Research On Reputation Incentive Scheme For Consensus Protocols Of Blockchain

Blockchain technology has become a new method to solve multi-party cooperative trust problems.Through consensus mechanisms,users can achieve consensus to accomplish a collaborative task without central control.In this thesis,the aim is to solve the technical problems often encountered in the process of reputation-based nodes participating in consensus:(1)How to effectively and reasonably establish fairness of a system,since the establishment of accounts does not require cost,nodes can create multiple accounts to participate in consensus,each account maintains a credibility,which can improve the probability of putting forward a block without increasing costs.This alters the fairness of the system.(2)How to effectively avoid users from abandoning old accounts to new accounts in order to evade the system penalty,when the credit value of an account is too low(such as down to a negative number),users have sufficient reasons to abandon the account before being punished without paying any penalty,as long as all the funds in the account are transferred to other accounts before abandoning,they can re-join with a new identity.(3)How to effectively avoid the latent problem of reputation,malicious users may potentially establish their own reputation and then implement a deception to withdraw it.This has happened before on exchanges and Bitcoin lending platforms,so the reputation mechanism must be carefully designed,and the economic costs of Byzantine Fault behavior analyzed.By designing a de-centralized consensus protocol and combining it with a reputation incentive mechanism,an identity scheme for consensus protocols is proposed: deploying an identity registration contract,the logic of the contract is to establish a global registry to store the identity declarations of the nodes participating in consensus.The node invocation contract must be registered successfully before it can participate in the consensus as a verifier.At the same time,the contract also maintains the credibility of the verifier.The main research content is a credit incentive scheme,the goal is to encourage members' cooperative behavior through a credit incentive mechanism,while limiting the participation of malicious members.Each new member entering the system will get a unique ID and initial reputation value,which accumulates with the consensus behavior of this member.Cooperative behavior can gain positive credit value,negative behavior gains vice versa.The return of positive reputation can encourage members to participate in consensus action to a higher degree,while negative reputation can reduce the weight of negative members in the consensus finding process.In a complex distributed environment,the establishment of a reputation system can improve the token incentive system,make it change from a one-time game to a multiple-round game,at the same time,it can resist stronger attacks,and give some users with lower computing power more opportunities to participate in the consensus.The main advantage of this scheme is that it is compatible with the most common PoX consensus protocol.This enables it to be integrated into existing networks and protocols.At the same time,in order to reflect the applicability of the reputation module in the Internet of Things,just to mention a few other possible applications,the current mainstream blockchain project,Ethereum,is used to develop and test the reputation module.Multiple miner nodes are run on a single machine,each node independently operates the Ethereum protocol to mine,and an infinite cycle of transaction generation and submission functions is run to the miner's local transaction pool.In order to facilitate the experiment to test some basic data,the miner registration contract is not used here,but the miner's address and reputation value in the Genesis Block are manually set.This block maintains this static list of miner's addresses in the Ethereum agreement.