BM-PAW: A Profitable Mining Attack in the PoW-based Blockchain System

Abstract

Mining attacks enable an adversary to procure a disproportionately large portion of mining rewards by deviating from honest mining practices within the PoW-based blockchain system. In this paper, we demonstrate that the security vulnerabilities of PoW-based blockchain extend beyond what these mining attacks initially reveal. We introduce a novel mining strategy, named BM-PAW, which yields superior rewards for both the attacker and the targeted pool compared to the state-of-the-art mining attack, PAW. BM-PAW attackers are incentivized to offer appropriate bribe money to other targets, as they comply with the attacker's directives upon receiving payment. We further find the BM-PAW attacker can circumvent the miner's dilemma through equilibrium analysis in a two-pool BM-PAW game scenario, wherein the outcome is determined by the attacker's mining power. We finally propose practical countermeasures to mitigate these novel pool attacks.