

Keywords

References

  1. Y. Cho, H. Yu, B. Kim, J. Cho, J. Kim, J. Lee, and H. Park, "Proposal of optimum equalizer hardware architecture for cable modem and analysis of various LMS algorithms," J. KICS, vol. 27, pp. 150-159, Feb. 2002.
  2. L. Bharani and P. Radhika, "FPGA implementation of optimal step size nlms algorithm and its performance analysis," IJRET, vol. 2, pp. 885-890, Jul. 2013. https://doi.org/10.15623/ijret.2013.0205027
  3. S. Haykin, Adaptive Filter Theory, Prentice Hall, Upper Saddle River, 4th Ed., 2001.
  4. R. Chinaboina, D. Ramkiran, H. Khan, M. Usha, B. Madhav, K. Srinivas and G. Ganesh, "Adaptive algorithms for acoustic echo cancellation in speech processing," IJRRAS, vol. 7, pp. 38-42, Apr. 2011.
  5. I. Santamaria, P. Pokharel, and J. Principe, "Generalized correlation function: Definition, properties, and application to blind equalization," IEEE Trans. Sign. Process., vol. 54, pp. 2187-2197, Jun. 2006. https://doi.org/10.1109/TSP.2006.872524
  6. W. Liu, P. Pokharel, and J. Principe, "Correntropy: Properties and applications in non-gaussian signal processing," IEEE Trans. Sign. Process., vol. 55, pp. 5286-5298, Nov. 2007. https://doi.org/10.1109/TSP.2007.896065
  7. N. Kim, S. Kang, and D. Hong, "Blind equalization based on maximum cross-correntropy criterion using a set of randomly generated symbols," J. KICS, vol. 35, pp. 33-39, Jan. 2010.
  8. N. Kim, "Practical approach to blind algorithms using random-order symbols and cross-correntropy," J. KICS, vol. 39A, pp. 149-154, Mar. 2014. https://doi.org/10.7840/kics.2014.39A.3.149
  9. N. Kim, "Performance analysis of correntropy-based blind algorithms robust to impulsive noise," J. KICS, vol. 40, pp. 2324-2330, Dec. 2015. https://doi.org/10.7840/kics.2015.40.12.2324
  10. E. Parzen, "On the estimation of a probability density function and the mode," Ann. Math. Stat. vol. 33, p. 1065, 1962. https://doi.org/10.1214/aoms/1177704472
  11. J. Proakis, Digital Communications, McGraw-Hill, 2nd Ed, 1989.