Future smart mobility should require heavier computations and more reliable data transfer, e.g., computational tasks for autonomous driving, LiDAR/radar/camera sensor data transmission via V2X communications. In this regard, the WMNL is investigating the problem of task offloading from multiple coexisting vehicles to road-side 5G mobile edge computing (MEC) servers (see the illustration below), where resource-limited smart vehicles can rely on more powerful computation resources (i.e., MEC servers) at the edge of the 5G network. To do so, vehicles should transmit all the necessary data (e.g., program code to run, internal states) to the MEC servers via 5G communication links. More specifically, we consider two 5G communication interfaces, cellular and mmWave, and try to capture the tradeoff in selectively utilizing them.
We are also investigating how wireless communications should be re-designed for Hyperloop, which is a levitating bullet-train travelling at 1,200km/h inside a near-vacuum metal tube (see the illustration below). The communication environment in the Hyperloop tube is unprecedented, since there exist ultra-fast mobile transceivers inside the waveguide-like structure (i.e., the tube). This makes it a unique communications environment that has never been studied before. We are undergoing an extensive EM analysis to understand and model the intra-tube communication channel. Our lab is also participating in the national consortium for developing the Korean version of Hyperloop (called Hypertube), where the WMNL is closely collaborating with ETRI for the communications part as a leading role at UNIST.
