Národní úložiště šedé literatury Nalezeno 53 záznamů.  1 - 10dalšíkonec  přejít na záznam: Hledání trvalo 0.00 vteřin. 
Centralized Cryptocurrency Exchange with Trusted Computing
Sasák, Tomáš ; Očenášek, Pavel (oponent) ; Homoliak, Ivan (vedoucí práce)
In this thesis, we propose, design, implement, and analyze a centralized cryptocurrency exchange using trusted computing. Most popular exchanges work in a way that requires the customer to fully rely on the exchange operators. Decentralized exchanges work on top of existing blockchains, which limits the number of tradable coins. Using a trusted computing platform, Intel SGX, it is possible for operators to make code open source, and the customer can attest that the exchange platform is running the published exchange code. Also, exchange private keys are encrypted in memory and safely stored on disk using SGX sealing. Combining this approach with a smart contract platform on a public blockchain, we can achieve non-equivocation of the exchange and create a better layer of security and transparency between the customer and the exchange. The exchange was implemented as a proof-of-concept using the programming language Go, using the EGo framework to build SGX enclave applications. For data storage, we combine key-value storage PebbleDB with relation database PostgreSQL, and exchange attempts to secure the integrity of these data. The exchange smart contract was implemented in Solidity and deployed on the Ethereum development environment. Our concept implementation is able to handle around 35 deposits per second and around 23 bids matched per second. We performed a security analysis on this model from outside, but also from inside. The single ledger update from the exchange in the smart contract costs 0.00255 ETH, which in total costs 3.6742 ETH per day.
Detekce anomálií u aplikačních firewallů
Pospěch, Jan ; Homoliak, Ivan (oponent) ; Očenášek, Pavel (vedoucí práce)
Cílem předložené bakalářské práce je popis procesu detekce anomálií u aplikačních firewallů. Práce je zaměřena na principy a základy detekce anomálií, čtenář se seznámí především s technikami a metodami strojového učení. Je popsán proces analýzy požadavků a odpovědí získaných ze systému pro ochranu webových aplikací a s jejich pomocí je vytvořen návrh systému. V praktické části je popsána implementace systému a testování na reálných datových sadách. Nejlepší výsledky vykazují algoritmy Decision tree a Random forest s hodnotou f1-score 0.9987. Z metod učení bez učitele nejlepší výsledky vykazuje Autoenkódér s hodnotou f1-score 0.8315.
Decentralizovaná aplikace pro Question/Answer sekci využívající blockchain
Čuhanič, Jakub ; Očenášek, Pavel (oponent) ; Homoliak, Ivan (vedoucí práce)
Tato práce se zaměřuje na vytvoření decentralizované mobilní aplikace, která nabízí Q&A sekci. Problém transparentnosti procesu dotazování v Q&A sekcích, cenzury otázek a spamu je vyřešen pomocí technologie blockchainu a jeho napojení na systém pro správu identit. Výsledné decentralizované aplikace jsou řešením na všechny tyto problémy. Jejich využití může být kdekoliv, kde je potřeba Q&A prostor, především ve veřejnoprávních médiích nebo na internetu. Díky všem vlastnostem aplikací vyplývajících z blockchainu by se mělo dosáhnout stavu, ve kterém nebudou moci moderátoři relací manipulovat ani cenzurovat uživatele a jejich otázky.
Secure and Efficient State Preservation in Ethereum-Based Smart Contract Platforms
Eršek, Martin ; Perešíni, Martin (oponent) ; Homoliak, Ivan (vedoucí práce)
This thesis presents a solution to mitigate state-related challenges in Ethereum. By analyzing Ethereum internals and relevant literature, the identified problems are addressed. A proof-of-concept implementation showcases promising results, such as up to a 33% reduction in membership proof size and nearly a 50% decrease in mean transaction execution time compared to existing solutions. Furthermore, the proposed solution enables effective parallel processing with over a 97% chance of non-conflicting execution for 8 random native transfer transactions. Future work involves implementing a parallel execution client and integrating the solution into production, advancing Ethereum's state management capabilities.
An Empirical Evaluation of Selfish Mining Strategies on Various Blockchains
Kubík, Ján Jakub ; Košťál,, Kristián (oponent) ; Homoliak, Ivan (vedoucí práce)
The aim of this thesis is to enhance blockchain security by deepening the understanding of selfish mining attacks across various consensus protocols. Previous research was mainly focused on single protocols with one attacker, and limited studies have been conducted on multiple attackers in the Nakamoto consensus protocol. To address this gap, a discrete, static, stochastic simulation framework was developed to analyze selfish mining with multiple attackers in diverse consensus protocols, including Nakamoto, Subchain, and Strongchain. Through the use of this framework, the thresholds for successful selfish mining on Nakamoto were replicated, the Strongchain thresholds were verified, and a new threshold for two attackers in Strongchain was discovered. In the case of Subchain, independent, selfish mining on weak and strong blocks was investigated, revealing profitability in strong blocks with slightly higher thresholds than Nakamoto. The main contribution of this thesis lies in the development of a framework with verified thresholds and the exploration of selfish mining for the studied protocols, which can be easily extended for new consensus protocols, thereby enabling researchers to evaluate a protocol's resilience against selfish mining attacks.
Analysis of the Move Programming Language for Blockchain Platforms
Šmehýl, Adam ; Homoliak, Ivan (oponent) ; Perešíni, Martin (vedoucí práce)
This thesis studies the Move programming language, focusing on its usability for developing applications (smart contracts or programs) on blockchain platforms. Two key aspects are considered: first, a comparison of programming models using Move to widely used models of EVM-compatible platforms like Ethereum and the popular Solana blockchain; and second, the implementation of the same program in Solidity on Ethereum, Rust on Solana, and Move on Aptos. Criteria for comparison include deployment and execution costs, processing speed, code readability, and overall development experience. A detailed analysis of Move's unique features, such as resource management, the use of generics, and other security enhancements in programming, is conducted. The results demonstrate Move's potential for extensive use in the blockchain field, with its strong emphasis on secure coding and resource management contributing to the growing interest within the blockchain community.
Anomaly Recognition in Advanced Detection Systems
Poposki, Vasil ; Homoliak, Ivan (oponent) ; Očenášek, Pavel (vedoucí práce)
The objective of this work is to implement an anomaly detection system using artificial intelligence techniques that can detect anomalies by learning the system behavior. The proposed approach is effective in identifying novel or unknown anomalies that traditional rule-based methods may miss in network traffic data. However, the implementation of such a system involves addressing challenges such as data processing and feature extraction. This work discusses different methods of data analysis and intrusion detection approaches in Extended Detection and Response systems and the challenges we face in today’s expanding security technologies.
Anomaly Detection by IDS Systems
Gawron, Johann Adam ; Homoliak, Ivan (oponent) ; Očenášek, Pavel (vedoucí práce)
The goal of this thesis is to familiarize myself, and the reader, with the issues surrounding anomaly detection in network traffic using artificial inteligence. To propose and subsequently implement a methodology for creating an anomaly classifier for network communication profiles. The classification method should be able to efficiently and accurately identify anomalies in network traffic to avoid generating false outputs. During the research of the issue, IDS systems, various types of attacks, and approaches to anomaly detection and classification were examined. In evaluating the effectiveness, several standard methods were examined and used to express the quality of classifiers.
Incentive Strategies for Transaction-Fee Regime of Proof-of-Work Blockchains
Budinský, Rastislav ; Perešíni, Martin (oponent) ; Homoliak, Ivan (vedoucí práce)
In this thesis, we review the undercutting attacks in the transaction-fee-based regime of Proof-of-Work (PoW) blockchains with the longest chain fork-choice rule. Next, we focus on the problem of fluctuations in mining revenue and the mining gap -- i.e., a situation, in which the immediate reward from transaction fees does not cover miners' expenditures. To mitigate these issues, we propose a solution that splits transaction fees from a mined block into two parts -- (1) an instant reward for the miner of a block and (2) a deposit sent to one or more Fee-Redistribution Contracts ($\mathcal{FRC}$s) that are part of the consensus protocol. At the same time, these redistribution contracts reward the miner of a block with a certain fraction of the accumulated funds of the incoming fees over a predefined time. This setting enables us to achieve several interesting properties that are beneficial for the incentive stability and security of the protocol. With our solution, the fraction of \textsc{Default-Compliant} miners who strictly do not execute undercutting attack is lowered from the state-of-the-art result of 66\% to 30\%.
Detekce anomálií ve Wi-Fi komunikaci
Lička, Zbyněk ; Homoliak, Ivan (oponent) ; Očenášek, Pavel (vedoucí práce)
Práce se zabývá detekcí anomálií při komunikaci pomocí technologie IEEE 802.11 (Wi-Fi) na linkové vrstě OSI. K detekci anomálií byla zvolena metoda neuronových sítí, konkrétně LSTM rekurzivních neuronových sítí. Na začátku je popsána oblast zaměření a motivace k detekci anomálií v prostředí počítačových sítí. Poté jsou popsány různé techniky detekce anomálií, které jsou v oblasti počítačových sítí běžně používány. Pokračuje analýza požadavků na systém umožňující detekci anomálií a volba vhodné metody pro tyto účely. Dále je popsán návrh a způsob implementace systému a zvoleného modelu. Poté pokračuje zpráva o testování, vyhodnocení experimentů a diskuze.

Národní úložiště šedé literatury : Nalezeno 53 záznamů.   1 - 10dalšíkonec  přejít na záznam:
Chcete být upozorněni, pokud se objeví nové záznamy odpovídající tomuto dotazu?
Přihlásit se k odběru RSS.