Analysis of Symmetric Cryptographic Algorithms

Nowadays computers have become an integral part of our lives. The Internet has changed the way we do everything, from banking to allowing us to work from home during a pandemic. More and more data is being transmitted through the Internet. The data might be sensitive and therefore we need to ensure the integrity and confidentiality to have a secure Internet.

Cryptography provides the tools necessary to secure communication on the Internet. This thesis deals with the analysis of block ciphers. These block ciphers are used for many applications due to their efficiency, from bulk encryption to IoT devices. How secure these block ciphers are is therefore fundamental for the security of the internet.

The way we measure the security of a block cipher is by trying to break parts of it and seeing how far we can get. Thus the more security analysis we do on these block ciphers, the more confidence we have in their security. Understanding how secure they are is therefore critical to our communication systems.

This thesis comes in two parts. The part introduces block ciphers and various techniques for their analysis. The second part contains four publications written during the PhD studies. The first publication presents a tool based on the division property which allows for rapid evaluation of a cipher. We used this tool to find several new and improved distinguishers.

The second publication presents a new quantum distinguisher for type-3 generalized Feistel networks based on Simon’s and Grover’s algorithms. The third publication is a formal analysis of the probabilities involved in boomerang attacks. In particular we prove that the common independence assumption is inherently flawed.

The fourth publication analyzes the re-identification attack, which is a problem that can arise when using format preserving encryption to encrypt a database.