DFRFT-Based Adaptive Channel Estimation for XL-MIMO Hybrid-Field Communications
IEEE Transactions on Communications · 2026 · 74 · 15769-15785
DOI: 10.1109/tcomm.2026.3742866 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Communications, 2026. Verfasst von Ruisong Weng und 6 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Gefördert von: Key Research and Development Projects, National Natural Science Foundation of China, Natural Science Foundation of China.
Abstract (Originalsprache)
With the evolution of future wireless communication systems toward large antenna arrays and high-frequency bands, near-field effects have become increasingly prominent. These effects cause traditional discrete Fourier transform (DFT)-based techniques to suffer from severe performance degradation, owing to quadratic phase terms and energy spread. Existing polar-domain codebooks face issues of non-orthogonality and high training overhead. To address these challenges, this paper proposes a hybrid-field channel estimation strategy based on the discrete fractional Fourier transform (DFRFT), which is applicable to both the sampling-type DFRFT and the eigenvector decomposition-type DFRFT. Specifically, we first verify that the near-field channel demonstrates significant sparsity in the DFRFT domain, derive the optimal DFRFT order and analyze the respective advantages of the sampling-type DFRFT and the eigenvector decomposition-type DFRFT. Then, a DFRFT-based adaptive dictionary orthogonal matching pursuit (D-ADOMP) algorithm is designed, where the DFRFT order is used for adaptive dictionary construction rather than explicit angle-distance parameter reconstruction. This algorithm iteratively determines the appropriate DFRFT order through a two-stage search and directly performs sparse recovery over the corresponding DFRFT-domain sensing dictionary, enabling the sequential separation of multipath components for high-precision channel reconstruction. Notably, the DFT is inherently a special case of the DFRFT when its order equals one, which allows the proposed scheme to naturally reduce to the DFT representation when the DFT-domain sparsity is sufficiently recovered. Moreover, we derive an effective near-field boundary from the perspective of DFT-domain sparsity, which characterizes when the near-field quadratic phase term causes noticeable DFT-domain energy spreading and identifies the region where DFRFT can provide a visible sparsity gain over DFT. Simulation results validate the effectiveness of the proposed algorithm, which exhibits superior performance to existing methods. Furthermore, the results demonstrate that the proposed scheme achieves favorable estimation accuracy with acceptable computational complexity and offers a feasible solution for 6G extremely large-scale multiple-input multiple-output (XL-MIMO) communications.
Förderung
- Key Research and Development Projects
- National Natural Science Foundation of China
- Natural Science Foundation of China
Zitieren
APA
Weng, R., Pan, C., Ren, H., Wang, R., Chen, X., Liu, W., & Wang, J. (2026). DFRFT-Based Adaptive Channel Estimation for XL-MIMO Hybrid-Field Communications. IEEE Transactions on Communications, 74, 15769-15785. https://doi.org/10.1109/tcomm.2026.3742866
Vancouver
Weng R, Pan C, Ren H, Wang R, Chen X, Liu W, et al. DFRFT-Based Adaptive Channel Estimation for XL-MIMO Hybrid-Field Communications. IEEE Transactions on Communications. 2026;74:15769-15785. doi:10.1109/tcomm.2026.3742866
BibTeX
@article{Weng2026,
author = {Weng, Ruisong and Pan, Cunhua and Ren, Hong and Wang, Ruizhe and Chen, Xianzhe and Liu, Wang and Wang, Jiangzhou},
title = {DFRFT-Based Adaptive Channel Estimation for XL-MIMO Hybrid-Field Communications},
journal = {IEEE Transactions on Communications},
year = {2026},
volume = {74},
pages = {15769--15785},
doi = {10.1109/tcomm.2026.3742866}
}
EUVIMED