#ThinkSix - Is it time for wireless communication to get smart(er) with AI/ML? Part 3.

How might an AI based air interface be realized? The first stage is likely to be the front end, using machine learning to model the non-linearities of the transmitter and outperforming existing methods to explore pre-distortion models (for example). Imperfections in current channel models leading to signal-noise degradation and mismatch have already been outperformed by autoencoders producing very different constellation models. This video looks into the future to see whether this work can transform the air interface for a 6G standard.