Generative AI for the Optimization of Next-Generation Wireless Networks: Basics, State-of-the-Art, and Open Challenges

TL;DR

This paper surveys the use of Generative AI (GAI) for optimizing next-generation wireless networks (xG), highlighting how models like GANs, GFlowNets, and Generative Diffusion Models (GDMs) can learn from real network data and generate diverse scenarios for offline exploration and robust optimization. It presents a structured review of GAI foundations, network paradigms (AIGC, ISAC, SemCom, and security), and a set of use cases across mobile, ISAC-enabled, and SemCom-enabled networks, culminating in a diffusion-based case study for load balancing, carrier aggregation, and backhauling in non-terrestrial networks. The paper discusses networking requirements, challenges, and future directions, including distributed learning, edge computing, memory management, and scalable hardware. The case study demonstrates how diffusion-based GAI integrated with reinforcement learning can address high-dimensional, dynamic NTN optimization problems, illustrating practical pathways toward 6G architectures.

Abstract

Next-generation (xG) wireless networks, with their complex and dynamic nature, present significant challenges to using traditional optimization techniques. Generative AI (GAI) emerges as a powerful tool due to its unique strengths. Unlike traditional optimization techniques and other machine learning methods, GAI excels at learning from real-world network data, capturing its intricacies. This enables safe, offline exploration of various configurations and generation of diverse, unseen scenarios, empowering proactive, data-driven exploration and optimization for xG networks. Additionally, GAI's scalability makes it ideal for large-scale xG networks. This paper surveys how GAI-based models unlock optimization opportunities in xG wireless networks. We begin by providing a review of GAI models and some of the major communication paradigms of xG (e.g., 6G) wireless networks. We then delve into exploring how GAI can be used to improve resource allocation and enhance overall network performance. Additionally, we briefly review the networking requirements for supporting GAI applications in xG wireless networks. The paper further discusses the key challenges and future research directions in leveraging GAI for network optimization. Finally, a case study demonstrates the application of a diffusion-based GAI model for load balancing, carrier aggregation, and backhauling optimization in non-terrestrial networks, a core technology of xG networks. This case study serves as a practical example of how the combination of reinforcement learning and GAI can be implemented to address real-world network optimization problems.

Figures (8)

• Figure 1: AI, machine learning, deep learning, and GAI: A hierarchical overview.
• Figure 2: Flow network ($F$) in a GFlowNet architecture: the root node ($S0$) starts with an inflow $Z$, and transitions between states ($S1, S2, S3$) occur via actions ($a1, a2, a3, a4, a5, a7$). Each state follows the flow balance equation ensuring the sum of incoming flows equals the sum of outgoing flows. The terminal state ($S4$) has an outflow denoted as $R(S4)$.
• Figure 3: Architecture of a GAN: the generator network takes training data and produces fake data, while the discriminator network distinguishes between real and fake data, outputting $0$ or $1$ depending on whether the input data is real or generated.
• Figure 4: Deriving the action probability given in \ref{['eq_softmax']} through the backward process of diffusion-based GADM algorithm.
• Figure 5: System model of a LEOS-based non-terrestrial network with carrier aggregation technology.