Research On Reputation Mechanism And Incentive Mechanism Based On Blockchain For Mobile Crowdsensing VANET

Since entering the 21st century,with the deepening of the Internet of Things,the VANET system has continued to develop,so that the scenario of Mobile Crowdsensing VANET has been widely used.As the current popular sensing scenario,the vehicle,as a sensing user,relies on the sensing device to interact with the surrounding people,cars,roads and the environment to complete large-scale and complex tasks.However,due to the highly dynamic and freely moving vehicles and complex topological changes,there are inevitably two major challenges,one,malicious users will upload false or low-quality data to threat system,and the other,the system lacks enough attraction to attract users to participate.Therefore,how to deal with the challenge effectively is the key problem in this scenario,which is of great significance to ensure the safe communication and enthusiasm for participation.Effective detection of malicious users and motivating their active participation are the basis for maintaining the security and reliability of the scenario and are crucial to its performance.Traditional research focuses on cryptography and third-party platforms to detect malicious users,so its applicability is poor due to its complex overhead,insufficient consideration of historical reputation value and unfairness of detection.In view of the enthusiasm of users to participate,the traditional research focuses on economic theory,gamification and other methods,which will inevitably cause the inefficiency of incentive and lack of consideration of data quality.Therefore,in view of the lack of existing research,in order to ensure the effectiveness of malicious user detection,encourage the fairness and effectiveness of user participation and system security,using blockchain to construct a reputation mechanism for malicious user detection,design an incentive mechanism for user participation enthusiasm.The main research work is as follows:First of all,in view of the current problems of malicious user detection in the Mobile Crowdsensing VANET,such as high overhead,poor detection fairness,centralized system vulnerable to single point attack and so on,in order to meet the requirements of system security and efficient detection,this paper constructs the blockchain into three parts,and designs a reputation mechanism for malicious user detection composed of three stages of data upload,data quality evaluation and reputation update.First,in order to reduce the data upload overhead,data aggregation and machine learning technology are combined to complete the data upload,and smart contracts are designed to verify the reliability of election.Then,the data quality evaluation algorithm is composed of four indicators: the history of the participating vehicle,the unbiased data,the leadership of the participating vehicle and the spatial force of the participating vehicle.then,logistic is introduced,and the model effectively describes the changes in reputation values.Finally,while realizing effective malicious user detection,the goal of low overhead,high performance and robustness against malicious users is achieved.Secondly,in view of the current problems of incentive schemes in the Mobile Crowdsensing VANET,such as poor availability,high cost,poor accuracy and ignoring the data quality itself,in order to meet the requirements of system security and data traceability and unforgeability,this paper designs an incentive mechanism composed of two stages of pre-task sensing and post-task sensing based on blockchain architecture.First,four influence factors are defined in terms of space,time,number of participating vehicles,and participating vehicle reputation value to obtain pre-task incentives.Then,fully combining the data quality evaluation algorithm,the reward and punishment cost function and resource compensation value are defined to obtain posttask incentives,and the exchange of incentive value and points is realized through the standard procedure of blockchain mounting.Finally,while effectively motivating users,the system achieves low overhead,high task completion rate and high-quality rate.Finally,this paper takes air quality detection as the application scenario of Mobile Crowdsensing VANET.Through security analysis and performance evaluation,and compared with different schemes,the effectiveness of the proposed method is shown.The reputation mechanism scheme for malicious user detection not only ensures the security of the system,but also reduces the time overhead,and cannot tolerate users with more than 50% malice.The incentive mechanism scheme for user participation enthusiasm can encourage users to achieve nearly 100% of the tasks in a short time,and the high-quality rate is increased by 14%,which effectively ensures that the Mobile Crowdsensing system can sense data with high-quality.