The advent of 5G brings the promise of high bandwidth and low latency, particularly with the adoption of mmWave technology for throughput-sensitive applications. However, the physical limitations of mmWave cells result in frequent handovers in commercial 5G networks. Would these handovers affect TCP performance? In this talk, I will share our recent research addressing this question. Our findings demonstrate that, contrary to expectations, these handovers often cause significant under-utilization of available bandwidth. I will outline our measurement methodology, highlight the improper interactions we uncovered between 5G and TCP, and introduce M2HO, a novel device-centric solution developed to mitigate this issue. The talk will conclude with insights into our ongoing efforts to further enhance the performance of 5G networks.

Cellular technology is constantly evolving. We are in the 5G era with much faster and better radio access than its previous generations (4G/3G). It is mainly attributed to more frequency spectrum resources and advanced radio access technologies. However, not all 5G benefits come without cost and pains. In this talk, I would like to introduce our recent studies to unveil unanticipated problems with 5G radio resource control (RRC). Compared to 4G/3G, 5G scales up radio resource used to serve mobile devices, namely, the number of serving cells increases from 1 to N. However, 5G RRC may not keep pace with the expanding radio resource in all cases, particularly RRC configurations may go wrong or fail to fully utilize radio resources available. As a result, 5G is much slower and worse than it can and should. Finally, I would like to share lessons gained and open questions to scale up RRC and network management in 5G and beyond network.