RTChain: A Reputation System with Transaction and Consensus Incentives for E-commerce Blockchain

Blockchain technology, whose most successful application is Bitcoin, enables non-repudiation and non-tamperable online transactions without the participation of a trusted central party. As a global ledger, the blockchain achieves the consistency of replica stored on each node through a consensus mechanism. A well-designed consensus mechanism, on one hand, needs to be efficient to meet the high frequency of online transactions. For example, the existing electronic payment systems can handle over 50,000 transactions per second (TPS), while Bitcoin can only handle an average of about 3TPS. On the other hand, it needs to have good security and high fault tolerance; that is, in the case when some nodes are captured by adversaries, the network can still operate normally. In this article, we establish a reputation system, called RTChain, to be integrated into the e-commerce blockchain to achieve a distributed consensus and transaction incentives. The proposed scheme has the following advantages. First, an incentive mechanism is used to influence the consensus behavior of nodes and the transaction behavior of users, which in turn influence the reputation scores of both nodes and users. That is, when a node correctly processes a transaction, it will receive the corresponding reputation value as a reward, and the reputation value will be reduced as punishment not only when the node is dishonest and violates the consensus agreement but also the transaction is not completed as required. Just like electronic transactions in the real world, the higher the reputation of the user, the more likely it is to be selected as the transaction partner. A user with a low reputation will be gradually eliminated in our system because it is difficult to complete the transaction. Second, RTChain uses a verifiable random function to generate the leader in each round, which guarantees fairness for all participants and, unlike PoW, does not consume a large amount of computing resources. Then our consensus mechanism selects the nodes with high reputation scores to reduce the number of nodes participating in the consensus, thus improving the consensus efficiency, so that RTChain’s throughput can reach 4,000TPS. Third, we built a reputation chain to implement the distributed storage and management of reputation. Finally, our consensus mechanism is secure against existing attacks, such as flash attacks, selfish mining attacks, eclipse attacks, and double spending attacks, and allows nodes that participate in the consensus to fail, as long as the reputation of the failure node does not exceed one-third of the total reputation. We build a prototype of RTChain, and the experimental results show that RTChain is promising and deployable for e-commerce blockchains.