| Description: |
Several master thesis projects will be offered within the scope of the InSecTT (Intelligent Secure Trustable Things) project, a pan-European effort with 52 key partners from 12 countries. The main goal in the InSecTT project is providing intelligent, secure and trustworthy systems for industrial applications, in particular for manufacturing industry. To achieve this goal, combination of different technologies and techniques such as Artificial Intelligence (AI), Machine Learning (ML), cybersecurity and Internet of Things (IoT) are considered in the project.
Historically industrial devices would have been isolated and they were theoretically air-gapped and didn't connect to anything. On an IIoT network, it's possible that what was previously a separate network for operational technology, responsible for monitoring and controlling physical devices such as pumps and valves, could be converged with the rest of the information technology network. While this can have benefits, the risk is that critical software that once ran in its own secure environment is now linked to a broader network, creating an easier target for hackers.
Intrusion Detection Solutions (IDS) for IIoT need to be customized to the nature of the devices. Small devices with limited resources need a solution tailored to the types of attacks they are likely to experience without overwhelming the limited memory and computing resources of the device. At the same time, the sophistication of the IDS must scale up to support more powerful gateway and control systems. The key is to monitor for, detect, and quickly report anomalous situations. This requires integration with a security management system where IDS events can be sent and viewed to determine if the anomalous events indicate a cyber-attack. Can we use machine learning to differentiate the attack packet flows, classify attacks and eventually stop them? Can we allow the learning algorithm to understand new kind of attacks and grow in intelligence? Can we use multi-agent approach? There is a need for an intelligent IDS that could detect different and dynamic attack patterns, and this research will develop such a system.
The project will be performed in close cooperation with other partners in the project, which include large industries, research institutes and universities, as well as small and medium-sized enterprises from both Sweden and the EU.
|
| Prerequisites: |
Minimum qualifications:
• Master studies in computer science or computer networks or in a related field
• Have experience or deep interest in network security, machine learning and artificial intelligence algorithms.
Preferred qualifications
• Have a background and experience in operations research, simulation, statistics, algorithms;
• You are curious, goal-oriented, flexible, ambitious and communicative;
• Are skilled in presenting and documenting your work in English, i.e. to describe the problem, the solution and the approach in an understandable manner;
• Have excellent programming skills (e.g. C, C++, Java and Python);
• Ability to read and understand state of the art research in network security and machine learning
|
