Aug 2026· Entropy· Vol 28, pp. 884· 0 citations· 29 references
Medicine
TL;DR
IC-GMRO is presented, a conditional information-bottleneck graph-clustering framework for structured representation learning in multi-agent resource optimization and distinguishes representation-level information guarantees from the idealized potential and projected-dual arguments used only to motivate the practical neural updates.
Abstract
Dynamic vehicular integrated sensing and communication (ISAC) requires representations that remain compact, decision-relevant, and structurally stable as mobility rewires interference and sensing relations. This paper presents IC-GMRO, a conditional information-bottleneck graph-clustering framework for structured representation learning in multi-agent resource optimization. At each control epoch, vehicles, roadside units, targets, and typed interactions form a temporal heterogeneous graph. A context-conditioned variational bottleneck suppresses nuisance variation while retaining action-relevant information; balanced soft graph clusters then convert the latent space into reusable coordination codes. Feasibility-masked policies jointly select association, beam, resource block, transmit power, and sensing-time ratio. The analysis distinguishes representation-level information guarantees from the idealized potential and projected-dual arguments used only to motivate the practical neural updates. Controlled simulations and component ablations show improved utility, sensing success, latency robustness, and cross-density robustness relative to greedy, flat, and graph-only baselines.
Vehicular Ad Hoc Networks (VANETs) are a major enabling technology for Intelligent Transportation Systems (ITS), facilitating safety-critical, traffic-efficient vehicle-to-vehicle (V2V) communication. Nevertheless, due to the highly dynamic nature of the vehicular environment, heterogeneous mobility patterns that ensure stable clustering and consistent cluster head (CH) selection pose a challenge. Existing embedding-driven and graph-based vehicle clustering methods rely on localized, one-hop neighbourhood aggregation and assume homogeneous neighbour influence. To overcome these challenges, the paper proposes an attention-guided deep embedding framework that jointly models long-range structural interactions and heterogeneous vehicular influence, rather than restricting learning to immediate neighbours. The proposed algorithm trains latent vehicular representations that understand long-range structural interactions that go beyond immediate one-hop neighbourhoods using Deep Nonnegative Matrix Factorisation (DANMF). To capture the heterogeneous vehicle influence (varying mobility, distance over time of vehicles), the dominant connectivity patterns are highlighted by using an ensemble attention mechanism. This incorporates interaction-factor attention to highlight the high-weighted interaction patterns of latent representation, whereas, through vehicle-level proximity attention, the stable and behaviourally related neighbours are selectively highlighted. Clustering is done according to the learned embeddings, and cluster heads are selected based on the strongest affinity of vehicles to cluster centres, to guarantee that CHs represent varying influence of mobility and structural significance. Extensive simulations reveal that the proposed model outperforms baselines, with a maximum throughput of 7.1 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:Mbps$$\end{document}, CH stability of 90%, and a minimum packet delay of 4.5 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:ms$$\end{document}, which translates to 36% higher throughput, 32% more stable CHs, and 40% reduced packet delay than state-of-the-art techniques. Thus, this way, the proposed model provides a robust solution for stable V2V communication. Not applicable.
Channel knowledge maps (CKMs) or radio maps are a key enabler for AI-native 6G networks, supporting proactive resource allocation, coverage optimization, and environment-aware wireless intelligence. Recent generative approaches have demonstrated improved realism and structural fidelity compared with purely discriminative regression, but often require many iterative sampling steps and incur non-negligible latency in dynamic scenarios. Motivated by the need for fast and high-fidelity CKM construction, we propose a conditional Flow Matching (FM) framework for channel gain map generation from lightweight environment priors. Specifically, our model takes as conditioning inputs (i) a binary/gray-scale building occupancy map and (ii) a transmitter (Tx) location mask, and learns a continuous-time probability flow that transports noise to the target channel gain map. We implement the flow field with a time-conditioned U-Net enhanced by multi-scale context aggregation and attention mechanisms (ASPP + CBAM), improving the model’s capability to capture sharp shadowing boundaries and long-range propagation patterns. At inference, the channel gain map is generated by solving an ODE with a small number of function evaluations, yielding efficient generating while preserving spatial structure. The proposed FM-based CKM constructor provides a promising alternative to diffusion models for workshop settings targeting wireless foundation models and AI-native 6G. The code is avilable at https://github.com/AiBiaoZ/FM-CKM
Experimental results show that, compared with the best-performing baseline in terms of RMSE for each forecasting task, TA-STGAT reduces RMSE by 10.89% and 11.29% in workload-representation prediction and traffic flow prediction, respectively, demonstrating its effectiveness for short-term edge-state forecasting.
A joint model update scheduling and resource allocation framework, aiming to maximize long-term semantic fidelity and resource efficiency of UAV edge intelligence systems, and a Lyapunov-guided discrete reinforcement learning algorithm that performs action space dimensionality reduction and transforms constraints into virtual queue stability problems.
Feng He, Alireza Furutanpey, Paolo Bellavista et al.· 0 citations
Results across heterogeneous urban scenarios show that the proposed framework improves vehicular communication performance in terms of packet delivery reliability and communication delay, while preserving competitive machine learning performance under deployment-oriented constraints.
Juan Pablo Astudillo León, Leticia Lemus Cárdenas, Luis J. de la Cruz Llopis et al.· IEEE Access· 0 citations
Evaluated in the most challenging scenario characterized by maximum vehicle density, dynamic communication delays, packet loss, severe data Non-IIDness, and complex multi-intersection interactions, the proposed method demonstrates superior collaborative performance.
Y. Zhao, Zeng-Wei Guan, W.-Q. Wang et al.· Advanced Electromagnetics· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
