Tensor network Kalman filter for LTI systems

Tensor network Kalman filter for LTI systems

Gedon, Daniel (Student TU Delft)
Piscaer, P.J. (TU Delft Team Raf Van de Plas)
Batselier, K. (TU Delft Team Jan-Willem van Wingerden)
Smith, C.S. (TU Delft Team Raf Van de Plas)
Verhaegen, M.H.G. (TU Delft Team Raf Van de Plas)

2019

An extension of the Tensor Network (TN) Kalman filter [2], [3] for large scale LTI systems is presented in this paper. The TN Kalman filter can handle exponentially large state vectors without constructing them explicitly. In order to have efficient algebraic operations, a low TN rank is required. We exploit the possibility to approximate the covariance matrix as a TN with a low TN rank. This reduces the computational complexity for general SISO and MIMO LTI systems with TN rank greater than one significantly while obtaining an accurate estimation. Improvements of this method in terms of computational complexity compared to the conventional Kalman filter are demonstrated in numerical simulations for large scale systems.

Curse of dimensionality
Kalman filter
Large scale systems
LTI systems
MIMO
SISO
Tensor train
Tensors

http://resolver.tudelft.nl/uuid:945b282c-9d21-4877-ba24-e346f2a5891c

https://doi.org/10.23919/EUSIPCO.2019.8902976

IEEE, Piscataway, NJ, USA

2020-05-18

978-9-0827-9703-9

Proceedings of the 27th European Signal Processing Conference (EUSIPCO 2019)

27th European Signal Processing Conference, EUSIPCO 2019, 2019-09-02 → 2019-09-06, A Coruna, Spain

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Institutional Repository

conference paper

© 2019 Daniel Gedon, P.J. Piscaer, K. Batselier, C.S. Smith, M.H.G. Verhaegen