EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Two-Stage Transformer Framework for Sparse-Array Direction-of-Arrival Estimation via Correlation Vector Recovery

Wenchao He iD, Yiran Shi iD, Hongxi Zhao iD, Hongliang Zhu, Chunshan Bao

Sensors · 15.05.2026 · 26 (10) · 3132

DOI: 10.3390/s26103132 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 15.05.2026. Verfasst von Wenchao He und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Mechanical and Electrical Engineering, Changchun Humanities and Sciences College, Changchun 130117, China. Die Arbeit nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate direction-of-arrival (DOA) estimation with high resolution is fundamental to many array sensing applications. In practice, however, sparse arrays with missing sensors and snapshot-limited observations often lead to incomplete and noisy second-order statistics, which substantially degrades the performance of conventional eigendecomposition-based estimators. In this paper, we propose a two-stage Transformer framework for sparse-array DOA estimation that explicitly separates correlation recovery from angle inference. The first stage operates in the correlation domain and learns to reconstruct a clean and complete correlation vector from partially observed measurements using masking-aware tokenization and global-context modeling. The recovered representation can be further converted into a structured covariance matrix, providing an interpretable interface to classical signal processing back-ends. Based on the recovered features, the second stage adopts a Transformer regressor to directly predict multi-source DOAs. Extensive simulations on a large-scale dataset with SNRs from −5 to 10 dB and various snapshot numbers demonstrate that the proposed method delivers robust accuracy and improved stability in low-SNR and snapshot-limited regimes, while maintaining competitive performance at higher SNRs. Additional evaluations with an ESPRIT back-end further confirm that the recovery-based covariance yields more reliable DOA estimation than conventional difference–coarray processing, with particularly evident gains under challenging noise conditions.

Schlagwörter

Stichwörter der Autor:innen

covariance completiondirection-of-arrival estimationsparse arraytransformer

Autor:innen und Institutionen

  • School of Mechanical and Electrical Engineering, Changchun Humanities and Sciences College, Changchun 130117, China.
  • College of Communications Engineering, Jilin University, Changchun 130015, China.

ORCID-Profile: Wenchao He, Yiran Shi, Hongxi Zhao

Zitieren

APA

He, W., Shi, Y., Zhao, H., Zhu, H., & Bao, C. (2026). A Two-Stage Transformer Framework for Sparse-Array Direction-of-Arrival Estimation via Correlation Vector Recovery. Sensors, 26(10), 3132. https://doi.org/10.3390/s26103132

Vancouver

He W, Shi Y, Zhao H, Zhu H, Bao C. A Two-Stage Transformer Framework for Sparse-Array Direction-of-Arrival Estimation via Correlation Vector Recovery. Sensors. 2026;26(10):3132. doi:10.3390/s26103132

BibTeX
@article{He2026,
  author = {He, Wenchao and Shi, Yiran and Zhao, Hongxi and Zhu, Hongliang and Bao, Chunshan},
  title = {A Two-Stage Transformer Framework for Sparse-Array Direction-of-Arrival Estimation via Correlation Vector Recovery},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {10},
  pages = {3132},
  doi = {10.3390/s26103132}
}
Kostenlos starten →