BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium: Realizing the Potential of NextG Networks: Real-Tim
 e Communications and 5G Digital Twins
DTSTART:20260513T150000
DTEND:20260513T160000
DTSTAMP:20260916T055359Z
UID:b9a78ea4b5d43b206184a7fe0837c56aa52de55258d3012572d36428
CATEGORIES:Conferences - Seminars
DESCRIPTION:By: Kyle Jamieson - Princeton University\n\nAbstract\nWhy hav
 e 5G networks fallen short of their promises circa 2017? Many argue the fo
 cus should shift to making cellular networks more resilient. This talk wil
 l present two examples research at the Princeton Advanced Wireless Systems
  lab to this end. \n\nFirst\, interactive applications over 5G have becom
 e ubiquitous\, yet their Quality of Experience (QoE) often degrades under 
 spectrum scarcity. This stems from a fundamental tension: public 5G networ
 ks must multiplex many users over limited radio resources\, while interact
 ive traffic demands timely and reliable delivery. We observe that within a
 n interactive session\, only a subset of subflows is critically important 
 for QoE. I present our recent work on StreamGuard\, a practical 5G archite
 cture that exploits this insight to enable subflow-level\, QoE-aware prior
 itization.\n\nSecond\, current and future applications demand ultra-low la
 tency and consistent throughput\, yet frequently traverse 5G cellular netw
 orks\, so cope with volatile packet dynamics\, as 5G base station schedule
 rs dynamically react to user workloads and wireless channel conditions. Th
 e task of evaluating network algorithms in these environments is hamstrung
  by current tools: record-and-replay emulators sever the feedback interact
 ion 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 Neur
 alEmu\, a high-fidelity\, machine learning-based emulation framework that 
 learns complex 5G scheduler resource allocation behaviors directly from ex
 tremely high-resolution network telemetry tools.\n\nBio\nKyle Jamieson is 
 Professor of Computer Science at Princeton University\, and Associated Fac
 ulty with the Department of Electrical and Computer Engineering\, the Prin
 ceton NextG Initiative\, and the Princeton Quantum Initiative. He received
  his Bachelor's\, Master's\, and Doctoral degrees from the Massachusetts I
 nstitute of Technology. He is a Distinguished Member of the ACM and a Fell
 ow of the IEEE.\n\nThe Princeton Advanced Wireless Systems Lab (paws.princ
 eton.edu) designs\, builds\, and evaluates wireless systems\, innovating i
 n networking\, sensing\, and computation. This material is based upon work
  supported by the National Science Foundation under grants CNS-2223556 a
 nd OAC-2429485. Any opinions\, findings\, and conclusions or recommendatio
 ns expressed in this material are those of the author(s) and do not necess
 arily reflect the views of the National Science Foundation.\n\nMore inform
 ation\n 
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420 https://epfl.zoom.us/
 j/64039590686
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR