PARTICLE FILTERING FOR HIGHDIMENSIONAL REACTION DIFFUSION PDE STATE ESTIMATION UNDER NON-GAUSSIAN NOISE
Scientific works/Elmi eserler · 22.07.2026 · 914
DOI: 10.58225/mpmma.2026.914 ↗Kurz gefasst
Fachartikel in der Zeitschrift Scientific works/Elmi eserler, online erschienen am 22.06.2026. Verfasst von Tahir Gadjiev und 1 weiteren Autor:innen. Die Arbeit nennt 12 Literaturangaben.
Abstract (Originalsprache)
This paper presents a robust Bayesian filtering framework for state estimation in high-dimensional nonlinear partial differential equations governed by reaction–diffusion dynamics. A Reduced-Order Robust Particle Filter (RORPF) is proposed, combining Proper Orthogonal Decomposition (POD)-based model reduction with a mixture-model robust likelihood to address the limitations of conventional filters under sparse observations, strong nonlinearities, and heavy-tailed noise. By projecting the system onto a low-dimensional subspace, the proposed method reduces particle degeneracy and alleviates the curse of dimensionality while improving robustness to outliers. Numerical experiments demonstrate that RORPF achieves accuracy comparable to the Extended Kalman Filter in low-dimensional problems and significantly outperforms classical Particle Filters in high-dimensional settings, providing enhanced stability, higher estimation accuracy, and up to an 18-fold reduction in computational cost.
Zitieren
APA
Gadjiev, T., & Hajiyev, Z. (2026). PARTICLE FILTERING FOR HIGHDIMENSIONAL REACTION DIFFUSION PDE STATE ESTIMATION UNDER NON-GAUSSIAN NOISE. Scientific works/Elmi eserler, 914. https://doi.org/10.58225/mpmma.2026.914
Vancouver
Gadjiev T, Hajiyev Z. PARTICLE FILTERING FOR HIGHDIMENSIONAL REACTION DIFFUSION PDE STATE ESTIMATION UNDER NON-GAUSSIAN NOISE. Scientific works/Elmi eserler. 2026:914. doi:10.58225/mpmma.2026.914
BibTeX
@article{Gadjiev2026,
author = {Gadjiev, Tahir and Hajiyev, Zaur},
title = {PARTICLE FILTERING FOR HIGHDIMENSIONAL REACTION DIFFUSION PDE STATE ESTIMATION UNDER NON-GAUSSIAN NOISE},
journal = {Scientific works/Elmi eserler},
year = {2026},
pages = {914},
doi = {10.58225/mpmma.2026.914}
}
EUVIMED