“AI based Machine-to-Machine Communications in 6G” by Sinem Coleri

6G wireless solutions need to meet the anticipated demands of Machine-to-Machine (M2M) communications with the total number of devices expected to be about 26 billion in 2026 according to Ericsson Mobility Report. The wide variety of the requirements of massive M2M scenarios in terms of energy, reliability, delay and throughput, and the inclusion of a new spectrum of control applications, such as extended reality, remote surgery, autonomous vehicle platoons, increase the complexity of M2M system design significantly. This high complexity can only be handled by developing artificial intelligence (AI) based communication techniques, technologies and architectures while addressing the semantics of the control systems. In this talk, I will introduce extreme value theory-based channel modeling and communication techniques detecting rare events in ultra-reliable communication together with federated learning techniques for channel estimation.