Research On Trust-based Dynamic Access Control Model For Medical Big Data

The rapid development and popularization of computer science and Internet technology has brought people from the era of network 2.0 to big data.This era,based on information technology products,such as big data and cloud computing,has brought huge challenges and development opportunities to all walks of our society in the form of comprehensive coverage.As one of the most important field which related to people's livelihood,the development of medical treatment in big data is also highly valued by the whole society.Whether in national policy,theoretical technology or applied research,it is a top priority of all sectors of society today.However,the various problems that arise in the process of its application cannot be ignored,it even seriously hinders the development of medical big data.Among them,the most concerned issue is the security of such data that involves more personal privacy,which is also the foothold of my research.As one of the major methods to ensure data security,access control has always played an irreplaceable role in protecting information resources.The basic principle of it is to restrict the operation and rights of visitors to achieve the purpose of resource protection.As a type of information resource,medical big data,protected by access control method will be a way to effectively solve the problem of data security.Given that most of the security issues of data in the application process can be summarized as trust issues,this paper uses questionnaire,Delphi method and cluster analysis method to establish a targeted medical big data trust evaluation system,which based on medical big data,evaluation system and related theoretical research,and determines 14 evaluations for users indicators and 23 evaluation indicators for cloud nodes.In addition,this paper introduces fuzzy theory into the trust evaluation algorithm,learns from the classic trust evaluation model establishment process,and builds a trust evaluation algorithm model based on information entropy to fit the direct trust values and the global trust value,which lays a foundation for the next step in building an access control model.Consider that the issues such as access control rate,model complexity,applicability,and Sybil attack resistance are still the focus of research on access control models,this paper proposes a trust-based dynamic access control model based on the previous research.First,a two-way selection mechanism is proposed.Only when both parties meet the trust threshold requirements can they enter the interactive phase,which controls the access volume from the basic phase,reduces the resource occupation rate,and improves the control success rate.Then the thesis designs a dynamic control mechanism,gives the control process,and designs the trust value reduction coefficient and increase coefficient,which timely changes the node trust value and improves the timeliness of the trust value in the model.What's more,since the model continuously changes the trust value based on user behavior,it can effectively prevent Sybil attacks.Finally,five sets of simulation comparison experiments are designed to verify the effectiveness of the model.Through experimental results and analysis,it can be seen that the model proposed in this paper performs well in terms of controllability,time complexity,number of accessible resources,interaction success rate,and interaction stability.