EUVIMEDEUVIMED entdecken
Crossref

Time-Varying Channel Modeling and Tracking for UAV Near-Field Communications

Huizhu Han, Wei Liu, Jinkun Zhu, Xingwang Li, Jing Lei

IEEE Transactions on Communications · 2026 · 74 · 16093-16106

DOI: 10.1109/tcomm.2026.3734294 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Communications, 2026. Verfasst von Huizhu Han und 4 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Gefördert von: Youth Independent Innovation Fund of the National University of Defense Technology, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the use of higher frequency bands and extremely large-scale multiple-input multiple-output arrays, uncrewed aerial vehicle (UAV) communication links may operate in the near-field region, where spherical-wave propagation and platform motion lead to complex channel variations. This paper investigates the modeling and tracking of UAV-assisted near-field time-varying channels. First, a spherical-wave channel model is established to characterize the effects of UAV motion and attitude variation on the array-path geometry. Then, a subarray-based Doppler model is developed, in which the line-of-sight Doppler shift varies across subarrays, whereas the non-line-of-sight components exhibit path-dependent Doppler shifts. Based on this model, a subarray-based reduced-dimensional Kalman filtering (KF) framework is developed for low-complexity channel tracking. A reduced-dimensional dictionary-domain state-space model is constructed, and an expectation–maximization framework incorporating KF and Rauch–Tung–Striebel smoothing is employed to estimate the channel evolution parameters from pilot observations, followed by recursive KF tracking using the learned model. Building on this reduced-dimensional KF framework, a KalmanNet-enhanced tracking method is further developed to improve tracking accuracy and robustness against process-noise covariance mismatch. By retaining the learned state-transition model and the known pilot observation model, the KalmanNet-enhanced tracker employs a recurrent neural network to estimate the Kalman gain from innovation and state-update features, thereby reducing its sensitivity to inaccurate process-noise statistics. Simulation results demonstrate the effectiveness of the proposed reduced-dimensional tracking framework and show that KalmanNet further enhances tracking accuracy and robustness against process-noise mismatch.

Förderung

  • Youth Independent Innovation Fund of the National University of Defense Technology
  • National Natural Science Foundation of China

Zitieren

APA

Han, H., Liu, W., Zhu, J., Li, X., & Lei, J. (2026). Time-Varying Channel Modeling and Tracking for UAV Near-Field Communications. IEEE Transactions on Communications, 74, 16093-16106. https://doi.org/10.1109/tcomm.2026.3734294

Vancouver

Han H, Liu W, Zhu J, Li X, Lei J. Time-Varying Channel Modeling and Tracking for UAV Near-Field Communications. IEEE Transactions on Communications. 2026;74:16093-16106. doi:10.1109/tcomm.2026.3734294

BibTeX
@article{Han2026,
  author = {Han, Huizhu and Liu, Wei and Zhu, Jinkun and Li, Xingwang and Lei, Jing},
  title = {Time-Varying Channel Modeling and Tracking for UAV Near-Field Communications},
  journal = {IEEE Transactions on Communications},
  year = {2026},
  volume = {74},
  pages = {16093--16106},
  doi = {10.1109/tcomm.2026.3734294}
}
Kostenlos starten →