The future sixth generation (6G) networks will bring about wider frequency band, faster transmission rate and higher system capacity, for which the challenges of increasingly large and complex and networks, types of terminals and equipment, and complex and diverse business types will need to be addressed. Artificial Intelligence (AI) advanced theories and technologies will need to be utilized to meet the complex requirements of 6G systems. Smart communication uses AI to solve a series of problems in communication systems to realize the intelligence of network elements and network architectures, the intelligence of connected objects, and the intelligence of data processing. The 6G wireless network will encompass all aerial-ground-sea communications and will produce a large quantity of diverse data. Such data will be distributed on different networks, systems, and network elements. However, if trained together, these data will produce high transmission costs and bring about security risks. Therefore, distributed joint learning will be the key technology to realize multi-user smart communication in 6G networks.

The design of distributed learning architecture and optimization of parameter communication mode will become significant factors which effect the output efficiency of AI applications in 6G networks. Additionally, through data-based model structure search, auto machine learning can automatically find the best neural network structure, allowing the network to learn and train rules independently, and to realise self-evolution of the network. Through the interaction process between agent and environment, reinforcement learning can sense spectrum, energy, cache, computing, and other resources, and promote the deeper integration of communication, computing, and storage. The uncertainty of environmental model and the long-term goal when selecting behavior strategies ought to be considered to ensure the adaptability of service, arrangement, and management to improve the autonomy of 6G networks.

This Special Issue aims to attract and encourage submissions of recent intelligent information processing techniques for the multi-source and multi-type data in 6G networks. We particularly welcome novel research focusing on high-performance intelligent computing structures. We also invite submissions addressing the outlook and challenges of intelligence techniques. Original research and review papers are welcome.

Potential topics include but are not limited to the following:

• The design of 6G intelligent network structure
• Distributed and swarm learning architecture in 6G networks
• Federated learning architecture in 6G networks
• AI-based data security and privacy protection in the 6G network
• Automatic machine learning-based network structure evolution
• Reinforcement learning-based computing and storage in 6G networks
• Application of deep learning in network data management
• High performance computing models in smart communication
• AI-based intelligent information processing techniques for multi-source and multi-type data