The School of Computer Science at the University of Windsor is pleased to present…
Building the Foundations for Self-Defending Networks on Programmable Infrastructure
Colloquium Presentation by: Dr. Rhongho Jang
Date: Friday, 9 October 2026
Time: 10:00 – 11:00 am
Location: Erie Hall, room 3123
Abstract:
The Internet's infrastructure has become both the battlefield and the weapon of modern cyberattacks. Botnets now launch floods of up to 31.4 Tbps, while stealthy adversaries hide inside traffic that looks legitimate. As attackers become increasingly AI-empowered, they can adapt, evade, and scale faster than ever. Yet today's defenses still follow an architecture designed for a different era. Traffic is sent to remote scrubbing centers, intelligence is centralized far from the data, and protection goes only to those who can pay for it. This talk presents in-network defense systems built on programmable network infrastructure. The central challenge is that network devices offer line-rate processing but scarce memory and rigid pipelines, while learning-based detection needs rich and complete data. I will discuss how to reconcile these demands through hardware-conscious data structures, collaborative measurement across switches, and generative models that forecast attack behavior from partial observations. Prototypes on Intel Tofino switches and eBPF/XDP servers show that in-network defense can operate at line rate, scale across entire networks, and detect attacks earlier and more reliably than existing approaches.
Keywords: Programmable Data Plane, In-network Computing, Stream Data Summary, ML/AI-assisted Attack Detection.
Biography:
Rhongho Jang is an Assistant Professor in the Department of Computer Science at Wayne State University, where he directs the Networked Intelligence and Distributed Security (NIDS) Lab. He holds a dual Ph.D. in Computer Science and Engineering. His research lies at the intersection of in-network computing, programmable data planes, and generative machine learning for cyber defense, with a focus on bridging the deployment gap between programmable hardware and AI. His research is supported by the National Science Foundation, including a 2026 NSF CAREER Award. He received the Best Paper Award at WISE 2024 and the Best Fast Abstract Award at IEEE/IFIP DSN 2021. He is active in the research community, serving on the Technical Program Committees of USENIX Security 2027 and IEEE INFOCOM 2027 and as Technical Program Chair of IEEE DSC 2025.