A Variegated Look at 5G in the Wild: Performance, Power, and QoE Implications

Abstract

Motivated by rapidly expanding deployment of 5G services, we carry out an in-depth study to understand how 5G impacts application performance and energy efficiency. We examine the effects of 5G deployment strategies such as Non-Standalone (NSA) 5G and Standalone (SA), radio bands and protocol specific properties (e.g., RRC state transitions and power profiles) on performance and power usage experienced by user equipment (smart phones) and applications running on them. Unlike previous studies, we focus on both 5G mmWave – which provides significantly higher throughput than 4G LTE – and low-band radio (with more comparable throughput with 4G LTE). Our work answers the key questions of how to best utilize 5G for common applications such as video streaming and web browsing, and aid intelligent throughput-power trade-offs.

Publication
Proceedings of the 2021 ACM SIGCOMM 2021 Conference
Shuowei Jin
Shuowei Jin
PhD Candidate at University of Michigan CSE

My research interests include efficient LLM inference algorithms/systems, machine learning systems, and mobile systems.