Empowering Wireless Networks With Artificial Intelligence-Generated Graph

In wireless communications, transforming the networks into graphs and processing them using deep learning models is one of the mainstream network optimization approaches. In this direction, the generative AI (GenAI) shows great potential in graph analysis, processing, and generation, offering a broader exploration space for graph-based network optimization. Therefore, this article proposes to use GenAI-based graph generation to support wireless networks. Specifically, we first explore applications of graphs in wireless networks. Then, we introduce and analyse common GenAI models from the perspective of graph generation. On this basis, we propose a framework that incorporates the conditional diffusion model, which can be trained with reward functions and conditions customized by network designers and users. Once trained, it can generate new graphs based on input conditions, thereby addressing various issues in wireless networks. Finally, using the link selection in an integrated sensing and communication (ISAC) as an example, the effectiveness of the proposed framework is validated.