Naeem, Faisal, Ali, Mansoor, Kaddoum, Georges, Faheem, Yasir, Zhang, Yan, Debbah, Merouane et Yuen, Chau.
2025.
« A survey on genAI-driven digital twins: Toward intelligent 6G networks and metaverse systems ».
IEEE Open Journal of the Communications Society, vol. 6.
pp. 10365-10402.
Prévisualisation |
PDF
Kaddoum-G-2025-33202.pdf - Version publiée Licence d'utilisation : Creative Commons CC BY. Télécharger (10MB) | Prévisualisation |
Résumé
Sixth-Generation (6G) networks aim to deliver unprecedented network performance by facilitating intelligent, ultra-low-latency, and massively connected applications that seamlessly integrate the physical and digital domains through context-aware operation. These applications work across physical and digital environments. Within this broader shift, digital twins (DTs) have demonstrated notable improvements in overall network performance by creating high-fidelity digital counterparts of physical 6G systems. These DTs give researchers and operators a way to view network behavior as it evolves, to forecast likely performance patterns, and – crucially – to adjust key processes such as beamforming, resource allocation, and interference management. Even so, the value of DT-based optimization is limited by several practical factors. Their effectiveness depends a great deal on access to reliable and sufficiently rich data, and the inherent complexity of 6G environments often makes accurate modeling and efficient resource coordination challenging. This paper examines how a range of generative artificial intelligence (GenAI) models can be used alongside DTs to strengthen resource allocation and improve security in 6G networks. It also sets out a GenAI-enabled DT framework for various 6G-enabling applications, highlighting the potential roles of different GenAI models in supporting semantic communications, the metaverse, integrated sensing and communication (ISAC), AI-generated content (AIGC), and reconfigurable intelligent surfaces (RIS). This paper concludes by drawing attention to emerging conceptual frameworks for DT–GenAI integration. It notes several research challenges that have yet to be resolved, and outlines future directions for deploying GenAI-augmented DTs to achieve intelligent, adaptive, and resilient 6G networks.
| Type de document: | Article publié dans une revue, révisé par les pairs |
|---|---|
| Professeur: | Professeur Kaddoum, Georges |
| Affiliation: | Génie électrique |
| Date de dépôt: | 08 janv. 2026 18:43 |
| Dernière modification: | 10 janv. 2026 19:12 |
| URI: | https://espace2.etsmtl.ca/id/eprint/33202 |
Actions (Authentification requise)
 |
Dernière vérification avant le dépôt |