Research And Implementation Of Consensus Mechanism In Blockchain Electronic Health Record System

At present,the traditional electronic health record systems used by medical institutions are mostly independent of each other,and it is difficult to achieve information interaction between different systems.At the same time,the medical data in the traditional electronic health record system is stored in a local database of the medical institution,and may be hacked or stolen by internal personnel,and data security is difficult to guarantee.In addition,the medical institution holds the patient’s medical data and brings the corresponding privacy protection issues.The blockchain electronic health record system utilizes blockchain technology to fully protect the patient’s privacy and ownership of medical data while achieving secure storage and efficient sharing of medical data.The consensus mechanism is a protocol that ensures that all nodes on the blockchain are synchronized with each other and that the legality,sequence,and result of the transaction are consistently recognized and added to the blockchain.In order to avoid the medical data in the blockchain electronic health record system from being attacked by malicious nodes,the design of the consensus mechanism must first ensure security.In addition,the consensus process,as a performance bottleneck of the blockchain system,needs to improve performance in terms of throughput and latency to meet system performance requirements.This thesis aims to develop a Byzantine fault-tolerant consensus mechanism for a blockchain electronic health record system based on the Hyperledger Fabric framework.This thesis compares and analyzes the mainstream consensus mechanism in the blockchain.The common POW and POS consensus mechanisms in the public chain are unable to meet the performance requirements of the blockchain electronic health record system because of their extremely low throughput performance.The Kafka consensus mechanism performs very well in terms of throughput and is suitable for handling high frequency services,but cannot tolerate Byzantine errors,which is inconsistent with the security requirements of the system.The PBFT consensus mechanism is relatively balanced in all aspects,but there is still room for optimization in its complex communication process.Aiming at the insufficiency of PBFT consensus mechanism and system requirements,this thesis proposes and studies a consensus mechanism for the blockchain electronic health record system,Dynamic Mode Byzantine Fault Tolerance(DMBFT).The consensus mechanism mainly makes two optimizations in terms of performance: one is to optimize the single mode of the consensus mechanism to the dynamic mode for the characteristics of the system node,and to accelerate the consensus in the case of no failure;the second is to apply the BLS aggregation signature scheme to the consensus process.Reduce network traffic and improve signature verification efficiency.In terms of security,two optimizations are also made: one is to use message digest and digital signature technology to protect the integrity and non-repudiation of transaction data,and to strengthen the security protection of the consensus process;the second is to design activities based on reputation value.The node selection strategy reduces the security impact of the faulty node on the system consensus process.In order to verify the optimization effect of the DMBFT consensus mechanism,the consensus mechanism was applied to the blockchain electronic health record system,and a reasonable system architecture was designed.The core business of medical record submission,authorization,medical record transaction and so on closely related to the consensus process was carried out.achieve.Finally,the DMBFT consensus mechanism is tested by initiating core transactions in the blockchain electronic health record system.Through the comparison test with the PBFT consensus mechanism,the optimization ability of the DMBFT consensus mechanism is evaluated and compared.In the non-fault mode of DMBFT consensus mechanism,the average throughput of DMBFT consensus mechanism is 27% higher than that of PBFT consensus mechanism,which meets the performance requirements of the system.In terms of fault tolerance,the maximum number of fault nodes that the system can handle is as expected,and the fault processing process brings temporary performance degradation within an acceptable range to meet the security requirements of the system.It is verified that the DMBFT consensus mechanism can work effectively in the block chain electronic health record system,and achieves the expected results in terms of performance and security.