IC Colloquium: Realizing the Potential of NextG Networks: Real-Time Communications and 5G Digital Twins
By: Kyle Jamieson - Princeton University
Abstract
Why have 5G networks fallen short of their promises circa 2017? Many argue the focus should shift to making cellular networks more resilient. This talk will present two examples research at the Princeton Advanced Wireless Systems lab to this end.
First, interactive applications over 5G have become ubiquitous, yet their Quality of Experience (QoE) often degrades under spectrum scarcity. This stems from a fundamental tension: public 5G networks must multiplex many users over limited radio resources, while interactive traffic demands timely and reliable delivery. We observe that within an interactive session, only a subset of subflows is critically important for QoE. I present our recent work on StreamGuard, a practical 5G architecture that exploits this insight to enable subflow-level, QoE-aware prioritization.
Second, current and future applications demand ultra-low latency and consistent throughput, yet frequently traverse 5G cellular networks, so cope with volatile packet dynamics, as 5G base station schedulers dynamically react to user workloads and wireless channel conditions. The task of evaluating network algorithms in these environments is hamstrung by current tools: record-and-replay emulators sever the feedback interaction that exists between application end points and a commercial operator’s proprietary 5G scheduler, while full-stack simulators rely on overly simplistic scheduling logic. To bridge this reality gap, we present NeuralEmu, a high-fidelity, machine learning-based emulation framework that learns complex 5G scheduler resource allocation behaviors directly from extremely high-resolution network telemetry tools.
Bio
Kyle Jamieson is Professor of Computer Science at Princeton University, and Associated Faculty with the Department of Electrical and Computer Engineering, the Princeton NextG Initiative, and the Princeton Quantum Initiative. He received his Bachelor's, Master's, and Doctoral degrees from the Massachusetts Institute of Technology. He is a Distinguished Member of the ACM and a Fellow of the IEEE.
The Princeton Advanced Wireless Systems Lab (paws.princeton.edu) designs, builds, and evaluates wireless systems, innovating in networking, sensing, and computation. This material is based upon work supported by the National Science Foundation under grants CNS-2223556 and OAC-2429485. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
More information
Abstract
Why have 5G networks fallen short of their promises circa 2017? Many argue the focus should shift to making cellular networks more resilient. This talk will present two examples research at the Princeton Advanced Wireless Systems lab to this end.
First, interactive applications over 5G have become ubiquitous, yet their Quality of Experience (QoE) often degrades under spectrum scarcity. This stems from a fundamental tension: public 5G networks must multiplex many users over limited radio resources, while interactive traffic demands timely and reliable delivery. We observe that within an interactive session, only a subset of subflows is critically important for QoE. I present our recent work on StreamGuard, a practical 5G architecture that exploits this insight to enable subflow-level, QoE-aware prioritization.
Second, current and future applications demand ultra-low latency and consistent throughput, yet frequently traverse 5G cellular networks, so cope with volatile packet dynamics, as 5G base station schedulers dynamically react to user workloads and wireless channel conditions. The task of evaluating network algorithms in these environments is hamstrung by current tools: record-and-replay emulators sever the feedback interaction that exists between application end points and a commercial operator’s proprietary 5G scheduler, while full-stack simulators rely on overly simplistic scheduling logic. To bridge this reality gap, we present NeuralEmu, a high-fidelity, machine learning-based emulation framework that learns complex 5G scheduler resource allocation behaviors directly from extremely high-resolution network telemetry tools.
Bio
Kyle Jamieson is Professor of Computer Science at Princeton University, and Associated Faculty with the Department of Electrical and Computer Engineering, the Princeton NextG Initiative, and the Princeton Quantum Initiative. He received his Bachelor's, Master's, and Doctoral degrees from the Massachusetts Institute of Technology. He is a Distinguished Member of the ACM and a Fellow of the IEEE.
The Princeton Advanced Wireless Systems Lab (paws.princeton.edu) designs, builds, and evaluates wireless systems, innovating in networking, sensing, and computation. This material is based upon work supported by the National Science Foundation under grants CNS-2223556 and OAC-2429485. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
More information
Practical information
- General public
- Free
Contact
- Host: Haitham Al Hassanieh