EUVIMEDEUVIMED entdecken
Crossref

Gaussian–Student’s t Mixture-Based Robust Probability Hypothesis Density Filter Under Heavy-Tailed Process and Measurement Noises

Qiutiao Xue, Guisheng Liao, Han Zou, Sunyong Wu

IEEE Sensors Journal · 2026-11 · 26 (21) · 31731-31745

DOI: 10.1109/jsen.2025.3620134 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Sensors Journal, 2026-11. Verfasst von Qiutiao Xue und 3 weiteren Autor:innen. Die Arbeit nennt 48 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Central Guided Local Science and Technology Development Funds Project, Guangxi Key Research and Development Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In multitarget tracking (MTT), the presence of heavy-tailed noise can significantly degrade tracking accuracy, which is critical in integrated sensing and communication (ISAC) systems. To address this problem, we propose a novel robust probabilistic hypothesis density (PHD) filter based on a Gaussian–Student’s t mixture distribution (GSTMD). The proposed algorithm models noise as a GSTMD, which is a linear combination of Gaussian and Student’s t distributions. By varying the parameters of this distribution, it can adapt to different heavy-tailed noises. However, selecting appropriate prior parameters for the distribution is often challenging. Therefore, we introduce the variational Bayesian (VB) method to adjust the parameters of the GSTMD in real time, deriving the posterior distribution of the distribution’s parameters and the marginal likelihood function. On this basis, we further derive the closed recursive formulation of PHD filtering based on the GSTMD. Simulation results show that the proposed algorithm can effectively improve the MTT accuracy in heavy-tailed noise scenarios.

Förderung

  • National Natural Science Foundation of China
  • Central Guided Local Science and Technology Development Funds Project
  • Guangxi Key Research and Development Project
  • Open Fund of the Key Laboratory of Data Analysis and Computation of Guangxi Colleges and Universities
  • Open Fund of Guangxi Center for Applied Mathematics
  • Open Fund of the Key Laboratory of the Ministry of Education of the People’s Republic of China for the year of 2021 for the project of “Cognitive Radio and Inf…
  • Innovation Project of GUET Graduate Education

Zitieren

APA

Xue, Q., Liao, G., Zou, H., & Wu, S. (2026). Gaussian–Student’s t Mixture-Based Robust Probability Hypothesis Density Filter Under Heavy-Tailed Process and Measurement Noises. IEEE Sensors Journal, 26(21), 31731-31745. https://doi.org/10.1109/jsen.2025.3620134

Vancouver

Xue Q, Liao G, Zou H, Wu S. Gaussian–Student’s t Mixture-Based Robust Probability Hypothesis Density Filter Under Heavy-Tailed Process and Measurement Noises. IEEE Sensors Journal. 2026;26(21):31731-31745. doi:10.1109/jsen.2025.3620134

BibTeX
@article{Xue2026,
  author = {Xue, Qiutiao and Liao, Guisheng and Zou, Han and Wu, Sunyong},
  title = {Gaussian–Student’s t Mixture-Based Robust Probability Hypothesis Density Filter Under Heavy-Tailed Process and Measurement Noises},
  journal = {IEEE Sensors Journal},
  year = {2026},
  volume = {26},
  number = {21},
  pages = {31731--31745},
  doi = {10.1109/jsen.2025.3620134}
}
Kostenlos starten →