DSCM-enabled microwave-band fiber-wireless-integration fronthaul transmission with CRARNet-based subchannel-wise AMR
Optics Express · 10.08.2026 · 34 (16) · 29812
DOI: 10.1364/oe.606299 ↗Kurz gefasst
Fachartikel in der Zeitschrift Optics Express, online erschienen am 03.08.2026. Verfasst von Jiahao Bi und 12 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben. Gefördert von: National Key Research and Development Program of China, National Natural Science Fund for Excellent Young Scientists Fund Program (Overseas), Beijing Institute of Technology Research Fund Program for Young Scholars.
Abstract (Originalsprache)
With the increasing demand for high-capacity and flexible fronthaul transmission in radio access networks (RANs), the fiber-wireless-integration (FWI) scheme has emerged as a promising transmission framework. To further enhance resource allocation flexibility and heterogeneous service support, digital subcarrier multiplexing (DSCM) can be introduced into FWI systems to simultaneously deliver multiple independent subchannels with different modulation formats. However, the receiver may not have prior knowledge of the transmitter-side modulation assignment, making automatic modulation recognition (AMR) necessary before subsequent modulation-dependent signal recovery. In addition, the joint effects of wireless transmission, electro-optic conversion, fiber delivery, and coherent reception significantly complicate subchannel-wise AMR. To address these challenges, we propose a microwave-band FWI fronthaul transmission scheme integrated with DSCM and a convolutional residual attention recurrent network (CRARNet)-based AMR. In the proposed scheme, the modulation format of each demultiplexed subchannel is automatically identified by the CRARNet-based AMR module using the modulation-format-agnostic preprocessed in-phase/quadrature (I/Q) sequence, which is obtained after a universal, modulation-format-independent digital signal processing (DSP) flow at the receiver. The CRARNet-based AMR module consists of three stages, namely shallow feature extraction, deep feature refinement, and temporal modeling, where multi-scale pre-activated residual blocks (MSPAs) and convolutional block attention modules (CBAMs) are employed to refine discriminative features, and bidirectional long short-term memory (BiLSTM) is employed to model temporal dependencies. The proposed scheme is experimentally demonstrated in a 17.5-GHz microwave-band FWI system over a 2-m wireless link and a 25-km single-mode fiber-28 (SMF-28) link. The transmitted DSCM signal consists of nine independent subchannels, each carrying symbols using quadrature phase-shift keying (QPSK), 8-ary quadrature amplitude modulation (8QAM), or 16-ary quadrature amplitude modulation (16QAM). Experimental results show that the proposed CRARNet-based AMR module achieves the best average probability of correct classification (PoCC) of 99.45% and outperforms representative baseline models under varying received optical power (ROP) conditions while maintaining a favorable accuracy-complexity trade-off. Furthermore, the adopted CRARNet-based AMR module exhibits excellent robustness across varying launched optical power and antenna height conditions. The results demonstrate that the proposed CRARNet-based AMR module provides an effective solution for subchannel-wise AMR in the DSCM-enabled microwave-band FWI fronthaul scheme.
Förderung
- National Key Research and Development Program of China
- National Natural Science Fund for Excellent Young Scientists Fund Program (Overseas)
- Beijing Institute of Technology Research Fund Program for Young Scholars
Zitieren
APA
Bi, J., Fu, C., Fu, Y., Li, X., Xie, T., Hong, Z., Zhang, Y., Chen, C., Li, X., Liu, H., Liu, Y., Zhang, Q., & Xin, X. (2026). DSCM-enabled microwave-band fiber-wireless-integration fronthaul transmission with CRARNet-based subchannel-wise AMR. Optics Express, 34(16), 29812. https://doi.org/10.1364/oe.606299
Vancouver
Bi J, Fu C, Fu Y, Li X, Xie T, Hong Z, et al. DSCM-enabled microwave-band fiber-wireless-integration fronthaul transmission with CRARNet-based subchannel-wise AMR. Optics Express. 2026;34(16):29812. doi:10.1364/oe.606299
BibTeX
@article{Bi2026,
author = {Bi, Jiahao and Fu, Chengang and Fu, Yukai and Li, Xinying and Xie, Tangyao and Hong, Zhixin and Zhang, Yujie and Chen, Chen and Li, Xianshuang and Liu, Haonan and Liu, Yang and Zhang, Qi and Xin, Xiangjun},
title = {DSCM-enabled microwave-band fiber-wireless-integration fronthaul transmission with CRARNet-based subchannel-wise AMR},
journal = {Optics Express},
year = {2026},
volume = {34},
number = {16},
pages = {29812},
doi = {10.1364/oe.606299}
}
EUVIMED