Evaluating Attack Thresholds in Proof of Stake Blockchain Consensus Protocols

Abstract

This research paper delves into the analysis of attack thresholds within Proof of Stake (PoS) blockchain consensus protocols. In the PoS system, validators participate by staking a certain amount of funds, granting them the ability to propose and vote on new blocks in the network. The influence of validators in the consensus process is determined by the proportion of funds they have bonded, thus wielding voting power commensurate with their stake percentage. A critical vulnerability in PoS systems lies in the possibility of a single validator reaching the attack threshold of 1/3 stake, which grants them the capability to disrupt network consensus. This study examines the potential of a validator reaching the attack threshold through two distinct methods. Firstly, I explore the Superior Returns Attack Threshold (SRAT) method, wherein a network member aims to increase their token ownership by investing in high-risk, high-return securities. The objective is to accumulate sufficient wealth to stake and attain an attack stake. Secondly, I investigate the Random Attack Threshold (RAT) method, whereby a validator inadvertently accumulates an attack stake due to stochastic randomness. My analysis reveals that intentional efforts to obtain 1/3 stake by a validator through the SRAT method yield a nonzero probability, posing potential security challenges for pure PoS systems. However, I demonstrate that the probability of a validator acquiring 1/3 stake through stochastic randomness with constant rewards, as observed in the RAT method, is nearly zero.