The traffic in today’s networks, 4G, 5G, mobile or otherwise, seems to be following nicely the exponential expectations projected each year. On the one hand, this is driven by and drives further CMOS scaling for the digital processing of information; on the other hand, this has pushed communication channels to use ever wider bandwidths. Unfortunately, not only the individual endpoint throughputs are increasing, but the amount of endpoints and their capabilities is skyrocketing as well.
Moreover, capacity as a KPI is being complemented by reliability and latency as use-cases branch out beyond the traditional human-centric communications and entertainment into areas such as industrial automation, AR/VR and autonomous vehicles. Especially the future communications between autonomous systems (whether this is AI or ML) are expected to create the next wave of traffic.
The wireless networks of tomorrow (6G) will be even more complex than today’s (5G) networks, featuring advanced techniques such as cell-free MIMO, advanced/distributed beamforming, new waveforms and frequencies above 100GHz. But we also see that these shifts can and will have an impact on the edge network. The purpose of this PhD research is to explore the network architecture trade-offs in this new environment, and model (in terms of throughput, energy efficiency, cost, latency) the use of high-frequency, large bandwidth, bursty meshed short-range communication links that will support these new users, new applications, using these new capabilities. In short, to design the next generation network for mobile and interacting AI's, which are supported by virtualized edge clouds.
The successful PhD candidate will be part of a large IMEC team working on the research, implementation and prototyping of future communications systems: experts in digital, analog and mm-wave ASIC design, wireless communications, PHY processing, MAC and higher layers, machine learning.
To get some sense of the domain: