Improvement of Multi-Mechanism PBFT Consensus Algorithm Based on Attribute-Based Encryption Grouping

The PBFT consensus algorithm is widely used in consortium blockchains due to its good performance in Byzantine fault tolerance, but problems still exist in some aspects such as random election of master nodes, high communication overhead, and poor network scalability. Many researchers use node grouping to solve the above problems, but there are still some deficiencies in the credibility mechanism, grouping algorithm and consistency mechanism. To address the shortcomings of existing researches, this paper proposes a multi-mechanism PBFT consensus improvement algorithm based on attribute-based encryption grouping (ABE-MPBFT). Firstly, based on the nodes' positivity, honesty and computational ability, we design the credibility evaluation mechanism to establish the grouping criteria, then construct the credibility reward and penalty mechanism to update the nodes' grouping. Secondly, attribute sets are assigned to nodes, and node grouping is spontaneously realized using CP-ABE in the process of consensus message encryption and decryption transmission, which solves the problem of large computational overhead and reuse of traditional grouping algorithms. According to different node properties of inter-group and intra-group, we design two-phase, three-phase and four-phase consistency mechanisms for each group to further reduce the communication overhead. Finally, it is proved through experiments that this algorithm can effectively reduce the number of communication, and improve consensus delay and TPS.