The Journal of the Acoustical Society of Korea (한국음향학회지)

The Acoustical Society of Korea (한국음향학회)
권호 표지
DOI QR코드

DOI QR Code

A study on robust recursive total least squares algorithm based on iterative Wiener filter method

반복형 위너 필터 방법에 기반한 재귀적 완전 최소 자승 알고리즘의 견실화 연구

  • Lim, Jun Seok (Department of Electrical Engineering, Sejong University)
  • 임준석 (세종대학교 전자정보통신공학과)
  • Received : 2021.02.26
  • Accepted : 2021.04.16
  • Published : 2021.05.31
Download PDF
⟨ Previous Next ⟩

Abstract

It is known that total least-squares method shows better estimation performance than least-squares method when noise is present at the input and output at the same time. When total least squares method is applied to data with time series characteristics, Recursive Total Least Squares (RTS) algorithm has been proposed to improve the real-time performance. However, RTLS has numerical instability in calculating the inverse matrix. In this paper, we propose an algorithm for reducing numerical instability as well as having similar convergence to RTLS. For this algorithm, we propose a new RTLS using Iterative Wiener Filter (IWF). Through the simulation, it is shown that the convergence of the proposed algorithm is similar to that of the RTLS, and the numerical robustness is superior to the RTLS.

입력과 출력에 동시에 잡음이 존재하는 경우 최소 자승법 보다는 완전 최소 자승법이 더 우수한 추정 성능을 보인다는 것이 알려져 있다. 완전 최소 자승법을 시계열 특성을 가지는 데이터에 적용할 경우 보다 실시간 성을 더하기 위해서 Recursive Total Least Squares(RTS) 알고리즘이 제안되어 있다. RTLS는 알고리즘 내에 존재하는 역행렬 계산에서 수치적인 불안정성을 지닌다. 본 논문에서는 RTLS와 유사한 수렴성을 지닐 뿐만 아니라 수치적 불안정성을 줄이기 위한 알고리즘을 제안한다. 이 알고리즘을 위해서 Iterative Wiener Filter(IWF)를 적용한 새로운 RTLS를 제안한다. 시뮬레이션을 통해서 수렴성이 기존의 RTLS와 유사할 뿐만 아니라 수치적 견실성이 기존 RTLS보다 향상되었다는 것을 보인다.

Keywords

References

  1. C. Davila, "An efficient recursive total least squares algorithm for FIR adaptive filtering," IEEE Trans Signal Process. 42, 268-280 (1994). https://doi.org/10.1109/78.275601
  2. R. Arablouei, K. Dogancay, and S. Werner, "Recursive total least-squares algorithm based on inverse power method and dichotomous coordinate-descent iterations," IEEE Trans Signal Process. 63, 1941-1949 (2015). https://doi.org/10.1109/TSP.2015.2405492
  3. N. Choi, J. Lim, J. Song, and K. Sung," Adaptive system identification using an efficient recursive total least squares algorithm" (in Korean), J. Acoust. Soc. Kr. 22, 93-100 (2003).
  4. J. Lim, "Error in variable FIR typed system identification using combining total least mean squares estimation with least mean squares estimation" (in Korean), J. Acoust. Soc. Kr. 29, 97-101 (2010).
  5. J. Lim and Y. Pyeon, "FIR system identification method using collaboration between RLS (recursive least squares) and RTLS (recursive total least squares)" (in Korean), J. Acoust. Soc. Kr. 29, 374-380 (2010).
  6. J. Lim and H. Pang, "Reweighted l1 regularized TLS linear neuron for the sparse system identification," Neurocomputing, 173, 1972-1975 (2016). https://doi.org/10.1016/j.neucom.2015.08.020
  7. J. Lim and H. Pang, "Mixed norm regularized recursive total least squares for group sparse system identification," Int. J. Adapt. Control Signal Process. 30, 664-673 (2016). https://doi.org/10.1002/acs.2635
  8. J. Lim and H. Pang, "l1- regularized recursive total least squares based sparse system identification for the error-in-variables," SpringerPlus, 5, 1-9 (2016). https://doi.org/10.1186/s40064-015-1659-2
  9. A. H. Sayed, Fundamentals of Adaptive Filtering (Wiley, NewYork, 2003), pp. 212-280.
  10. B. Xi and Y. Liu, "Iterative Wiener filter," Electronics Letters, 28, 1892-1899 (2013).
  11. J. Lim, "L1-norm iterative wiener filter for sparse channel estimation," Circuits Syst Signal Process. 39, 6386-6393 (2020). https://doi.org/10.1007/s00034-020-01467-x