Second order Extended Ensemble Kalman Filter with Stochastically Perturbed Innovation

Authors

  • Cavin Oyugi Ongere Department of Mathematics, Masinde Muliro University of Science and Technology, P. O. Box 190-50100, Kakamega, Kenya
  • David Angwenyi Department of Mathematics, Masinde Muliro University of Science and Technology, P. O. Box 190-50100, Kakamega, Kenya https://orcid.org/0000-0002-6958-2817
  • Robert Oryiema Department of Mathematics, Masinde Muliro University of Science and Technology, P. O. Box 190-50100, Kakamega, Kenya

DOI:

https://doi.org/10.51867/scimundi.mathematics.4.2.20

Keywords:

Non-Linaer Filtering, Non-Linear State Space Dynamic Models, Bayesian Methods

Abstract

Studies have shown several forms of non-linear dynamic filters. However, Extended Kalman filters have proved to provide more accurate values of the state of dynamic systems over period of time. Though, the results of estimation by use extended Kalman filters are accurate, there is involvement of computation of high dimension covariance matrix that are very expensive. Although Bayesian methods offer a robust and accurate approach, they are often hindered by the computational complexity involved in computing high-dimensional matrices. This study introduces a new filter, the Second Order Extended Ensemble Filter with pertubed innovation (SoEEFPI), designed to numerically address the inversion of high-dimensional covariance matrices and then stochastically perturbing the innovation. The SoEEFPI is derived from the numerical expansion of the expected values of non-linear terms in the stochastically perturbed Kushner-Stratonovich equation, utilizing a second- order Taylor series expansion. Validation of the SoEEFPI is conducted on a three-dimensional stochastic Lorenz 63 model, with simulations performed using MATLAB software. In the validation process , SoEEKFPII is compared with First Order Extended Ensemble Filter (FoEEF), First Order Extended Kalman Bucy Filter (FoEKBF), Second order Extended Ensemble Filter (SoEEF), Bootstrap Particle Filter, and Second Order Extended Kalman Bucy Filter (SoEKBF). Results indicated that SoEEFPI outperformed the other filters (KBF, FoEEF, SoEEF) across all three variables of the Lorenz 63 model: x1, x2 and x3. While SoEKBF exhibited the lowest root mean square error (RMSE), its computational cost is significantly higher due to the integration of high-dimensional covariance, making SoEEFPI a more desirable option since its covariance computation is performed empirically.

Downloads

Download data is not yet available.

References

Angwenyi, D. (2019). Time-continuous state and parameter estimation with application to hyperbolic SPDEs, (Doctoral dissertation, Universita¨ t Potsdam).

Cohen, S. N., & Elliott, R. J. (2015). Stochastic calculus and applications (Vol. 2), New York: Birkha¨ user. https://doi.org/10.1007/978-1-4939-2867-5 DOI: https://doi.org/10.1007/978-1-4939-2867-5

Evensen, Geir (2003) The Ensemble Kalman Filter: theoretical formulation and practical implementation, Journal of Ocean Dynamics, volume:53, Serial:1616-7228, https://doi.org/10.1007/s10236-003-0036-9 https://doi.org/10.1007/s10236-003-0036-9 DOI: https://doi.org/10.1007/s10236-003-0036-9

Goodliff, M., Fletcher, S., Kliewer, A., Forsythe, J., & Jones, A. (2020). Detection of non-Gaussian behavior using machine learning techniques: a case study on the Lorenz 63 model, Journal of Geophysical Research: Atmospheres, 125(2), e2019JD031551.

https://doi.org/10.1029/2019JD031551 DOI: https://doi.org/10.1029/2019JD031551

Hazazi, Muhammad Asaduddin and Sihabuddin, Agus(2019)Extended Kalman filter in recurrent neural network: USDIDR forecasting case study, Indonesian Journal of Computing and Cybernetics Systems, 13. https://doi.org/10.22146/ijccs.47802 DOI: https://doi.org/10.22146/ijccs.47802

Itoˆ, K. (1944). 109. Stochastic Integral. Proceedings of the Imperial Academy, 20(8), 519-524.

Lamberton, D., & Lapeyre, B. (2011). Introduction to stochastic calculus applied to finance. Chapman and Hall/CRC.

https://doi.org/10.1201/9781420009941 DOI: https://doi.org/10.1201/9781420009941

Marron, M and Garcia, JC and Sotelo, MA and Cabello, M and Pizarro, D and Huerta, F and Cerro, J(2007)Comparing a Kalman filter and a particle filter in a multiple objects tracking application, IEEE International Symposium on Intelligent Signal Processing.

https://doi.org/10.1109/WISP.2007.4447520 DOI: https://doi.org/10.1109/WISP.2007.4447520

Midenyo, K., Angwenyi, D., & Oganga, D. (2024). Second Order Extended Ensemble Filter for Non-linear Filtering.

https://doi.org/10.51867/ajernet.mathematics.5.4.25 DOI: https://doi.org/10.51867/ajernet.mathematics.5.4.25

Neelima, M., & Prabha, I. S. (2024). Optimized deep network based spoof detection in automatic speaker verification system, Multimedia Tools and Applications, 83(5), 13073-13091. https://doi.org/10.1007/s11042-023-16127-w DOI: https://doi.org/10.1007/s11042-023-16127-w

Oryiema, Robert and David, Angwenyi and Kevin Midenyo (2021)Extended Ensemble Filter for High- dimensional Nonlinear State Space Models, Journal of Advances in Mathematics and Computer Science, 36. https://doi.org/10.9734/jamcs/2021/v36i530365 DOI: https://doi.org/10.9734/jamcs/2021/v36i530365

Robert, Oryiema and Angwenyi, David and Achiles, Nyogesa and On'gala, Jacob[2022]Initialization and Estimation of Weights and Bias using Bayesian Technique, Asian Journal of Probability and Statistics, 17. https://doi.org/10.9734/ajpas/2022/v17i230420 DOI: https://doi.org/10.9734/ajpas/2022/v17i230420

Sen, N., & Caines, P. E. (2016). Nonlinear filtering theory for McKean-Vlasov type stochastic differential equations, SIAM Journal on Control and Optimization, 54(1), 153-174. https://doi.org/10.1137/15M1013304 DOI: https://doi.org/10.1137/15M1013304

Downloads

Published

2024-11-22

How to Cite

Ongere, C. O., Angwenyi, D., & Oryiema, R. (2024). Second order Extended Ensemble Kalman Filter with Stochastically Perturbed Innovation. SCIENCE MUNDI, 4(2), 222–242. https://doi.org/10.51867/scimundi.mathematics.4.2.20