Consistent estimation of autoregressive parameters from noisy observations based on two interacting Kalman filters

The estimation of the parameters of an autoregressive process (AR) from noisy observations is still a challenging problem. In this paper, we propose to sequentially estimate both the signal and the parameters, avoiding a non-linear approach such as the extended Kalman filter. The method is based on two conditionally linked Kalman filters running in parallel. Once a new observation is available, the first filter uses the latest estimated AR parameters to estimate the signal, while the second filter uses the estimated signal to update the AR parameters. This approach can be viewed as a recursive instrumental variable-based method and hence has the advantage of providing consistent estimates of the parameters from noisy observations. A comparative study with existing algorithms illustrates the performances of the proposed method when the additive noise is either white or coloured.

[1]  Petre Stoica,et al.  Adaptive instrumental variable method for robust direction-of-arrival estimation , 1995 .

[2]  Steven Kay,et al.  Modern Spectral Estimation: Theory and Application , 1988 .

[3]  J. Dripps,et al.  Autoregressive spectral estimation of fetal breathing movement , 1989, IEEE Transactions on Biomedical Engineering.

[4]  Wei Xing Zheng.  Fast identification of autoregressive signals from noisy observations , 2005, IEEE Trans. Circuits Syst. II Express Briefs.

[5]  Wei Xing Zheng.  Autoregressive parameter estimation from noisy data , 2000 .

[6]  Steven Kay,et al.  The effects of noise on the autoregressive spectral estimator , 1979 .

[7]  Mohamed Najim,et al.  Speech enhancement as a realisation issue , 2002, Signal Process..

[8]  Mohamed Najim,et al.  A single microphone Kalman filter-based noise canceller , 1999, IEEE Signal Processing Letters.

[9]  Kwan Wong,et al.  Identification of linear discrete time systems using the instrumental variable method , 1967, IEEE Transactions on Automatic Control.

[10]  Wei Xing Zheng.  Unbiased identification of autoregressive signals observed in colored noise , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[11]  Gerhard Doblinger,et al.  Smoothing of noisy AR signals using an adaptive Kalman filter , 1998, 9th European Signal Processing Conference (EUSIPCO 1998).

[12]  P. Young.  An instrumental variable method for real-time identification of a noisy process , 1970 .

[13]  Carlos E. Davila,et al.  On the noise-compensated Yule-Walker equations , 2001, IEEE Trans. Signal Process..

[14]  Toshio Yoshimura,et al.  A modified extended Kalman filter for linear discrete-time systems with unknown parameters , 1981, Autom..

[15]  Petre Stoica,et al.  Instrumental Variable Methods for Arma Parameter Estimation , 1985 .

[16]  Inan Güler,et al.  AR spectral analysis of EEG signals by using maximum likelihood estimation , 2001, Comput. Biol. Medicine.

[17]  Otto Opitz,et al.  Classification and Data Analysis , 1999 .

[18]  Ehud Weinstein,et al.  Iterative and sequential Kalman filter-based speech enhancement algorithms , 1998, IEEE Trans. Speech Audio Process..

[19]  E. Stear,et al.  The simultaneous on-line estimation of parameters and states in linear systems , 1976 .

[20]  J. Treichler,et al.  Transient and convergent behavior of the adaptive line enhancer , 1979 .

[21]  Carlos E. Davila.  A subspace approach to estimation of autoregressive parameters from noisy measurements , 1998, IEEE Trans. Signal Process..

[22]  B. Friedlander.  Instrumental variable methods for ARMA spectral estimation , 1983 .

[23]  L. Marple.  A new autoregressive spectrum analysis algorithm , 1980 .

[24]  Charles W. Therrien,et al.  Discrete Random Signals and Statistical Signal Processing , 1992 .

[25]  Mohamed A. Deriche.  AR parameter estimation from noisy data using the EM algorithm , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.

[26]  Ehud Weinstein,et al.  Single-sensor active noise cancellation , 1994, IEEE Trans. Speech Audio Process..

[27]  C. Braun,et al.  Adaptive AR modeling of nonstationary time series by means of Kalman filtering , 1998, IEEE Transactions on Biomedical Engineering.

[28]  W. M. Carey,et al.  Digital spectral analysis: with applications , 1986 .

[29]  Wen-Rong Wu,et al.  Adaptive AR modeling in white Gaussian noise , 1997, IEEE Trans. Signal Process..

[30]  Md. Jahangir Hossain,et al.  Parameter estimation of multichannel autoregressive processes in noise , 2003, Signal Process..