EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Clock Asynchronism Calibration in Bistatic Sensing Systems via Expert‐Guided Cross‐Attention Mechanism

Haitao Zhao, Zhennan Qin, Wenchao Xia, Chun Zhu, Yadong Wang

IET Communications · 2026-01 · 20 (1)

DOI: 10.1049/cmu2.70186 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IET Communications, online erschienen am 06.07.2026. Verfasst von Haitao Zhao und 4 weiteren Autor:innen. Die Arbeit nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Major Project, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the rapid evolution of wireless communication networks, it has become increasingly desirable for base stations to provide not only data transmission but also environmental perception capabilities. To meet such dual demands, integrated sensing technologies have attracted growing attention, among which bistatic sensing provides enhanced target detection through bistatic geometric configuration. However, bistatic sensing performance critically depends on the sampling alignment between transceivers. In practice, frequency discrepancies between crystal oscillators lead to clock asynchronism, causing sampling misalignment and inter‐symbol aliasing, thereby inducing systemic biases in target localization accuracy. To address this issue, this paper establishes a clock‐asynchronism sensing model that incorporates inter‐symbol aliasing effects and reveals how sampling shifts distort range measurements without affecting angle estimation. A cross‐attention‐based neural network is further developed to accurately estimate the sampling offsets by fusing expert knowledge derived from cross‐correlation peak shifts with raw signal features. Simulation results demonstrate that the proposed expert‐guided cross‐attention network (ECA‐Net) achieves an exact‐match offset prediction accuracy of 96.3% and a remarkably low offset prediction mean absolute error (MAE) of 0.08 over a signal‐to‐noise ratio (SNR) range of −5 to 10 dB. Furthermore, the implementation of cyclic‐shift compensation effectively diminishes the localization error to a sub‐meter level in bistatic sensing scenarios.

Förderung

  • National Science and Technology Major Project
  • National Natural Science Foundation of China

Zitieren

APA

Zhao, H., Qin, Z., Xia, W., Zhu, C., & Wang, Y. (2026). Clock Asynchronism Calibration in Bistatic Sensing Systems via Expert‐Guided Cross‐Attention Mechanism. IET Communications, 20(1). https://doi.org/10.1049/cmu2.70186

Vancouver

Zhao H, Qin Z, Xia W, Zhu C, Wang Y. Clock Asynchronism Calibration in Bistatic Sensing Systems via Expert‐Guided Cross‐Attention Mechanism. IET Communications. 2026;20(1). doi:10.1049/cmu2.70186

BibTeX
@article{Zhao2026,
  author = {Zhao, Haitao and Qin, Zhennan and Xia, Wenchao and Zhu, Chun and Wang, Yadong},
  title = {Clock Asynchronism Calibration in Bistatic Sensing Systems via Expert‐Guided Cross‐Attention Mechanism},
  journal = {IET Communications},
  year = {2026},
  volume = {20},
  number = {1},
  doi = {10.1049/cmu2.70186}
}
Kostenlos starten →