Deutsch Intern
  • [Translate to Englisch:]
Chair of Computer Science III


Telefon (0931) 31-88475
Telefax (0931) 31-86632
E-Mail michael.seufert@uni-wuerzburg.de
Homepage www.mseufert.de
Raum A206
Anschrift Lehrstuhl für Informatik III
Am Hubland
D-97074 Würzburg

Current Projects

  • UserNet (ML-based Monitoring and Management of QoE for User-centric Communication Networks, Emmy Noether Junior Research Group funded by DFG)

In order to allow QoE monitoring for arbitrary Internet applications, the interplays between QoE and user interactions is investigated and modelled based on measurements and subjective studies. In addition, ML methods are adapted to the domain in order to apply them to encrypted network traffic. This allows to quantify the QoE by monitoring interactions and the resulting changes in the encrypted application traffic. Based on this, a data-driven improvement of QoE and QoE fairness is enabled by using reinforcement learning to find optimal network configurations by interacting with the dynamic network environment. By means of powerful, software-defined networking (SDN) technologies like P4, together with available computing resources in the network, such fine-granular models can now be implemented in the network for the first time, such that network management becomes more dynamic. Thus, the implementation of the required ML-based algorithms and components and their integration into network operation is researched.

  • Group-based Communications (Analysis and Modelling of Group-based Communication in Internet Applications for Improved Edge Caching, funded by DFG)

The goal of this research project is analyzing the novel communication paradigm of group-based communication. The findings are used to design edge caching mechanism, which can reduce the network load imposed by user-generated contents, and increase the Quality of Experience of end users.

  • Wintermute  (AI-based network assessment, policy definition, and enforcement of security in complex networks, funded by BMBF)

In this project, we aim to ensure safe communication with the help of artificial intelligence (AI) and machine learning (ML) under two main aspects: privacy and usability. While the preservation of secure communication is important, a violation of privacy has to be prevented. Additionally, the goal is to provide a user-friendly solution to intrusion detection, which supports specialists in their decision making by providing an assessment of the current network situation.

  • Trendscout Survey: Decentralized Data Network Architectures (DeDaNe, funded by an industry partner)

The goal of the project is to give an overview on decentralized data network architectures and to survey relevant focus areas in order to identify potential topics for future research.


Previous Projects

Projects at University of Würzburg (2019-)

  • WebQKAI (AI-based approaches to identify Web QoE key performance indicators in the network, funded by AIT Austrian Institute of Technology)

The objective of the WebQKAI project is to infer web QoE key performance indicators (KPIs) from data collected by network devices, which provide insights for operators with respect to network operations and maintenance. Therefore, relevant and important network scenarios have to be investigated in this project. The goal of the project is to estimate the web QoE in specific scenarios (including web accesses in web browsers or mobile apps) with high accuracy.

  • KIKDIN  (Usage of AI methods for the parametrization of convergent, deterministic, industrial networks, funded by Siemens Predevelopment Nürnberg)

Within the KIKDIN project, the chair of communication networks of the University of Würzburg and SIEMENS investigate the usage of AI methods for the parametrization of convergent, deterministic, industrial networks. The long-term goal is to assign and parametrize network configurations for different application types and different network infrastructures in an automated and dynamic way. The assignment shall be based on statistical methods, which especially includes the usage of machine learning as an AI method.

  • QoE-Care (Quality of Experience of business applications, funded by an industry partner)

This project focusses on the relationship between the perceived quality of the performance of business applications by the employees and the technical performance data of such applications.

  • EduQoE (Quality of Experience Monitoring and Analysis in Education-purpose Networks, collaboration with Plan Ceibal)

Digitized and online education services need reliable and high quality access networks to avoid performance degradations, which impact the quality of the education. In this project, we teamed with Plan Ceibal to monitor and analyze the QoE in their networks, with the eventual goal of improving their services.



Projects at AIT Austrian Institute of Technology (2018-2019)

I worked as a scientist and postdoc at AIT Austrian Institute of Technology in the Digital Insight Lab at the Center for Digital Safety & Security. I was involved in the following projects:

  • Big-DAMA (Big Data Analytics for Network Traffic Monitoring and Analytics, funded by Vienna Science and Technology Fund (WWTF))

The Big-DAMA project is conceiving novel scalable techniques and big-data frameworks capable to analyze both online network traffic data streams and offline massive traffic datasets. The goal is exploring scalable online and offline data mining and machine learning-based techniques to monitor and characterize extremely large network traffic datasets.

  • ViCrypt (Video Quality Metrics Analysis for (non)Encrypted Video Traffic, funded by an industry partner)

In this project, both encrypted and non-encrypted video traffic was monitored and analyzed. Relevant features were extracted from the network traffic as input to machine learning models. The goal of the project was to predict video QoE metrics and user actions with high accuracy.

  • EduQoE (Quality of Experience Monitoring and Analysis in Education-purpose Networks, collaboration with Plan Ceibal)

Digitized and online education services need reliable and high quality access networks to avoid performance degradations, which impact the quality of the education. In this project, we teamed with Plan Ceibal to monitor and analyze the QoE in their networks, with the eventual goal of improving their services.


Projects at University of Würzburg (2013-2017)

During my PhD, I worked as a researcher at University of Würzburg, Chair of Communication Networks, and was involved in teaching, as well as the following research projects:

  • QoE³ (Quality of Experience Estimation at the Edge, funded by an industry partner)

The goal of the project is to investigate and assess real-time monitoring solutions or estimators for Quality of Experience (QoE) for HTTP adaptive video streaming (HAS) at the network edge.

  • Mobi-QoE (Monitoring and Analysis of Quality of Experience in Mobile Broadband Networks, funded by European Commission H2020 project MONROE (Measuring Mobile Broadband Networks in Europe))

The objective of Mobi-QoE is to extend MONROE’s testbed to the QoE domain by integrating novel software-based QoE-capable measurement tools and QoE models for popular end-user services (e.g., YouTube, Facebook, Spotify). In crowdsourced field trials, the extensions will be evaluated and the QoE models will be refined to enable experiments for QoE-based performance analysis of mobile broadband networks with the MONROE testbed.

  • What's Up (WhatsApp against depression of children and adolescents, funded by Startnext crowdfunding campaign)

What's Up is an interdisciplinary collaboration with psychologists at University of Tübingen to analyze the communication of depressive children and adolescents in WhatsApp with the help of WhatsAnalyzer. The goal is to establish an early-warning system for depressive phases, which can effectively be used in the treatment of depression. In this early phase of the project, a pilot study is planned to assess the research prospects and to prepare future proposals for funding.

  • INPUT (In-Network Programmability for Next-Generation Personal Cloud Service Support, funded by European Commission H2020)

The goal of INPUT is to design a novel edge cloud supported infrastructure for personal cloud services. Within this project, our focus is mainly on monitoring, orchestration, and consolidation algorithms for optimized configuration of network/cloud infrastructure and allocation of virtualized cloud services with respect to energy consumption and Quality of Experience.

  • SmartenIT (Socially-Aware Management of New Overlay Application Traffic combined with Energy Efficiency in the Internet, funded by European Commission FP7)

SmartenIT was dedicated to incentive-compatible cross-layer optimization of network operation by exploiting social awareness, QoE awareness, and energy efficiency. Within the project, we acted as work package leader for "Traffic Measurements and Scenario Definitions" and were involved in the design and evaluation of traffic mechanisms for the end-user focused scenario. We developed HORST (home router sharing based on trust), a socially-aware traffic management mechanism, which enables WiFi offloading, content prefetching/caching, and content delivery on user-owned home routers. The implemented prototype of RB-HORST (HORST combined with RB-Tracker for content delivery) won the Best Demo Award at LCN 2015.

  • SmartQoE (Measurement Concept and Trendscouting for QoE in Mobile Context, funded by FTW Telecommunication Research Center Vienna)

SmartQoE was an industry collaboration with FTW Telecommunication Research Center Vienna. The goal of the project was to develop a measurement concept for monitoring QoE-relevant application data for video streaming on a smartphone. Therefore, we developed and implemented the Android-based YoMoApp (YouTube Monitoring App), which was used by FTW in lab and field studies, and evaluated the collected data. Moreover, a trendscouting was part of the project to estimate the potential of new research topics on QoE.

Projects at FTW Forschungszentrum Telekommunikation Wien (Telecommunication Research Center Vienna, 2012-2013)

Before I joined the Chair of Communication Networks of the University of Würzburg, I worked at Forschungszentrum Telekommunikation Wien (FTW, Telecommunication Research Center Vienna) in the area of user-centered interactions and communication economics (2012-2013). During that time, I was involved in the following projects:

  • ACE 2.0, ACE 3 (Advancing the Customer Experience)

The goal of the ACE projects was Quality of Experience assessment and monitoring of broadband services in wireless and wireline networks, conducted together with industry partners such as A1 Telekom Austria, Telekom Austria Group, Kapsch CarrierCom and Vodafone Group.

  • NGMN P-SERQU (Service Quality Definition and Measurement, project of Next Generation Mobile Networks Alliance)

The scope of research of the NGMN P-SERQU was to identify KPIs to monitor video QoE delivered over wireless networks. One of the original project objectives was to develop a correlation between radio and core network QoS KPIs to predict QoE of streamed video. Primary reason for that objective was to provide mobile network operators a mechanism to predict QoE based on the QoS measurements obtained with existing tools in their network.

Other Projects (not related to telecommunications)

After finishing my studies, I worked at the Chair of Programming Languages and Methodology of the University of Würzburg (2012), and was involoved in two projects:

  • Digitization Workflow for Old Manuscripts (University Library of Würzburg)

In this project, a digitization workflow for the old manuscripts of the University Library of Würzburg was developed and implemented. My tasks were implementing fault- and inconsistency-tolerant document structure (i.e., page numbers, tables of contents, indexes) extraction algorithms on digitized and OCR-ed manuscripts, and a web-based form for digitization orders.

  • SpringerMaterials, SpringerReference (Springer Science+Business Media)

The goal of this project with Springer was to develop and maintain the web portals of SpringerMaterials (resource database for physical and chemical data in material science) and SpringerReference (scientific online publishing system). My tasks were search engine optimization, web analytics, as well as implementation of support for image and video galleries.