Assessing blockchain selfish mining in an imperfect network: Honest and selfish miner views

Abstract

Bitcoin and Ethereum, the most famous blockchain-based cryptocurrencies, both use Proof-of-Work protocol to achieve consensus, which is vulnerable to selfish mining. The existing researches on selfish mining analysis focused on Bitcoin, or ignored the blockchain details, or only investigated mining revenue without considering system performance and security.This paper aims to quantitatively evaluate the influence of selfish mining in an imperfect blockchain network from the perspective of honest miners (system performance and security) and selfish miners (selfish mining revenue ratio). We develop a novel Markov model to capture the behaviors of honest and selfish miners in both Bitcoin and Ethereum. Our model can also capture natural forks (occurring due to block propagation delay) and the varying distance between uncle and nephew blocks in Ethereum. Formulas are derived to calculate mining revenue, system performance and security metrics. The proposed model and metric formulas are validated by (1) comparing our numerical with simulation results, and (2) comparing our numerical results with the existing work results. Numerical analysis is carried out to investigate selfish mining impact over diverse parameters. These quantitative results can help detect whether there are any selfish miners in the system, help design blockchain reward mechanisms and enhance security.