Martin Perešíni from the Institute of Intelligent Systems will defend his dissertation in June

We invite you to the dissertation defense of Ing. Martin Perešíni from the Department of Intelligent Systems, FIT, BUT, which will take place on Monday, June 29, 2026, at 12:00 p.m. in conference room G108. The thesis advisor for the dissertation titled “Cross-Layer Security Analysis of Blockchain Systems” is Doc. Ing. Ivan Homoliak, Ph.D.

Perešíni earned his engineering degree at FIT, Brno University of Technology. Today, he is a member of the BlockSec@FIT (Blockchain and System Security Research Group) and Security@FIT research groups and teaches the course “Blockchains and Decentralized Applications” at the faculty. He has a long-standing collaboration with international institutions, including Prof. Muhammad Shafiqu of New York University Abu Dhabi and Mario Larangeira of the Institute of Science Tokyo. Throughout his professional career to date, he has received numerous awards. To name just a few, he won Best Paper in the CryptoEx track at the IEEE ICBC 2024 conference; at the ETHReS 2026 conference, he and the students he advised received two awards: for Best Paper and Best Poster.

In his research, Martin Perešíni focuses on blockchain as a comprehensive, multi-layered security ecosystem—not merely as a cryptographic protocol. When people hear the term “cryptocurrency attack,” most imagine a hacker “breaking” the encryption. However, in his work and research, Martin Perešíni shows that in practice, blockchains fail in exactly this way only very rarely. Much more often, the problems stem from poorly designed economic incentives, flawed assumptions in the protocol design, and shortcomings in implementation. And here’s the key point: These vulnerabilities are usually hidden at the boundaries between the layers of a complex system. Perešíni’s work therefore relies on a method known as cross-layer analysis. This means that he does not examine security in isolation, but across the layers of the system — because a blockchain is not an indivisible whole. It consists of layers resembling the floors of a building: We’re talking about the consensus layer (how thousands of computers around the world agree on a single common result without a “central authority”), the network layer (how nodes send data to one another), and the application layer (wallets and applications that the average user interacts with). Blockchain research typically examines each “floor” separately. In his work, Perešín shows that the most dangerous vulnerabilities, however, arise at the junctions between the “floors,” where the assumptions of one layer clash with the reality of another.

To substantiate his claim, he combined several approaches: game theory (where participants are modeled as players seeking their own profit), extensive computer simulations using real-world data (with more than 8,000 simulated nodes), the design and prototyping of new protocols, practical security research on 36 real-world cryptocurrency wallets, and advanced cryptography in the form of zero-knowledge proofs (zk-SNARKs). These approaches led to the following main conclusions of the study. Several of them pertain to so-called selfish mining — that is, a situation in which a participant secretly withholds discovered blocks to gain an unfair advantage. The prevailing belief (derived from Bitcoin’s practice) was that such an attack is only profitable if one controls roughly one-third (33%) of the network’s computing power. However, Perešín’s analysis involving multiple attackers shows that if five attackers strike simultaneously, the profitability threshold drops to just about 14–17% for each. Attacks thus become economically viable much sooner than previously assumed. Furthermore, Perešíni has formally proven that the popular “random transaction selection” rule in the new generation of so-called DAG blockchains (which, for greater speed, store blocks not in a single chain but in a branching graph) does not hold up from an incentive perspective: A rational participant always has a reason to deviate from the rules. The fact that the problem persists even on a realistic scale was confirmed by the aforementioned large-scale simulations. At the intersection of the consensus and network layers, Perešíni then contributed to the design of a Proof-of-Stake protocol with built-in “onion routing” — that is, the same principle of layered encryption on which, for example, the Tor anonymization network is based. This makes it difficult to determine who specifically proposed a block, while maintaining throughput. At the application layer, he and his colleagues developed a formal classification of how cryptocurrency wallets actually verify users. The unsettling finding is this: Despite interfaces that appear to have multiple security steps, most of the 36 wallets examined actually offer only single-factor authentication confirmed by the blockchain. The final piece of the “puzzle” in the conclusions of Perešíni’s dissertation and the research as a whole addresses wallet synchronization on mobile devices using zk-SNARKs — cryptographic proofs that allow one to verify that something is true without revealing the data itself, and which can be verified very quickly. The benefit here is very concrete: storage requirements on the phone will drop from 71 MB to 3.5 MB, and users will no longer need to trust third-party servers, as correctness will be guaranteed directly by mathematics (i.e., cryptography).

Martin Perešíni’s research scope is broad. He first became interested in cybersecurity issues related to blockchain toward the end of his engineering studies. At the time, he was considering whether to pursue a doctoral program at FIT or to take a job in industry. “But I felt that since I had already graduated with hard work and good results, it would simply be a shame to go into a corporation and deal with operational matters. I started approaching individual advisors with questions about possible dissertation topics. And I was also introduced to Associate Professor Homoliak, who was doing a postdoc in Singapore at the time and was returning to Brno. While abroad, he had been intensively studying blockchains, and after his return, he got me excited about this topic as well,” Perešíni describes the beginnings of his doctoral studies. What drew him to blockchain was the novelty of the technology from the perspective of computer security and its enormous, still untapped potential. “Blockchain technology is still, I’d say, finding its way. It hasn’t found its true use case yet. Moreover, it’s primarily associated in the media with ‘pump-and-dump’ schemes, and there are also many widely held myths surrounding it, such as the absolute anonymity of transactions and users. Nevertheless, interesting concepts like central bank digital currency — which would already leverage central bank authority—are currently being intensively discussed,” Perešíni comments on the current state of the issue. He sees a number of interesting technical challenges in this field in the near future — blockchain is, in his words, a sophisticated way “to do trustworthy things in an untrustworthy environment,” with the help of its pillars: mathematics/cryptography, distributed systems, and economic incentives. In an era of massive AI adoption, this may be the path to ensuring security, privacy, and verifiability. Specifically, he mentions trusted computations, digital tokenization of physical assets, and the issue of consensus protocols — which allow independent nodes to agree on the state of a database, the order of transactions, and so on, even without a central authority — as topics for the near future. “For example, Associate Professor Homoliak and I are working on the Proof of Useful Work consensus protocol, which aims to transform the Proof of Work concept from energy-intensive ‘hash guessing’ into a useful computation — specifically, the generation of zk-SNARK proofs,” says Perešíni, highlighting one of their current research challenges.

And how does Martin Perešíni see his immediate future? “I’m in the process of negotiating a potential postdoc position, and I’d obviously be happy to take it. There are plenty of research tasks ahead of us,” Perešíni concludes the interview about his academic career to date.

We wish Martin Perešíni a successful defense and the best of luck in his future career.