Deep Unfolding-Based Sensing-Assisted Channel Estimation With Imperfect Radar Arrays

Page view(s)
0
Checked on
Deep Unfolding-Based Sensing-Assisted Channel Estimation With Imperfect Radar Arrays
Title:
Deep Unfolding-Based Sensing-Assisted Channel Estimation With Imperfect Radar Arrays
Journal Title:
IEEE Journal on Selected Areas in Communications
Keywords:
Publication Date:
16 September 2025
Citation:
Yang, J., Ai, B., Chen, W., Yang, S., Wang, N., & Yuen, C. (2026). Deep Unfolding-Based Sensing-Assisted Channel Estimation With Imperfect Radar Arrays. IEEE Journal on Selected Areas in Communications, 44, 401–415. https://doi.org/10.1109/jsac.2025.3610425
Abstract:
In vehicle-to-everything (V2X) scenarios, the high dynamic characteristics of V2X environments impose significant challenges on communication channel estimation, where the emerging integrated sensing and communication technology could serve as a vital tool for achieving accurate channel estimation. This paper leverages radar-sensed angle information to assist in communication channel estimation and proposes a deep unfolding-based radar-assisted channel estimation network (Radar-CEnet). Specifically, for the radar module, to address the challenges posed by insufficient data in imperfect arrays, we employ a model-agnostic meta-learning with a convolutional neural network (MAML-CNN) approach to achieve high-precision direction-of-arrival (DOA) estimation. Then, the angle information obtained by the radar module, as prior knowledge, is used for channel estimation. Building on this, we design a novel soft-thresholding shrinkage function and propose the Radar-CEnet algorithm to efficiently estimate the sparse channel. Finally, we rigorously prove the convergence of the Radar-CEnet algorithm and demonstrate that it achieves a lower estimation error. Experimental results show that the proposed Radar-CEnet outperforms existing traditional methods and deep learning-based approaches in channel estimation performance. At an SNR of 20dB, the proposed Radar-CEnet method reduces the NMSE from –23.75 dB to –27.15 dB compared to the learning-based iterative soft-thresholding method, achieving an estimation accuracy improvement of approximately 54%.
License type:
Publisher Copyright
Funding Info:
This research / project is supported by the MOE and MOF, China - Fundamental Research Funds for the Central Universities
Grant Reference no. : 2022JBQY004

This research / project is supported by the Natural Science Foundation of China - NA
Grant Reference no. : 62221001, U2468201, and W2421083

This research / project is supported by the State Scholarship Fund, China Scholarship Council - NA
Grant Reference no. : 202307090109

This research / project is supported by the Agency for Science, Technology and Research Singapore - Manufacturing, Trade, and Connectivity Programmatic
Grant Reference no. : M22L1b0110

This research / project is supported by the Ministry of Education Singapore - Academic Research Fund Tier 2
Grant Reference no. : T2EP50124-0032

This research / project is supported by the National Research Foundation - Future Communications Research and Development Programme (FCP)
Grant Reference no. : FCP-NTURG- 2024-025
Description:
© 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
ISSN:
1558-0008
0733-8716
Files uploaded:

File Size Format Action
5.pdf 650.68 KB PDF Open