Research Of Blockchain Data Communication Performance Optimization Considering Trust Degree And Weight

With the rapid expansion of internet information,data storage management tends to be distributed.In a distributed storage environment,users pay more and more attention to the efficiency of user interaction experience and the reliability of information interaction.However,user's working productivity is often restricted by the efficiency of data communication between distributed sites.An important goal of distributed data management is to improve the efficiency of data transmission and to ensure the reliability of data transmission.Blockchain technology is one of the new technologies that support the development of MIS.And it provides a distributed data management solution for storage,authentication,transfer and exchange.This dissertation focuses on solving data transmission problem of blockchain from three aspects: optimizing communication efficiency,ensuring communication reliability and improving service fairness.And it proposes different performance optimization strategies for blockchain data communication considering multiple impact factors,such as nodes' transmitting capacity,nodes' trust degree,weight,service request priority.The main contributions of the dissertation are as follows:1)In order to optimize communication performance from the communication topology structure and the communication mechanism,a multi-link concurrent communication tree model is proposed.This model based on tree structure introduces a concurrent communication mechanism and embodies the concept that every node does its best to transmit and the capable are always busy.This model can improve communication efficiency because it can maximize nodes' communication capacity.The analysis of model performance indicates that the tree topology structure has good malfunction isolation because it avoids the transmission bottleneck and redundancy retransmission.And the entire communication process is multitask concurrent transmission without intermittence or delay in order to improve concurrency and transmission efficiency.In addition,the small proportion of nodes,which are responsible for forwarding data,can effectively reduce the difficulty of node maintenance and node management on blockchain.2)In order to further optimize transmission efficiency by ensuring transmission stability and reliability,an integrated factor communication tree algorithm(or IFT for short)is proposed.This dissertation first gives node trust evaluation based on AHP time frame update strategy,which fully embodies trust characteristics(e.g.multi-attribute,multi-granularity,dynamic change with time).Through the above method to calculate the node trust degree,and based on a multi-link concurrent communication tree model,IFT algorithm is proposed.The theory proof and simulation results indicate that the communication tree constructed based on IFT algorithm has the characteristics of high reliability and low latency.3)In order to solve the impact on communication performance caused by transmission delay between nodes,this dissertation proposes a multi-link multi-factor weighted communication tree algorithm(or MMWT for short),which introduces the concept of communication weights,is based on a multi-link concurrent communication tree model and comprehensively considers three impact factors including the node communication link number,the node trust degree and the communication weight.In MMWT algorithm,nodes,which have powerful transmitting capacity,high-trust,and small-weight with other nodes,are selected to locate in the top or upper part of the communication tree by setting the trust threshold and the weight threshold in order to shorten the concurrent communication time and to improve communication reliability.Simulation results indicate that this algorithm obtains better communication performance than the existing algorithms in concurrent communication time,average end-to-end delay,concurrent degree,trust degree for the communication tree,average link stress and tree depth.4)In order to solve the problem of blockchain node's performance degradation caused by node failure on blockchain,a blockchain communication algorithm considering node failure is proposed.The basic idea of the algorithm is to set the response threshold to decide whether the node is failure,and to use precaution measures to prevent branch error in data transmission.Theory inferential reasoning shows the relationship between node's depth located in communication tree and the number of its cascading failure node,and the relationship between the number of failure nodes and the concurrent communication time.Simulation results show that when the failure nodes are less than 15%,communication efficiency is not increased significantly.In addition,failure node with low capacity has little influence on communication performance and failure node with powerful capacity results in a reduced overall communication performance.5)In order to promote fairness of node service request on blockchain under ensuring the blockchain communication efficiency,a blockchain communication algorithm considering node service priority is proposed.This algorithm is based on a multi-link concurrent communication tree model,and selects several nodes with powerful transmitting capacity to locate in the top or upper part of the communication tree by minor adjustment of node service request queue in order to improve the concurrency and shorten the transmission delay.Through simulation experiment the optimal adjust ratio range of node service request queue is between 5% and 10%.Simulation results indicate that the proposed algorithm is effective in communication efficiency and communication fairness.