Research Of Block Withholding Attack Mitigation Strategy Based On Game Theory

With the emergence of specialized mining machines such as Application-specific integrated circuit,the computing power of miner in the bitcoin network is enhanced,and the difficulty of mining is correspondingly increased.The miners joining the pool are the most common way of mining in the bitcoin network.However,there is a block withholding attack between the pools.The pool attack will increase the revenue,but the revenue of mutual attack is lower than that not attacking each other.The attack or not is the dilemma of the pool.This is the proof of work consensus algorithm's mining dilemma corresponds to the prisoner's dilemma in game theory.Block withholding attack not only damage the interests of the pool,but also pose a threat to the stability of the Bitcoin network.The mitigation strategy for block withholding attack in single pool is not applicable to block withholding attacks between pools.Therefore,in this thesis,the sub-strategy of Zero-determinant(ZD)is set for the mining dilemma of pools: pinging strategies,extortion strategies,and tolerance strategies optimize the pool's revenue in dilemma and optimize pool's strategy choices to mitigate attacks.This thesis validates the validity of the ZD sub-strategy based on the Iterated Prisoner's Dilemma model in the Python environment.Simulation results show that the pool adopts a pining strategy to unilaterally control the revenue of the attacking pool,and control its revenue between no-attack and mutual attack;the revenue of the pool using the extortion strategy is higher than that of the attacking pool,or the same revenue as the attacking pool.The adoption of a tolerant strategy can promote the selection of no-attack strategies in the pool,so that the game between the pools does not attack each other.Therefore,the use of the ZD sub-strategy can improve the profit in the dilemma of the pool,and can also avoid the dilemma of the pool falling into mutual attack and achieve the purpose of mitigating the attack.