AI Native Air Interface for 6G Sub-THz Radio Access

The sixth generation (6G) of mobile communication systems will be disruptive. 6G means >1Tbit/s wireless data for immersive applications (vis-a-vis, holographic telepresence, metaverse, digital twins, remote surgery, robotics and AI in manufacturing...).

Also, 6G data will be delivered over a network of wireless access networks using multiple wireless pathways (satellite, aerial, terrestrial and subterrestrial). A new wideband air-interface capable of adapting to different operational environments is required.

Advances in artificial intelligence and machine learning (AI/ML) provide a timely opportunity to address this research challenge.The aim of this project is to develop an AI native air interface to provide reliable communication in 6G radio access networks.

The specific objectives are: 1) to develop an AI model based on deep neural networks and large language models to realise, configure and manage an 6G air interface; 2) to train the AI model using combined measured and simulated data sets; and 3) to demonstrate the AI air interface for high data rate beamformed access in 6G bands.