Welcome to IEEE Network - The Magazine of Global Internetworking. IEEE Network, published bimonthly, offers readers topics of interest to the networking community. As such, IEEE Network provides a focus for highlighting and discussing major computer communications issues and developments. The articles are intended to be surveys or tutorials, slanted towards the practical, and comprehensible to the nonspecialist, as well as practitioners.

Special Issue on 6G Network AI Architecture for Customized Services and Applications
Submission Date: 2022-08-15

The Fifth Generation (5G) mobile systems have adopted cloud computing and edge computing to support customized services in different application scenarios, such as smart cities, metaverse, interactive Virtual Reality (VR) games, intelligent manufacturing, and autonomous driving. The current 5G network architecture decouples the basic functions of data sensing, communication, and computing at user terminals, mobile networks, and cloud/edge, respectively. Cross-domain resource coordination and service orchestration require in-depth domain knowledge and rich experiences, and hence are very complicated and time-consuming. It is therefore very challenging to effectively coordinate heterogenous resources in distributed facilities for providing agile, stable, and customized services with guaranteed Quality of Experience (QoE) for everyone in dynamic mobile environments.

Thanks to the advancements of different Artificial Intelligence (AI) technologies, such as tiny AI, multi-skilled AI, federated AI, collective AI, collaborative AI, and semantic-oriented AI, the Sixth Generation (6G) mobile systems will develop an intelligent, collaborative, and adaptive network architecture with pervasive AI capabilities existing in 6G systems to address this big challenge. Specifically, a service-oriented approach should be applied to the design and evaluation of such a 6G network AI architecture, which incorporates ubiquitous sensing, storage, communication, computing, control, and AI resources from the cloud to the edge. Cross-domain resources and AI algorithms will be fully shared in the network and effectively orchestrated to customize service provisioning, enhance personalized QoEs, and optimize system performance. With pervasive AI capability, the 6G network AI architecture can adaptively support all kinds of computing-intensive, delay-constrained, security-assured, and privacy-sensitive services and applications for everyone, anywhere and anytime. In summary, the ambitious goal of 6G is to satisfy every user’s individual, integrated, and dynamic service requirements in different application scenarios, network conditions, operation situations, and security environments. This motivates us to conduct active research, developments, and experiments on 6G network AI architecture.

This Special Issue aims at bringing together recent advances on 6G network architecture, pervasive and collaborative AI algorithms, cross-domain service requirements, and security issues. It welcomes original ideas and innovative approaches on network architectures, design methodologies, service schemes, AI algorithms, collaborative protocols, security and privacy, and practical systems from both academia and industry. Potential topics include but are not limited to the following:
- Analytical models of customized service requirements
- Cross-domain performance metrics for QoE evaluation
- Theories for analyzing the service capacity of 6G systems
- Design methodologies of 6G network AI architectures
- 6G network AI architectures with distributed resources and AI capabilities
- Cross-domain resource management and orchestration
- Security and privacy issues of 6G network AI architectures
- 6G network architectures for semantic-oriented communications
- Network economics for collaborative services and applications
- Networking protocols for collaborative services and applications
- Heterogenous network resources coordination and management
- Complexity, robustness, and reliability with heterogenous resources
- Energy efficiency and low carbon emission in customized services
- Al algorithms for customized service provisioning
- AI algorithms for end-to-end QoE guarantees
- Al algorithms for network performance optimization
- Standardization activities on 6G network AI architecture
- 6G wireless testbeds and real-world experiments
- Collaborative distributed learning and federated learning for privacy-preserved 6G customized services