Abstract:With the rapid development of enabling technologies like VR and AR, we human beings are on the threshold of the ubiquitous human-centric intelligence era. 6G is believed to be an indispensable cornerstone for efficient interaction between humans and computers in this promising vision. 6G is supposed to boost many human-centric applications due to its unprecedented performance improvements compared to 5G and before. However, challenges are still to be addressed, including but not limited to the following six aspects: Terahertz and millimeter-wave communication, low latency and high reliability, energy efficiency, security, efficient edge computing and heterogeneity of services. It is a daunting job to fit traditional analytical methods into these problems due to the complex architecture and highly dynamic features of ubiquitous interactive 6G systems. Fortunately, deep learning can circumvent the interpretability issue and train tremendous neural network parameters, which build mapping relationships from neural network input (status and specific requirements of a 6G application) to neural network output (settings to satisfy the requirements). Deep learning methods can be an efficient alternative to traditional analytical methods or even conquer unresolvable predicaments of analytical methods. We review representative deep learning solutions to the aforementioned six aspects separately and focus on the principles of fitting a deep learning method into specific 6G issues. Based on this review, our main contributions are highlighted as follows. (i) We investigate the representative works in a systematic view and find out some important issues like the vital role of deep reinforcement learning in the 6G context. (ii) We point out solutions to the lack of training data in 6G communication context. (iii) We reveal the relationship between traditional analytical methods and deep learning, in terms of 6G applications. (iv) We identify some frequently used efficient techniques in deep-learning-based 6G solutions. Finally, we point out open problems and future directions.

Augmented Reality Based on Network Physics and 6G for Immersive Experience of Digital Media Art

Cloud-based Fusion Rendering for AR Applications

Toward Immersive Communications in 6G

The Application of 6G and Augmented Reality Technology in Education and Training

Embracing Radiance Field Rendering in 6G: Over-the-Air Training and Inference with 3D Contents

Implementing VR Action and Music Intelligent Interaction Based on 6G Wireless Sensing Technology

Exploring the 6G Potentials: Immersive, Hyper Reliable, and Low-Latency Communication

Haptic Repurposing with GenAI

Spatial Channel State Information Prediction with Generative AI: Towards Holographic Communication and Digital Radio Twin

Realizing XR Applications Using 5G-Based 3D Holographic Communication and Mobile Edge Computing

Entering hyper-reality: “Resonance-In-Sight,” a mixed-reality art installation

Towards Holographic Video Communications: A Promising AI-driven Solution

Digital Twin of a Network and Operating Environment Using Augmented Reality

Deep Learning in the Ubiquitous Human–Computer Interactive 6G Era: Applications, Principles and Prospects

Intelli-AR Preloading: A Learning Approach to Proactive Hologram Transmissions in Mobile AR

The Role of Game Networking in the Fusion of Physical and Digital Worlds through 6G Wireless Networks

Goal-oriented Semantic Communications for Avatar-centric Augmented Reality

ARNature: augmented reality style colorization for enhancing tourism experience

Holistic Network Virtualization and Pervasive Network Intelligence for 6G

Thing2Reality: Transforming 2D Content into Conditioned Multiviews and 3D Gaussian Objects for XR Communication

Full-colour 3D holographic augmented-reality displays with metasurface waveguides