Research On Double-Layer PBFT Consensus Based On Trust Management

As the communication network enters the 5G era,IOV(Internet-Of-Vehicle)as an intelligent network infrastructure has gradually become an important technology that is inseparable from people’s life and travel.IOV collects and shares information through vehicles to form a huge data network and realize the interconnection between vehicles.However,there are still problems of secure data sharing in IOV,such as privacy leakage,inaccurate information,etc.The existing mature business models often adopt centralized data management,which cannot well solve the data sharing and security problems of large-scale distributed IOV system.So,distributed decentralized blockchain technology becomes a hot research topic.Facing the demand of IOV,how to design the consensus mechanism in blockchain is a key point.The traditional PO W consensus mechanism is difficult to be applied in the IOV system with limited computation and communication resources due to the high demand of computation resources,while the state machine based PBFT consensus also has the problems of high communication overhead and too arbitrary selection of master nodes,and the IOV environment also has high requirements on the efficiency of consensus.To address the above problems,this paper conducts research on PBFT mechanism from two aspects of consensus protocol,security and consensus efficiency,and the results achieved are as follows.(1)In the traditional PBFT consensus algorithm,there is a drawback that the selection of the master node is too arbitrary and the malicious behavior of the consensus node does not have a penalty mechanism.In view of these two drawbacks,the paper proposes a set of trust management model for consensus nodes and a method for selecting the master node of consensus mechanism.The paper gives a trust management model for the IOV environment,establishes the trust factor,gives the trust value calculation method,establishes the reward and punishment mechanism for each consensus node,and takes the trust level of the vehicles around the node into consideration.Finally,for the characteristics of IOV network,the master node is jointly selected by integrating the trust level of nodes themselves,the trust level of vehicles and the busy degree of nodes,which effectively improves the fault tolerance and efficiency of consensus.(2)To address the problems of high communication overhead and low consensus efficiency of traditional PBFT consensus algorithm when the number of nodes is large,a double-layer PBFT consensus algorithm based on trust management,referred to as bPBFT,is proposed.This consensus algorithm adopts a hierarchical structure,divides nodes into upper and lower layers,and uses a dynamic threshold-based batch consensus strategy to design and finally simplifies the consensus protocol of the upper layer.Compared with the existing double-layer PBFT consensus,it can effectively reduce the probability of consensus blocking between the lower and upper layers,improve the consensus throughput,and reduce the communication overhead.Finally,theoretical analysis and experimental simulation test results show that bPBFT consensus can effectively reduce the probability of consensus blocking between lower and upper layers,reduce communication overhead,and effectively improve the throughput and consensus efficiency in a large number of nodes environment compared with the existing two-layer PBFT consensus,which has better performance than the existing two-layer PBFT consensus scheme.