← Back to all news
arXiv Machine Learning By Junyu Liao, Lili Chen, Tian Han, Jamie Evans

Graph Neural Networks for Resource Allocation in Integrated Sensing and Communication

Read the original on arXiv Machine Learning →

The paper introduces GNN-ISAC, a graph neural network–based power allocation algorithm for vehicular integrated sensing and communication networks. It models two edge types—inter‑provider interference and intra‑provider power‑budget sharing—to maximize communication sum‑rate while meeting sensing SINR constraints. Trained without labels via a sensing‑penalized sum‑rate loss, GNN‑ISAC achieves near‑optimal performance compared to WMMSE, with lower sensing‑link violations, reduced computation, and better generalization to larger networks.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv AI
Oct 15

GREAT: A Graph Edge Attention Network for the Electric Vehicle Routing Problem with Asymmetric Costs

The paper introduces GREAT, a Graph Edge Attention Network that operates directly on edge space to handle asymmetric energy costs in the Electric Vehicle Routing Problem (EVRP). Integrated into a reinforcement learning framework, GREAT jointly optimizes routing and charging decisions, outperforming a strong HGS-based heuristic by up to 8.78% in energy savings and reducing runtime by a factor of 60. It also shows competitive performance against a custom adaptive large neighborhood search and achieves state‑of‑the‑art results on other asymmetric routing benchmarks such as TSP, CVRP, and Orienteering.

By Attila Lischka, Filip Rydin, Jiaming Wu, Morteza Haghir Chehreghani, Bal\'azs Kulcs\'ar
llmsreinforcement-learningbenchmarks
More like this →
arXiv AI
Sep 18

AI-Driven Real-Time Relay Optimisation in Smart Urban NR-V2X Networks via Learning-to-Optimise Graph Neural Networks

The paper introduces an AI-driven Learning-to-Optimise framework that uses Graph Neural Networks to select real‑time multi‑hop relays in NR‑V2X networks for smart cities. By modelling the vehicular network as a graph and training a GINE model with MILP‑derived optimal decisions, the method achieves near‑optimal connectivity, improving connectivity by up to 11.3% and speeding up computation by up to 100×. This enables scalable, real‑time network control suitable for Industry 4.0 and smart‑city deployments.

By Giambattista Amati, Federica Mangiatordi, Emiliano Pallotti, Simone Angelini
More like this →