Master Seminar - Hot Topics in Edge Computing and QoS (2026)
Course Description
Emerging data-intensive applications such as Autonomous Vehicles (AVs), the Extended Reality (AR/VR/MR), and mission-critical IoT demand more than just raw processing power: they require a seamless integration of computation and communication.
This Master’s level seminar explores the symbiotic relationship between Edge Computing and Next-Generation Network Quality of Service (QoS). While Edge Computing addresses latency by moving cloud utilities and intelligence to the network edge (e.g., base stations, resource-constraind devices), proximity alone is insufficient. To fully realize the potential of the Edge, the underlying wireless infrastructure must actively support these transactions.
This seminar will explore recent advances from a combined perspective and investigate the most challenging problems in this research field. The topics may include but not limited to:
- Real-time and data-intensive applications for Edge Computing (e.g. Video Analytics, AVs, AR/VR/MR)
- Edge AI or Edge Intelligence (Their motivation, advantages, and challenges)
- Distributed/collaborative/collective computing in the continuum of device, edge, and cloud (This falls into the wide range of distributed systems)
- Hardware-accelerated Edge Computing (e.g. FPGAs, Accelerators, SoCs, GPUs)
- Edge-aware network architectures for QoS provisioning (networks can be 5G/6G, Wi-Fi, Non-terrestrial networks, satellite, and heterogeneous networks between them)
- System-level orchestration of computing and network resources (e.g., MEC placement and selection, NFV deployment and management with edge infrastructure)
- AI-integrated network optimization and autonomous radio and computing resource management (e.g., task offloading, joint optimization across radio and computing domains)
Pre-course meeting
There will be no pre-course meeting. The structure and procedure of the seminar are outlined in these Seminar Course Informational Slides
Previous Knowledge Expected:
- Interested in distributed systems, networking, or cloud computing.
- Have technical background in computer science, computer engineering, software engineering, or a related field.
- Have basic knowledge of computer networks, operating systems, programming languages.
- Ability to read and present in English. Good team work and communication skills.
Moodle page
[TBD] Seminar Hot Topics in Edge Computing (IN2107, IN4417)
Time and location
The seminar sessions will be hold on-site in Room: MI 01.07.023
Only reasonable excuses can lead to an online participation.
All sessions are happening between 3PM-5:30PM on Thursday except Session#2 due to holiday reasons.
April 16: Introduction&QA Session - Provide students with basic informations and FAQs
April 17: Paper Assignment - Instructors assign papers to participants according to preferences. (No session)
April 30: Seminar Session #1
May 13 4-6pm Wed: Seminar Session #2
May 28: Seminar Session #3
You can also subscribe to this Google Calendar - Edge Comp. Seminar
Objective
Upon completion of the seminar, the students will
- have broadened their knowledge on the current state of Edge Computing and QoS Provisioning in Wireless Networks research and their gaps
- gain critical thinking and analytical skills to review and evaluate research papers
- become familiar with reading, reviewing, and presenting academic papers in a setting similar to a scientific peer review process
Languages of Instruction
English
Teaching and Learning Method
> (Transfer of Skills) Workload for Students
Workload:
Given a list of papers, each participant will be required to:
- Write 3 reviews (2 for pure review, 1 for review and presenting) (about 6-12 hours workload, 40% of final grade)
- Present 1 paper during the semester (about 4-10 hours workload, 40% of final grade)
- Attend and participate sessions (about 2-3 hours/session x 3 session, 20% of final grade)
Workflow:
1. Each participant could submit their preference (2-3 preference to topics)
2. The lecturer will assign papers to them for review (2 papers) and present (1 paper)
3. The participants finish and submit reviews before the DDLs(sessions). There will be a paper presenting schedule, please find the corresponding sessions for your papers.
4. The presenter present the paper (15 ± 5 mins) and the corresponding reviewers participate in the discussion (15 ± 5 mins)
(The exact numbers may slightly vary due to the registered students.)
Further Reading
- S. Keshav. "How to read a paper"
- William G. Griswold, "How to Read an Engineering Research Paper"
- Graham Cormode. 2009. "How NOT to review a paper: the tools and techniques of the adversarial reviewer."
- Timothy Roscoe, ETH Zürich, 2007. Writing reviews for systems conferences
- J Smith. "The Task of the Referee"
- EuroSys Shadow TPC guidelines (Credits and Copyrights to Jing & Xingda)
Contact
Please do subject your email with “[Seminar Edge Computing] - your query” to both of us, thanks.
- Wei Geng <wei.geng@tum.de> Office Hours: 13-13:30 Wed (Email preferred)
- Hyerin Kim <hyerin.kim@tum.de>