Hide menu

SaS Seminars

Software and Systems Research Seminar Series


The SaS Seminars are a permanent series of open seminars of the Division of Software and Systems (SaS) at the Department of Computer and Information Science (IDA), Linköping University. The objective of the seminars is to present outstanding research and ideas/problems relevant for SaS present and future activities. In particular, seminars cover the SaS research areas software engineering, programming environments, system software, embedded SW/HW systems, computer systems engineering, realtime systems, parallel and distributed computing, and theoretical computer science. - Two kinds of seminars are planned:

  • talks by invited speakers not affiliated with SaS,

  • internal seminars presenting lab research to whole SaS.

The speakers are expected to give a broad perspective of the presented research, adressing the audience with a general computer science background but possibly with no specific knowledge in the domain of the presented research. The normal length of a presentation is 60 minutes, including discussion.

The SaS seminars are coordinated by Ahmed Rezine.



Recent / Upcoming SaS Seminars (2018)



Towards next-generation collaborative network defense

Dr. Emmanouil Vasilomanolakis, Technical university of Darmstadt, Germany.

Tuesday, June 19th, 2018, kl 13:15 room Alan Turing.

Abstract:
Cyber-attacks have nowadays become more frightening than ever before. The growing dependency of our society on networked systems aggravates these threats; from interconnected corporate networks and industrial control systems to smart households, the attack surface for the adversaries is increasing. I argue that network security requires, besides traditional methods, more out of the box and (pro)active research. In this talk, I will discuss the three main pillars of my current and future research, namely: i) offensive security, ii) attacker deception, and iii) collaborative intrusion detection. In particular, the first part of the talk will focus on offensive security (e.g., taking the perspective of an adversary to anticipate new attacks) and attacker deception (e.g., honeypots) and will examine the motivation, the challenges and possible next steps for this line of research. The second part of the talk will be emphasizing on collaborative intrusion detection and will combine a discussion of research challenges along with lessons learned from a number of projects that I have been involved to date.

Bio of speaker:
Dr. Emmanouil Vasilomanolakis is a senior researcher at Technische Universität Darmstadt. His research interests include collaborative intrusion detection, honeypots and botnet monitoring. Emmanouil received a PhD, for his thesis "On Collaborative Intrusion Detection", from the Technische Universität Darmstadt (Germany) in 2016 and a diploma (Dipl.-Inform.) and MSc (IT Security) from the University of the Aegean (Greece) in 2008 and 2011 respectively.




Customization methodologies for embedded systems

Dr. Lazaros Papadopoulos National Technical University of Athens, Greece.

Thursday, May 24th, 2018, kl 13:15 room Alan Turing.

Abstract:
The efficient deployment of applications in embedded devices is challenging, due to the limited resources that the embedded systems provide and the high requirements of embedded applications. Customizing applications based on underlying platform specifications often provides significant gains in terms of performance and energy efficiency. In this presentation, various customization methodologies for dynamic applications and Convolutional Neural Networks will be described, which are applicable in various levels of abstraction.

Bio of speaker:
Dr. Lazaros Papadopoulos is a Research associate in Microprocessors and Digital Systems Lab of the School of Electrical and Computer Engineering of National Technical University of Athens. His research interests include runtime optimisation methodologies for low energy consumption in embedded systems and special-purpose embedded processors. He co-authored 17 publications in international journals and conferences and participated in several EC projects (MNEEME, 2PARMA, AEGLE).




Fog Computing with Reference Attribute Grammars

Prof. Dr. Uwe Aßmann Technical University of Dresden, Germany.

Friday, April 20th, 2018, kl 14:15 room Alan Turing.

Abstract:
Fog Computing is a new reference architecture for sensor networks at the edge of the cloud. Its basic idea is to process data locally, in complex sensor nodes, gateways and routers, before moving them into the cloud. Thereby, it protects privacy by default and delivers speed (low latency, high bandwidth). Its business perspectives for European companies are tremendous, because starting from the sensor markets, they can penetrate the future cloud computing markets. Fog Computing has a very interesting application area, Robotic Co-Working. The talk reports about a case study in fog computing: the Chair of Software Engineering has combined several co-working robots, such as the KUKA LBR iiwa or the UR-10, with a sensor-equipped jacket and glove. A fog collects and aggregates all sensor data via a gateway, a laptop, and a robot server. Starting from this example, we show how to program a fog, with a world statechart and an adaptive software platform based on contexts and roles. Fog computing will enable life with wearables and machines in smarter rooms. Fogs mean smartness. Fogs can be programmed with event-based RAGs if events are admitted as input to an attributed tree. We give some examples of the models that are required for modeling fogs.

Bio of speaker:
Uwe Assmann is professor for software engineering at the Technical University of Dresden, Germany. He is also a former member of PELAB. More information can be found at http://www1.inf.tu-dresden.de/~ua1/




Real-Time Fault Localisation using Machine Learning and Microservices Principles

Armin Catovic Technology Specialist, Systems & Technology, Ericsson AB.

Monday, January 22nd, 2018, 13:15 room Alan Turing.

Abstract:
Linnaeus is a lightweight microservice that implements simple yet powerful machine learning principles for fault detection and localisation on a continuous log stream. It can be used in both standalone telecom nodes as well as virtualised network functions (VNFs). This seminar covers both the machine learning aspects as well as software design principles. In terms of software design, Linnaeus exposes the current microservices trend seen in the telecom industry, as a natural next step in network function virtualisation and general software architecture.

Bio of speaker:
Armin has been at Ericsson last 10 years. Originally from Melbourne (Australia) he completed a double degree in Computer Science and Telecom Engineering. Armin started his career as a contract Delphi programmer and consultant, before finally entering Ericsson. He worked in Ericsson's global services unit on large-scale 3G/WCDMA radio integration projects in Australia, Singapore, Indonesia and Bangladesh, before moving to Linköping R&D on the then-new 4G/LTE radio system. He worked as a system tester, developer and systems engineer in Linköping, before moving to Stockholm where he is currently working as a senior developer in the context of machine learning and AI.




Previous SaS Seminars



Page responsible: Christoph Kessler
Last updated: 2018-05-28