@article{QING2026, 
author = {Chaojin QING and Zhiying LIU and Wenquan HU and Liang ZHAO and Mian YE},
title = {Environmental sensing-assisted off-grid channel estimation in UAV-enabled OTFS-ISAC systems},
year = {2026},
journal = {Chinese Journal of Aeronautics},
volume = {39},
number = {7},
keywords = {Channel estimation, Delay-Doppler domain, Environmental sensing, Off-grid, Orthogonal time frequency space},
url = {https://www.sciopen.com/article/10.1016/j.cja.2025.103923},
doi = {10.1016/j.cja.2025.103923},
abstract = {Unmanned Aerial Vehicle (UAV)-enabled Orthogonal Time Frequency Space (OTFS)-Integrated Sensing And Communication (ISAC) is a promising technique that simultaneously provides the enhanced Sensing and Communication (S&C) services, while holding significant potential to mitigate the high-Doppler effect. To achieve the advantages of UAV-enabled OTFS-ISAC systems, accurate channel estimation is crucial. However, this process faces a significant challenge of the off-grid phenomenon in the Delay-Doppler (DD) domain, promoting the development of off-grid channel estimation methods. Regrettably, most existing off-grid channel estimation methods mainly focus on optimizing signal modeling, resulting in significantly high computational complexity and leaving substantial room for accuracy improvement. To address these issues, inspired by the benefits of environmental sensing, a sensing-assisted off-grid channel estimation method is proposed in this paper. Specifically, passive UAV radar is employed to sense the location and velocity of ground User Equipment (gUE). Based on Ground-to-Air (G2A) communication link, a radio map is constructed, which captures comprehensive environmental features and signal blockage information. We extract the delay and Doppler priors from this environmental information, which provide crucial insights in DD domain channel from a perspective of propagation environment. Subsequently, by leveraging this prior information, a Sensing-Assisted Grid Calibration-based OGSBI (SAGCSBI) method and a Path Refinement-based Enhanced Channel Estimation (PR-EnCE) method are developed. The proposed methods reduce off-grid errors and mitigate path superposition by integrating environmental information and signal modeling. Simulation results verify the superior performance of the proposed channel estimation methods over state-of-the-art methods and demonstrate the robustness against parameter variations.}
}