Research On Efficient Transaction Mechanism Of Blackchain Based On Smart Contract

Blockchain is a new application model that integrates multiple technologies such as cryptography,smart contracts,consensus mechanisms,and P2 P networks.It has the characteristics of decentralization,non-tampering,etc.,and can realize peer-to-peer transactions in an environment that is not completely trusted.The transaction mechanism is the core part of the blockchain,which undertakes the task of ensuring the correctness of transactions and data consistency,as well as the task of promoting the stability and effectiveness of consensus.Currently,the most widely used are public chains and alliance chains.In the public chain,due to the majority of the proof-of-work mechanism,the transaction cost is too high and the consensus time is difficult to grasp.Moreover,long-term transaction costs are too high,which will lead to a decline in platform transaction throughput.In the alliance chain,due to the disclosure of the master node's identity and a small proportion,the transaction risk increases,leading to transaction bottlenecks.Moreover,the lack of incentive mechanism leads to unbalanced load of network nodes.Therefore,it is extremely important to study efficient blockchain transaction mechanisms based on smart contracts.This article conducts research on the efficient transaction mechanism of blockchain.First,it analyzes the problems existing in the Ethereum transaction mechanism,and proposes a plan for optimizing the transaction cost of Ethereum.The purpose is to reduce transaction costs and accurately grasp the consensus time.Then studied the Hyperledger Fabric transaction mechanism,and proposed a scheme to improve the Fabric transaction mechanism by using a credit model and a verifiable random function.The purpose is to improve the security and transaction efficiency of the Fabric network,and to improve network load balance.The main content and results of the paper are as follows:(1)In response to the difficulty of balancing transaction costs and consensus time in Ethereum,this paper analyzes the factors affecting transaction costs in the Ethereum transaction mechanism,and proposes a genetic algorithm and an improved e Xtreme Gradient Boosting(XGBoost)algorithm Combined Ethereum smart pricing mechanism.This mechanism uses the XGBoost algorithm to model and analyze the Ethereum transaction data,and combines the genetic algorithm to optimize the XGBoost model.Therefore,the improved XGBoost model can be used to predict the GAS price of the next block of Ethereum.The thesis evaluates the model in terms of transaction cost,transaction success,error value and so on through experiments.(2)Aiming at the vulnerability of endorsement nodes in Hyperledger Fabric,this paper analyzes the reasons why endorsement nodes are vulnerable,and proposes a Hyperledger Fabric transaction mechanism optimization scheme based on an improved credit model.The credit value is introduced to evaluate the nodes,the candidate sets of endorsing nodes are selected by node levels,and the endorsing nodes are selected by using a verifiable random function.Experimental results show that compared with the original transaction mechanism,the proposed scheme has significant improvements in security.(3)Aiming at the problems of low node activity and transaction efficiency in the Fabric network,a credit evaluation mechanism is proposed.Stimulate the network activity of nodes and reduce the load of other nodes,which is conducive to the load balance of the Fabric network.And use the credit value to improve the judgment strategy of the endorsement result,and further improve the transaction efficiency.Through experimental demonstration,the proposed scheme has better performance in terms of transaction efficiency and communication cost.