Abstract
A novel regularised algorithm is presented for the adaptive recursive least-squares Ladder-Lattice prediction and filtering. The design of the filter is based on formulating the Lattice section as a special implementation of the factorised recursive least-squares algorithm, assuming the data and the parameter tracking method meet certain specific criteria.
The proposed estimator provides on a linear array of n processors a data throughput comparable with the triangular systolic implementation of the recursive square root information filters on n(n - 1)/2 processors.
Unlike the standard Recursive Least Squares Lattice-Ladder algorithm the proposed regularised estimator parameter tracking technique can be interpreted as introduction of a set of forgetting factors tuned to control the character of the adaptation. The estimator reduces considerably the numerical problems related to overparametrisation or poor data.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Peterka, V. (1981). Bayesian approach to system identification In P. Eykhoff (Ed.) Trends and Progress in System Identification,Pergamon Press, Eindhoven, Netherlands.
Gentleman, W. M. and H. T. Kung (1981). Matrix Triangularisation by Systolic Arrays. Proc. SPIE, Vol. 298, Real Time Signal Processing IV, 19–26.
Mc Whirter, J.G. (1983). Recursive Least Squares minimization using a Systolic Array, Proc. SPIE, Vol. 431, Real Time Signal Processing VI, 105.
Kulhavý, R. (1987). Restricted exponential forgetting in real-time identification. Automatica, 23, 589–600.
Gaston F. M. F. and G. W. Irwin (1989). The systolic approach to information Kalman filtering, Int. Jour. Control, 15, No. 1, 225–228.
Gaston F.M.F., G. W. Irwin and J. G. Mc Whirter (1990). Systolic Square Root Covariance Kalman Filtering, Jour. of VLSI Sig. Processing, No. 2, 37–49.
Ling, F., D. Manolakis, and J.G. Proakis (1986). A recursive modified Gram-Schmidt algorithm for least squares estimation. IEEE Trans., ASSP-34, 829–835.
Bierman, G. J. (1977). Factorization Methods for Discrete Sequential Estimation. Academic Press, New York.
Ljung, L., T. Söderström (1983). Theory and Practice of Recursive Identification. Cambridge, MA, MIT Press.
Hägglund, T. (1983). The problem of forgetting old data in recursive estimation. Proceedings of the IFAC Workshop on Adaptive Systems in Control and Signal Processing, San Francisco, Ca.
Parkum, J.E., Poulsen, N.K. and J. Hoist (1992). Recursive forgetting algorithms. Int. J. Control 55, 109–128.
Kulhavý, R. and M. Kárný (1984). Tracking of Slowly Varying Parameters by Directional Forgetting. Preprints 9th IFAC World Congress, Budapest. Vol.X, 178–183.
Kulhavý, R. (1986). Directional Tracking of Regression-Type Model Parameters. Preprints of the 2nd IFAC Workshop on Adaptive Systems in Control and Signal Processing. Lund, 97–102.
Ling, F., D. Manolakis, and J.G. Proakis (1986). Numerically Robust Least-Squares Lattice-Ladder Algorithms with Direct Update of the Reflection Coefficients IEEE Trans., ASSP-34, 837–845.
Kadlec, J. (1991). A Recursive Gramm-Schmidt Identification with Directional Tracking of Parameters. Preprints of the the 9-th IFAC/IFORS Symposium on Identification and System Parameter Estimation. Budapest, Hungary, 1707–1712.
Kulhavý, R. and M. B. Zarrop (1993). On a general concept of forgetting. To appear in Int. J. Control.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1993 Springer Science+Business Media New York
About this chapter
Cite this chapter
Kadlec, J., Gaston, F.M.F., Irwin, G.W. (1993). Regularised Lattice-Ladder Adaptive Filter. In: Kárný, M., Warwick, K. (eds) Mutual Impact of Computing Power and Control Theory. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-2968-2_18
Download citation
DOI: https://doi.org/10.1007/978-1-4615-2968-2_18
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-6291-3
Online ISBN: 978-1-4615-2968-2
eBook Packages: Springer Book Archive