Resource Allocation Driven by Large Models in Future Semantic-Aware Networks
Haijun Zhang, Jiaxin Ni, Zijun Wu, Xiangnan Liu, V. C. M. Leung
TL;DR
The paper tackles resource allocation in semantic-aware networks empowered by large models to reduce data transmission for data-intensive applications. It proposes a large-model-driven architecture that uses RelTR for image-to-text scene graphs and CLIP-based semantic importance to guide transmission, paired with a semantic transmission quality metric defined by $I_s$ and $P_d$ and a diffusion-model-based power allocation to maximize $\sum_{j=1}^N I_j \times (1 - P_{d_j})$. Key contributions include (i) a semantic network architecture that transmits textual semantic triplets rather than raw data, (ii) a formal semantic transmission quality metric combining semantic importance and triplet drop probability, and (iii) a diffusion-based scheme that learns optimal power allocation under dynamic wireless channels, showing improved semantic QoS over baselines in simulations. The work demonstrates a path toward energy- and spectrum-efficient, task-oriented wireless networks, while highlighting open challenges in multimodal data handling, adaptive model tuning, and scalable deployment in resource-constrained settings.
Abstract
Large model has emerged as a key enabler for the popularity of future networked intelligent applications. However, the surge of data traffic brought by intelligent applications puts pressure on the resource utilization and energy consumption of the future networks. With efficient content understanding capabilities, semantic communication holds significant potential for reducing data transmission in intelligent applications. In this article, resource allocation driven by large models in semantic-aware networks is investigated. Specifically, a semantic-aware communication network architecture based on scene graph models and multimodal pre-trained models is designed to achieve efficient data transmission. On the basis of the proposed network architecture, an intelligent resource allocation scheme in semantic-aware network is proposed to further enhance resource utilization efficiency. In the resource allocation scheme, the semantic transmission quality is adopted as an evaluation metric and the impact of wireless channel fading on semantic transmission is analyzed. To maximize the semantic transmission quality for multiple users, a diffusion model-based decision-making scheme is designed to address the power allocation problem in semantic-aware networks. Simulation results demonstrate that the proposed large-model-driven network architecture and resource allocation scheme achieve high-quality semantic transmission.