Research On Key Technologies Of Scalable Blockchain Based On In-chain Optimization

In recent years,blockchain technology has attracted widespread attention from all walks of life and become a key direction for independent innovation and breakthroughs in Chinese new generation of information technology.With the advantages of openness,transparency,anti-tampering,and decentralization of blockchain,researchers found that blockchain has broad application prospects not only in the field of digital currency,but also in finance,auditing,medical care,notarization,copyright and even social networking.However,the current blockchain technology still has scalability problems such as low transaction efficiency,high latency,and low throughput while deploying in practical applications.The performance bottleneck of the blockchain mainly comes from the consensus mechanism and the chained data structure.Therefore,from the perspective of inchain optimization,this thesis focuses on the optimization of the consensus mechanism and the blockchain with a DAG(directed acyclic graph)structure to explore solutions that can provide the scalability of the blockchain.The main contributions are as follows:(1)Aiming at the existing consensus mechanisms are not scalable enough to provide low latency,high throughput and security while supporting large-scale networks.A new twolevel consensus mechanism called RBFT(raft cluster Byzantine fault tolerance)was proposed,which combines the high efficiency of Raft and the high security of PBFT(practical Byzantine fault tolerance).In RBFT,the nodes were firstly divided into several groups.Each group adopts the improved Raft mechanism to reach consensus and select leader.Then the leaders of each group form the network committee,and the network committee adopts PBFT mechanism for consensus.Then,the consistency,security and liveness of the RBFT consensus mechanism are discussed.Finally,the simulation and experimental results verify that RBFT has the advantages of low communication overhead,low latency,high throughput,and tolerant Byzantine fault while applying in large-scale networks.Thus,improving the scalability of blockchain effectively.(2)The Hashgraph DAG blockchain paradigm has two disadvantages.One is the low consensus efficiency caused by the communication redundancy of the random Gossip protocol.And another one is the continuously increasing of Event size as the graph grows,which results in high bandwidth requirements for device data processing.To address these problems,an improved DAG paradigm DABG(directed acyclic block graph)is proposed.Firstly,a tree-based Gossip network is constructed with federated learning,and the treebased Gossip protocol is used to replace the native random Gossip protocol,which effectively reduces communication redundancy and improves consensus efficiency.The size of Event in DABG is fixed in a block rather than increasing during transmission and then the blocks after the consensus are connected through the heaviest chain rules to form a DAG,which effectively reduces the pressure on the equipment.Finally,the sequencing of block orders and security about view split attack issues are discussed,and the simulation results show that DABG can effectively reduce the communication redundancy,thereby improving the consensus efficiency of the DAG blockchain system.