论文简报
cs.IT 2605.27930v1 值得读

Optimization of CF-mMIMO Systems for the Coexistence between eMBB+ and mMTC+: From Analytical to GNN-Aided Designs

Sergi Liesegang, Lou Salaün, Chung Shue Chen, Stefano Buzzi

发布日期：2026-05-27 04:01 相关性：1.0000 价值：0.7600 分类：cs.IT eess.SP
arXiv 摘要 PDF 原文 网页正文

摘要

This paper investigates uplink multiple access for the coexistence of enhanced mobile broadband+ (eMBB+) and massive machine-type communications+ (mMTC+) in terminal-centric cell-free massive MIMO (CF-mMIMO) systems. We propose a non-orthogonal scheme in which low-rate mMTC+ transmissions are spread across the time-frequency grid shared with eMBB+ users, enabling efficient resource reuse. In the presence of imperfect channel state information, we derive closed-form expressions for the achievable rates of both services based solely on statistical channel knowledge. For mMTC+ devices, the analysis also incorporates finite blocklength (FBL) modeling to capture short-packet transmissions. To support heterogeneous service requirements, we formulate a power-control problem that maximizes the minimum energy efficiency of mMTC+ devices subject to quality-of-service constraints on eMBB+ users. The resulting nonconvex problem is solved via sequential fractional programming, accounting for both the Shannon and FBL regimes. To enable real-time operation, we further propose a graph neural network (GNN) with multi-head attention to approximate the model-based solution. Constraint satisfaction during training is enforced via an augmented Lagrangian loss. Numerical results demonstrate effective multiplexing of the two data services and show that the proposed GNN algorithm achieves near-optimal performance with a significantly lower computational complexity.

相关性判断

high
相关方向
wireless_communications cell_free_mimo finite_blocklength multiple_access power_control
判断依据

Directly in cs.IT and eess.SP; studies uplink multiple access, achievable rates, finite blocklength analysis, and power control for CF-mMIMO coexistence, which is squarely within information theory and communications.

价值判断

High relevance to cs.IT wireless communications with a complete analytical-to-optimization pipeline for CF-mMIMO coexistence. Technical contribution spans closed-form achievable rates under imperfect CSI, finite blocklength mMTC+ modeling, max-min energy-efficiency power control, and GNN approximation. Worth deeper reading for researchers tracking CF-mMIMO, heterogeneous service multiplexing, or learning-aided wireless optimization, though not clearly field-defining from the provided evidence alone.

核心问题与主要方法

核心问题

Uplink coexistence, rate analysis, and power control for eMBB+ and mMTC+ in terminal-centric CF-mMIMO under imperfect CSI

场景：Scalable uplink CF-mMIMO with distributed APs, subset AP-user association, spread-spectrum mMTC+, and shared time-frequency resources with eMBB+

主要方法

mMTC+ transmissions are spread across all available PRBs using PN sequences, enabling shared-resource coexistence with eMBB+ while trading spreading gain against a 1/N rate pre-log penalty. Achievable rates are derived under imperfect CSI using use-and-then-forget lower bounds; mMTC+ analysis requires handling high-order moments and adds an FBL penalty term for short packets. The power-control objective is a max-min fractional program over eMBB+ and mMTC+ powers, targeting mMTC+ energy efficiency while enforcing eMBB+ QoS and mMTC+ reliability constraints. Sequential FP constructs convex/concave surrogate bounds and applies Dinkelbach-style fractional programming loops to obtain stationary solutions for Shannon and FBL regimes. The learning surrogate maps LSF coefficients on a heterogeneous line graph to power coefficients using multi-head attention, with augmented Lagrangian terms penalizing QoS constraint violations.

关键贡献与后续阅读

关键贡献

Formulates a non-orthogonal uplink coexistence scheme for eMBB+ and mMTC+ in terminal-centric CF-mMIMO where mMTC+ spread-spectrum signals reuse eMBB+ time-frequency resources. Derives imperfect-CSI achievable-rate lower bounds for both services using statistical channel knowledge, including finite-blocklength rate modeling for mMTC+ short-packet operation. Defines a fairness-oriented power-control problem that maximizes the minimum mMTC+ energy efficiency while satisfying eMBB+ QoS, mMTC+ QoS, transmit-power, and reliability constraints. Develops sequential fractional-programming solutions for both infinite-blocklength and finite-blocklength regimes, using surrogate bounds and Dinkelbach-type fractional optimization. Proposes a heterogeneous multi-head-attention GNN over AP-terminal link nodes to approximate the analytical power-control solution with lower inference complexity and constraint-aware augmented Lagrangian training.

研究启发

How sensitive are the GNN feasibility and EE results to deployment distributions outside the five simulated training scenarios? Do the closed-form mMTC+ moment expressions remain usable beyond the uncorrelated-fading specialization mentioned in the excerpt? How much offline cost is required to generate supervised labels with sequential FP, and does that cost dominate practical deployment updates? Are the reported near-optimal GNN results robust under different AP association policies, traffic activity models, or pilot contamination regimes?

限制与不确定性

Novelty may be incremental because the components are familiar: CF-mMIMO, FBL, fractional programming, and GNN-based optimizer approximation. Evidence is based on abstract-level and structure analysis, not independent validation of derivations or numerical results. GNN usefulness depends on training data generation and generalization, which may be limited by offline analytical supervision.

参考文献

48 条
  1. [1] (2022-Huawei, White Paper, Jan.) 6G: the next horizon . External Links: Link Cited by: §I , §VII .
  2. [2] M. Avriel (2003) Nonlinear programming: Analysis and methods . Courier Corporation . Cited by: § V-A .
  3. [3] A. K. Bairagi et al. (2021) Coexistence mechanism between eMBB and uRLLC in 5G wireless networks . IEEE Trans. Commun. 69 ( 3 ), pp. 1736–1749 . External Links: Document Cited by: § I-A .
  4. [4] E. Bjornson and P. Giselsson (2020) Two applications of deep learning in the physical layer of communication systems [lecture notes] . IEEE Signal Process. Mag. 37 ( 5 ), pp. 134–140 . External Links: Document Cited by: § I-A .
  5. [5] E. Björnson and L. Sanguinetti (2020) Scalable cell-free massive MIMO systems . IEEE Trans. Commun. 68 ( 7 ), pp. 4247–4261 . External Links: Document Cited by: § II-C 1 , § III-A , footnote 4 , footnote 6 .
  6. [6] S. Boyd and L. Vandenberghe (2004) Convex optimization . Cambridge Univ. Press . External Links: ISBN 9780521833783 , LCCN 03063284 Cited by: §V , § VI-C , § VII-B 2 , § VII-C .
  7. [7] S. Boyd et al. (2011) Distributed optimization and statistical learning via the alternating direction method of multipliers . Foundations and Trends® in Machine learning 3 ( 1 ), pp. 1–122 . Cited by: §I , § VI-C .
  8. [8] S. Buzzi et al. (2016) A survey of energy-efficient techniques for 5G networks and challenges ahead . IEEE J. Sel. Areas Commun. 34 ( 4 ), pp. 697–709 . External Links: Document Cited by: §I .
  9. [9] S. Buzzi et al. (2026-early access,) Why user-centric cell-free distributed MIMO systems will be the disruptive 6G technology . IEEE Commun. Mag. ( ), pp. 1–7 . External Links: Document Cited by: §I .
  10. [10] U. Challita et al. (2020) When machine learning meets wireless cellular networks: Deployment, challenges, and applications . IEEE Commun. Mag. 58 ( 6 ), pp. 12–18 . External Links: Document Cited by: § VI-B .
  11. [11] J. Crouzeix and J. A. Ferland (1991) Algorithms for generalized fractional programming . Mathematical Programming 52 ( 1 ), pp. 191–207 . Cited by: §V , §V .
  12. [12] Ö.T. Demir, E. Björnson, and L. Sanguinetti (2021) Foundations of user-centric cell-free massive mimo . Now Publishers . External Links: ISBN 9781680837902 Cited by: §I , §I , § III-A , §VII , §VII , footnote 6 .
  13. [13] G. Di Gennaro et al. (2026) A general framework for scalable UE-AP association in user-centric cell-free massive MIMO based on recurrent neural networks . IEEE Trans. Commun. 74 ( ), pp. 3103–3119 . External Links: Document Cited by: §VII .
  14. [14] W. Dinkelbach (1967) On nonlinear fractional programming . Management science 13 ( 7 ), pp. 492–498 . Cited by: §V , Algorithm 1 .
  15. [15] M. Eisen et al. (2019) Learning optimal resource allocations in wireless systems . IEEE Trans. Signal Process. 67 ( 10 ), pp. 2775–2790 . External Links: Document Cited by: §I .
  16. [16] Y. C. Eldar, A. Goldsmith, D. Gündüz, and H. V. Poor (2022) Machine learning and wireless communications . Cambridge Univ. Press . Cited by: § VI-C .
  17. [17] M. Elwekeil et al. (2023) Power control in cell-free massive MIMO networks for UAVs URLLC under the finite blocklength regime . IEEE Trans. Commun. 71 ( 2 ), pp. 1126–1140 . External Links: Document Cited by: §II , §VII , §VII .
  18. [18] (2023-Ericsson, Document EAB-23:009890, Nov.) Ericsson mobility report . External Links: Link Cited by: §I .
  19. [19] G. Femenias and F. Riera-Palou (2026-early access,) Scalable eMBB/URLLC-enabled cell-free massive-MIMO with hardware impairments . IEEE Trans. Veh. Technol. ( ), pp. 1–16 . External Links: Document Cited by: § I-A .
  20. [20] (2017-3GPP, Technical Report 36.814 v9.2.0) Further advancements for e-utra physical layer aspects . Cited by: §II , §VII .
  21. [21] A. Ghosh and R. Ratasuk (2011) Essentials of lte and lte-a . Cambridge Univ. Press . Cited by: §VII .
  22. [22] M. Grant and S. Boyd (2024) CVX: MATLAB software for disciplined convex programming, version 2.2.2 . Note: http://cvxr.com/cvx Cited by: § V-A , § VII-C .
  23. [23] J. Guo and C. Yang (2022) Learning power allocation for multi-cell-multi-user systems with heterogeneous graph neural networks . IEEE Trans. Wireless Commun. 21 ( 2 ), pp. 884–897 . External Links: Document Cited by: § I-A , §I .
  24. [24] J. Hoydis et al. (2021) Toward a 6G AI-native air interface . IEEE Commun. Mag. 59 ( 5 ), pp. 76–81 . External Links: Document Cited by: §I .
  25. [25] G. Interdonato et al. (2023) On the coexistence of eMBB and URLLC in multi-cell massive MIMO . IEEE O. J. Commun. Soc. 4 ( ), pp. 1040–1059 . External Links: Document Cited by: § I-A , § II-A , § II-B , § III-A .
  26. [26] N. Jindal et al. (2008) Bandwidth partitioning in decentralized wireless networks . IEEE Trans. Wireless Commun. 7 ( 12 ), pp. 5408–5419 . External Links: Document Cited by: §IV .
  27. [27] E. A. Jorswieck (2024) Next-generation multiple access: From basic principles to modern architectures . Proc. IEEE 112 ( 9 ), pp. 1149–1178 . External Links: Document Cited by: §I .
  28. [28] S. Liesegang and S. Buzzi (2025) Coexistence of eMBB+ and mMTC+ in uplink cell-free massive MIMO networks . In 2025 IEEE Wireless Communications and Networking Conference (WCNC) , Vol. , pp. 1–6 . External Links: Document Cited by: Optimization of CF-mMIMO Systems for the Coexistence between eMBB+ and mMTC+: From Analytical to GNN-Aided Designs , § I-B , §I , § II-B , §II , § III-B , §V .
  29. [29] S. Liesegang et al. (2025) Design of RIS-aided mMTC+ networks for rate maximization under the finite blocklength regime with imperfect channel knowledge . IEEE Commun. Lett. 29 ( 11 ), pp. 2511–2515 . External Links: Document Cited by: § V-A .
  30. [30] S. Liesegang et al. (2026) Scalable integrated sensing and communications for multi-target detection and tracking in cell-free massive MIMO:A unified framework . IEEE Trans. Commun. 74 ( ), pp. 2777–2793 . External Links: Document Cited by: §VII .

底部评论

0 条根评论，可继续回复叠楼

0/2000