Abstract
Proof of Work (PoW) powered blockchains currently account for more than 90% of the total market capitalization of existing digital currencies. Although the security of Bitcoin has been analysed extensively, most analyses use restricted models that do not take parameters such as network delays or block intervals into account and do not allow a comparison of different blockchain instances with different consensus parameters. In this thesis, we provide a quantitative framework based on Markov Decision Processes (MDP) to analyse the security of different PoW blockchain instances with various parameters against selfish mining and double spending attacks. We devise optimal adversarial strategies for double-spending and selfish mining that take into account real world parameters such as network delays, different block sizes and block generation intervals that we capture through the stale block rate and we consider the impact of eclipse attacks. Additionally, we perform a practical security analysis of Ethereum and Stellar where we describe vulnerabilities in Ethereum and discuss the security of the current state of the Stellar network.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
