Increasing Bitcoin Revenue by Leveraging Rational Miners

Abstract

Blockchain technologies have received a lot of attention over the past years. The fundamental part of each blockchain is the underlying consensus mechanism that ensures the blockchain peers agree on the state of the blockchain ledger. One of the most popular blockchains, Bitcoin is a Proof of Work(PoW) based blockchain whose underlying fork choice rule is called the longest chain. In longest chain-based blockchains a popular attack strategy is selfish mining. In selfish mining, an attacker does not immediately publish the block. It keeps the adversarial fork private and strategically decides on the publishing time of the blocks included in this private fork. Selfish mining only applies when the attacker’s mining share is greater than 25%. Miners can collude with each other to achieve the lower bound for revenue gain. But, in the first epoch of selfish mining the block generation ratio does not increase for the selfish miner. Therefore in the first epoch, the revenue of a selfish miner does not increase compared to the revenue when it is mining honestly. Hence, miners may not agree to collude to avoid their loss. In this thesis, we propose the Risk-Free Collusive Selfish Mining strategy. In this new strategy, the attacker is incentivizing rational miners to follow its strategy. The strategy is designed in such a way that the collaborators do not lose anything, hence the rational miners are more likely to follow the attacker’s strategy. We have shown that an attacker with 20% or more computational power can gain more revenue than their fair share when colluding with collaborators who have 10% mining power. Though Risk-Free Collusive Selfish Mining is a viable strategy, there is one risk if the collaborators publish the shared block. Therefore, we propose Transaction Exclusion Attack strategy where the non-compliant miners cannot ruin the attacker’s strategy. In this strategy, the attacker only publishes the header instead of a full block. If there is a header in the system a dilemma between the rational miners arises regarding the choice between two strategies : (1) Mine an empty block on top of a block for which only header is published, (2) Mine on top of the previous block, in this attack scenario. We have analyzed if the rational miners are incentivized if they choose to mine on top of a block for which only header is published. The plot of the utility function shows that mining on top of a header benefits rational miners as long as the ratio of transaction fees and total fees is less than a certain value.