Dynamic Game Model for Ranking Bitcoin Transactions under GSP Mechanism

Abstract

Bitcoin is the first and most successful Blockchain system so far. In the Bitcoin system, miners use transaction attached fees as a driving force to mine a new block and package transactions, while users compete by bidding transaction fees for faster confirmation. Considering the particularity of Bitcoin trading system, we take time series into consideration to analyze the transaction rules of Bitcoin system from the perspective of multiple cycles and establish a dynamic game model related to time under Generalised Second Price(GSP) mechanism, and also confirm the model's superiority on saving users' fees, compared with the static game model. Also, we propose the quantification of the user experience quantified by calculating the price difference between the transactions uploaded by the same user within adjacent times, making the transaction process of the Bitcoin system no longer the final say of the transaction price. The dynamic game model shows that there is a perfect Bayesian game equilibrium solution in the payment decision, so there is no incentive for users to change the attached fee, and the whole system is maintained stably. In addition, we verify the dynamic game model from computational experiment. Firstly, it is proved that with the help of revenue discount, the cost saving of the dynamic model is generally higher than that of the static model. Then the user's revenue under the dynamic model is showing an upward trend, and the transactions order under the dynamic model is more stable than that under static model, which can be illustrated mathematically and computationally that the proposed dynamic game model in this paper will help all transactions be processed more efficiently in a uniform pipeline.