Full-Domain Coupler: A Wireless Native Neural Backbone for Channel Representation and Deduction

Zirui Chen; Ziqing Xing; Zhaoyang Zhang; Hongning Ruan; Yuzhi Yang; Zhaohui Yang; Chongwen Huang; Mérouane Debbah

doi:10.1016/j.eng.2026.09.010

Engineering ›› :202609010 DOI: 10.1016/j.eng.2026.09.010

research-article

Full-Domain Coupler: A Wireless Native Neural Backbone for Channel Representation and Deduction

Zirui Chen ^a^,^b^,^#
, Ziqing Xing ^a^,^b^,^c^,^#
, Zhaoyang Zhang ^a^,^b^,^c^,^*
, Hongning Ruan ^a^,^b
, Yuzhi Yang ^d
, Zhaohui Yang ^a^,^b
, Chongwen Huang ^a^,^b
, Mérouane Debbah ^d

Author information +

History +

PDF (3533KB)

Abstract

Data representation is a fundamental issue in deep learning. However, as wireless data scale and deeply couple across many physical domains such as time, space, and frequency, existing wireless artificial intelligence (AI) technologies lack dedicated representation solutions. Instead, they mainly rely on stitching general-purpose networks, a “tool-driven” paradigm that inevitably results in structural redundancy and bottlenecks in information flow. To fill this gap, this paper proposes Coupler, a wireless native-AI neural backbone designed for the representation learning of channel state information-the pivotal data in wireless systems. Leveraging the physical insights revealed by channel tensors, Coupler decomposes representation learning into individual domains on a layer-by-layer basis and then couples the learned domain-specific features through a dimension-staggered cascade. This full-domain interleaved learning architecture enables superior parameter efficiency and fine-grained multi-domain feature fusion. Based on this backbone, we use complex-domain multilayer perceptrons as spatial and frequency domain learners, while employing three optional mechanisms-convolution, attention, or gating-to capture temporal dependencies. This results in a series of efficient and extremely lightweight channel learning schemes with diverse functionalities, showcasing the compactness, versatility, and flexibility of Coupler. We evaluate these schemes on channel deduction, a general representation task encompassing channel estimation, interpolation, prediction, and feedback. Extensive experimental evaluations validate the significant performance gains and robust applicability, even for real-world measured data, demonstrating Coupler’s potential as a promising basic architecture in the design of wireless foundation models.

Keywords

Coupler / 6G native-AI / Deep learning / Interleaved learning / Channel representation / Channel deduction

Cite this article

Download citation ▾

Zirui Chen, Ziqing Xing, Zhaoyang Zhang, Hongning Ruan, Yuzhi Yang, Zhaohui Yang, Chongwen Huang, Mérouane Debbah. Full-Domain Coupler: A Wireless Native Neural Backbone for Channel Representation and Deduction. Engineering 202609010 DOI:10.1016/j.eng.2026.09.010

登录浏览全文

4963

注册一个新账户忘记密码

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	International Telecommunication Union. Framework and overall objectives of the future development of IMT for 2030 and beyond. ITU-R Recommendation M.2160-0. Report. Geneva: International Telecommunication Union; 2023 Nov.

[2]	Zhang P, Niu K, Wang X, Liu Y, Liang Z, Dong C, et al. ComAI: the convergence of communication and artificial intelligence. IEEE Commun Surv Tutor 2026; 28: 2163-97.

[3]	Zhang P, Xu W, Gao H, Niu K, Xu X, Qin X, et al. Toward wisdom-evolutionary and primitive-concise 6G: a new paradigm of semantic communication networks. Engineering 2022; 8: 60-73.

[4]	Chen Z, Zhang Z, Yang Z . Big AI models for 6G wireless networks: opportunities, challenges, and research directions. IEEE Wirel Commun 2024; 31(5): 164-72.

[5]	Hoydis J, Aoudia FA, Valcarce A, Viswanathan H . Toward a 6G AI-native air interface. IEEE Commun Mag 2021; 59(5): 76-81.

[6]	Zhang P, Niu K, Liu Y, Liang Z, Ma N, Xu X, et al. Way to build native AI-driven 6G air interface: principles, roadmap, and outlook. IEEE Trans Netw Sci Eng 2026; 13: 3551-65.

[7]	Wang X, Lu L, Li Q, Sun Q, Shi N, Chen Z, et al. A task-driven design approach for 6G AI-native architecture. Engineering 2026; 56: 87-103.

[8]	Liu Y, Wang X, Mei J, Boudreau G, Abou-Zeid H, Sediq AB . Situation-aware resource allocation for multi-dimensional intelligent multiple access: a proactive deep learning framework. IEEE J Sel Areas Commun 2021; 39(1): 116-30.

[9]	Qiao Y, Niu Y, Zhang X, Chen S, Zhong Z, Wang N, et al. Intelligent beam management based on deep reinforcement learning in high-speed railway scenarios. IEEE Trans Veh Technol 2024; 73(3): 3917-31.

[10]	Li X, Shi W, Zhang H, Peng C, Wu S, Tong W . The Agentic-AI core: an AI-empowered, mission-oriented core network for next-generation mobile telecommunications. Engineering 2026; 56: 104-19.

[11]	Chen Z, Zhang Z, Liu C, Xing Z . Towards wireless native big AI model: the mission and approach differ from large language model. Sci China Inf Sci 2025; 68(7): 170303.

[12]	Zhang J, Yu L, Liu S, Cai Y, Zhang Y, Xing H, et al. Wireless environmental information theory: a new paradigm toward 6G online and proactive environment intelligence communication. Engineering 2026; 56: 186-200.

[13]	Qi C, Dong P, Ma W, Zhang H, Zhang Z, Li GY . Acquisition of channel state information for mmWave massive MIMO: traditional and machine learning-based approaches. Sci China Inf Sci 2021; 64(8): 181301.

[14]	Guo J, Wen CK, Jin S, Li GY . Overview of deep learning-based CSI feedback in massive MIMO systems. IEEE Trans Commun 2022; 70(12): 8017-45.

[15]	Xiao Z, Zhang Z, Chen Z, Yang Z, Huang C, Chen X . From data-driven learning to physics-inspired inferring: a novel mobile MIMO channel prediction scheme based on neural ODE. IEEE Trans Wirel Commun 2023; 23(7): 7186-99.

[16]	Yang Y, Zhang Z, Chen Z, Yang Z, Liu L, Huang C, et al. A hybrid inference architecture incorporating neural network with belief propagation for AI receivers. IEEE Trans Wirel Commun 2025; 24(8): 6365-81.

[17]	Wang B, Shuai Z, Huang C, Yang Q, Yang Z, Jin R, et al. Multi-sources fusion learning for multi-points NLOS localization in OFDM system. IEEE J Sel Top Signal Process 2024; 18(7): 1339-50.

[18]	Chen Z. Analogical learning for cross-scenario generalization: framework and application to intelligent localization. 2025. arXiv:2504.08811.

[19]	Xing Z, Zhang Z, Chen Z, Ruan H, Yang Z, Feng Z . Multi-view wireless sensing via conditional generative learning: framework and model design. IEEE Trans Wirel Commun 2026; 25: 12691-708.

[20]	Soltani M, Pourahmadi V, Mirzaei A, Sheikhzadeh H . Deep learning-based channel estimation. IEEE Commun Lett 2019; 23(4): 652-5.

[21]	Li L, Chen H, Chang HH, Liu L . Deep residual learning meets OFDM channel estimation. IEEE Wirel Commun Lett 2020; 9(5): 615-8.

[22]	Liao Y, Hua Y, Cai Y . Deep learning based channel estimation algorithm for fast time-varying MIMO-OFDM systems. IEEE Commun Lett 2020; 24(3): 572-6.

[23]	Alrabeiah M, Alkhateeb A . Deep learning for TDD and FDD massive MIMO: mapping channels in space and frequency. In: Proceedings of the Asilomar Conference on Signals, Systems, and Computers; 2019 Nov 3-6; Pacific Grove, CA, USA. Piscataway: IEEE; 2019. p. 1465-70.

[24]	Chen Z, Zhang Z, Yang Z, Liu L . Channel mapping based on interleaved learning with complex-domain MLP-mixer. IEEE Wirel Commun Lett 2024; 13(5): 1369-73.

[25]	Zhang S, Liu Y, Gao F, Xing C, An J, Dobre OA . Deep learning based channel extrapolation for large-scale antenna systems: opportunities, challenges and solutions. IEEE Wirel Commun 2021; 28(6): 160-7.

[26]	Wen CK, Shih WT, Jin S . Deep learning for massive MIMO CSI feedback. IEEE Wirel Commun Lett 2018; 7(5): 748-51.

[27]	Guo J, Wen CK, Jin S, Li GY . Convolutional neural network-based multiple-rate compressive sensing for massive MIMO CSI feedback: design, simulation, and analysis. IEEE Trans Wirel Commun 2020; 19(4): 2827-40.

[28]	Song X, Wang J, Wang J, Gui G, Ohtsuki T, Gacanin H, et al. SALDR: joint self-attention learning and dense refine for massive MIMO CSI feedback with multiple compression ratio. IEEE Wirel Commun Lett 2021; 10(9): 1899-903.

[29]	Sun Y, Xu W, Liang L, Wang N, Li GY, You X . A lightweight deep network for efficient CSI feedback in massive MIMO systems. IEEE Wirel Commun Lett 2021; 10(8): 1840-4.

[30]	Chen Z, Zhang Z, Xiao Z, Yang Z, Wong KK . Viewing channel as sequence rather than image: a 2-D Seq2Seq approach for efficient MIMO-OFDM CSI feedback. IEEE Trans Wirel Commun 2023; 22(11): 7393-407.

[31]	Jiang W, Schotten HD . Deep learning for fading channel prediction. IEEE Open J Commun Soc 2020; 1: 320-32.

[32]	Luo C, Ji J, Wang Q, Chen X, Li P . Channel state information prediction for 5G wireless communications: a deep learning approach. IEEE Trans Netw Sci Eng 2020; 7(1): 227-36.

[33]	Jiang H, Cui M, Ng DWK, Dai L . Accurate channel prediction based on Transformer: making mobility negligible. IEEE J Sel Areas Commun 2022; 40(9): 2717-32.

[34]

Xiao Z, Zhang Z, Chen Z, Huang C, Zhong C, Debbah M . Mobile MIMO channel prediction with ODE-RNN: a physics-inspired adaptive approach. In: Proceedings of the IEEE Annual International Symposium on Personal, Indoor and Mobile Radio Communications; 2022 Sep 12-15; held virtually. Piscataway: IEEE; 2022. p. 1301-7.

[35]	Xiao Z, Zhang Z, Huang C, Chen X, Zhong C, Debbah M . C-GRBFnet: a physics-inspired generative deep neural network for channel representation and prediction. IEEE J Sel Areas Commun 2022; 40(8): 2282-99.

[36]	Bast SD, Guevara AP, Pollin S . CSI-based positioning in massive MIMO systems using convolutional neural networks. In: Proceedings of the IEEE Vehicular Technology Conference; 2020 May 25-28; Nice, France. Piscataway: IEEE; 2020. p. 1-5.

[37]	Pirzadeh H, Wang C, Papadopoulos H. Machine-learning assisted outdoor localization via sector-based fog massive MIMO. In: Proceedings of the IEEE International Conference on Communications; 2019 May 19-23; Shanghai, China. Piscataway: IEEE; 2019. p. 1-6.

[38]	Chen Z, Zhang Z, Xiao Z, Yang Z, Jin R . Deep learning-based multi-user positioning in wireless FDMA cellular networks. IEEE J Sel Areas Commun 2023; 41(12): 3848-62.

[39]	Tian G, Pjanić D, Cai X, Bernhardsson B, Tufvesson F. Attention-aided outdoor localization in commercial 5G NR systems. IEEE Trans Mach Learn Commun Netw 2024; 2: 1678-92.

[40]	Wang Z, Chen Z, Li R, Zhang Z, Yang Z . Synergizing global pattern learning and time order characterization in mobile channel prediction: an RWKV-based approach. 2026. arXiv:2606.16170.

[41]	Chen Z, Zhang Z, Yang Z, Huang C, Debbah M . Channel deduction: a new learning framework to acquire channel from outdated samples and coarse estimate. IEEE J Sel Areas Commun 2025; 43(3): 944-58.

[42]

Chen Z, Zhang Z, Xing Z, Yang Z . Spatial channel deduction: acquiring channel from approximate position and coarse estimate. In: Proceedings of the IEEE 36th International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC); 2025 Aug 31-Sep 4; Istanbul, Turkiye. Piscataway: IEEE; 2025. p. 1-6.

[43]	Xing Z, Chen Z, Zhang Z, Ruan H, Yang Z . Parameter-efficient channel deduction based on full-domain interleaved learning. In: Proceedings of the IEEE International Symposium on Personal, Indoor and Mobile Radio Communications; 2026 Sep 1-4; Singapore. Piscataway: IEEE; 2026.

[44]	Komninakis C, Fragouli C, Sayed AH, Wesel RD . Multi-input multi-output fading channel tracking and equalization using Kalman estimation. IEEE Trans Signal Process 2002; 50(5): 1065-76.

[45]	Wang T, Wen CK, Jin S, Li GY . Deep learning-based CSI feedback approach for time-varying massive MIMO channels. IEEE Wirel Commun Lett 2019; 8(2): 416-9.

[46]	Ruan H, Zhang Z, Chen Z, Xing Z, Yang Z . Geometry-aided channel deduction: a robust channel acquisition framework utilizing coarse scenario prompt. 2026. arXiv:2605.07425.

[47]	Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention is all you need. In: Advances in Neural Information Processing Systems; 2017 Dec 4-9; Long Beach, CA, USA. Red Hook: Curran Associates; 2017. p. 5998-6008.

[48]	Zhai S, Talbott N, Srivastava S, Guestrin C . An attention free Transformer. 2021. arXiv:2105.14103.

[49]	Alkhateeb A. DeepMIMO: a generic deep learning dataset for millimeter wave and massive MIMO applications. In: Proceedings of the Information Theory and Applications Workshop; 2019 Feb 10-15; San Diego, CA, USA. Piscataway: IEEE; 2019. p. 1-8.

[50]	Shi X, Chen Z, Wang H, Yeung DY, Wong WK, Woo W . Convolutional LSTM network: a machine learning approach for precipitation nowcasting. In: Advances in Neural Information Processing Systems; 2015 Dec 7-12; Montreal, QC, Canada. Red Hook: Curran Associates; 2015. p. 802-10.

[51]	Kingma DP, Ba J . Adam: a method for stochastic optimization. 2014. arXiv:1412.6980.

[52]	DICHASUS massive MIMO CSI dataset collection [Internet]. Stuttgart: University of Stuttgart; [cited 2026 Sep 16]. Available from: https://dichasus.inue.uni-stuttgart.de/.

[53]	Hsieh MH, Wei C . Channel estimation for OFDM systems based on comb-type pilot arrangement in frequency selective fading channels. IEEE Trans Consum Electron 1998; 44(1): 217-25.