Frontiers in Signal Processing

Download PDF (944.7 KB) PP. 19 - 36 Pub. Date: April 30, 2019

DOI: 10.22606/fsp.2019.32001

• Seiichi Nakamori^*
  Department of Technology, Faculty of Education, Kagoshima University, Kagoshima, Japan

This paper proposes the robust recursive least-squares (RLS) finite impulse response (FIR) filtering algorithm using the covariance information and the robust RLS Wiener FIR filtering algorithm in linear discrete-time stochastic systems with the parameter uncertainties. The observation and system matrices contain the uncertain parameters. The uncertain parameters cause the degraded signal. Theorem 2 proposes the robust RLS FIR filter using the covariance information of the state vector for the degraded signal, the cross-covariance information of the state vector for the signal with the state vector for the degraded signal, the observation matrices for the signal and the degraded signal, and the variance of the white observation noise. Here, it is assumed that the signal and the degraded signal are fitted to the finite-order autoregressive (AR) models. Theorem 3 proposes the robust RLS Wiener FIR filter. The robust RLS Wiener FIR filtering algorithm uses the system and observation matrices for the signal and the degraded signal, the variance of the state vector for the degraded signal, the cross-variance function of the state vector for the signal with the state vector for the degraded signal, and the variance of the white observation noise.

Robust RLS Wiener FIR filter; covariance information; Wiener- Hopf equation; uncertain parameters; degraded signal.

[1] A. H. Jazwinski, Limited memory optimal filtering. IEEE Trans. Automatic Control, vol.13, no.10, pp.558–563, 1968.

[2] W. H. Kwon and S. Han, Receding Horizon Control: Model Predictive Control for State Models, London, UK: Springer; 2005.

[3] S. Zhao S, Y. S. Shmaliy, P. Shi and C. K. Ahn, Fusion Kalman/UFIR filter for state estimation with uncertain parameters and noise statistics. IEEE Trans. Industrial Electronics, vol.64, no.4, pp.3075–3083, 2017.

[4] S. Zhao, Y. S. Shmaliy and F. Liu, Fast Kalman-like optimal unbiased FIR filtering with applications, IEEE Trans. Signal Processing, vol.64, no.9, pp.2284–97, 2016.

[5] S. Zhao, Y. S. Shmaliy and F. Liu, Fast Kalman-like optimal FIR filter for time-variant systems with improved robustness, ISA Transactions, Volume 80, pp.160–168, 2018.

[6] J. M. Pak, C. K. Ahn, Y. H. Mo and M. T. Lim, Moon Kyou Song, Maximum likelihood FIR filter for visual object tracking, Neurocomputing, 216, pp.543–553, 2016.

[7] X. Wang, W. Liu and Z. Deng, Robust weighted fusion Kalman estimators for systems with multiplicative noises, missing measurements and uncertain-variance linearly correlated white noises, Aerospace Science and Technology, vol.68, pp.331–344, 2017.

[8] S. Nakamori, Robust RLS Wiener signal estimators for discrete-time stochastic systems with uncertain parameters, Frontiers in Signal Processing, vol.1, no.3, pp.1–18, 2019.

[9] X. Zhu, Y. C. Soh and L. Xie, Design and analysis of discrete-time robust Kalman filters, Automatica, vol.38, no.6, pp.1069–1077, 2002.

[10] S. Nakamori, RLS Wiener FIR predictor and filter based on innovation approach in linear discrete-time stochastic systems, Frontiers in Signal Processing, vol.1, no.2, pp.49–61, 2017.

[11] A. P. Sage and J. L. Melsa, Estimation Theory with Applications to Communications and Control, New York: McGraw-Hill, 1971.