Cyber security thesis topics

More potential advisors

In Tallinn: Jüri Kivimaa, Rain Ottis, Andres Ojamaa, Tanel Tammet

In Tartu(UT): Dominique Unruh, Helger Lipmaa, Sven Laur, Bingsheng Zhang, Raimundas Matulevičius, Meelis Roos

AS Cybernetica (Tallinn): Aivo Kalu

AS Cybernetica (Tartu): Peeter Laud, Dan Bogdanov, Jan Willemson

You can look at the research background of the potential advisors at the Estonian Research Portal if you don't know the people from the courses. Many of the listed above are in the different fields of cryptography.

Proposed topics for thesis

Secure e-mail communication for Web shops

Advisor: Andres Ojamaa (andres.ojamaa@cs.ioc.ee)

The majority of present Web shops send confirmations of orders and other communication via unencrypted e-mail. However, these letters may contain and reveal sensitive information, such as the customer's home address, parcel tracking numbers or personal preferences. The goal of this project is to propose a general solution for protecting (security conscious) customers' privacy in Web shops by using e-mail encryption. In addition, a proof-of-concept implementation based on a free open source e-commerce platform should be developed. The proposed solution can employ existing standards and technologies, such as OpenPGP and GnuPG.

Network traffic generation for cyber defence exercises

Advisor: Andres Ojamaa (andres.ojamaa@cs.ioc.ee)

Cyber defence exercises need a lab environment that resembles a typical real world set-up as closely as possible. This includes infrastructure, such as computer networks, SCADA systems, operating systems, applications and server software, but also simulated network traffic. The characteristics of simulated traffic should look realistic to make the exercise challenging. This project includes developing and implementing software tools, for example, a web browser add-on module, to be used in such exercises to simulate realistic network traffic. This project may integrate traffic models published in the literature, and build on the results of initial prototyping.

Interactive hybrid network simulation based on Rich Components

Advisor: Andres Ojamaa (andres.ojamaa@cs.ioc.ee)

Large-scale computer network simulation is a computationally heavy task. To alleviate this, hybrid network simulations may be utilized that combine packet level and flow based simulation techniques for achieving an acceptable trade-off between performance and accuracy. Furthermore, interactivity of simulations makes experimentation and exploration of possible parameter space more efficient and fun. The goal of this project is to develop a large computer network model for simulating and analysing network traffic generated by malware outbreaks, botnets etc. The model would be based on powerful software components -- rich components -- that support visual programming, automatic synthesis of programs and interactivity of hybrid simulations. This project is based on and extends preliminary results from earlier promising experiments.