Research On Trust Evaluation For Cloud Service Providers With Privacy Preserving

With the development of distributed intelligent computing and the maturity of cloud computing technology,cloud services have developed rapidly.However,compared with the traditional service model,the cloud service environment is highly dynamic and complex,and users will face more security and privacy issues.In addition,the security incidents of famous cloud service providers have caused a serious crisis of trust between users and cloud service providers.This thesis focuses on the topic of how to build trust between users and cloud service providers.First,we conducted a survey on the research status of trust evaluation for cloud service providers,and then in view of the shortcomings in the existing framework,this thesis proposes two effective solutions.(1)This thesis proposes a trust evaluation framework with privacy protection for cloud service providers,which can against malicious feedback.The framework improves the reliability and integrity of the trust evaluation results by combining objective trust evaluation and subjective trust evaluation.In the subjective trust evaluation process,this framework calculates the Euclidean distance between users' feedback data and monitored QoS during cloud service operation to detect users' malicious feedback,which greatly improves the accuracy of users' malicious feedback detection.In addition,this framework uses differential privacy technology to add appropriate and controllable noise to the user's feedback evaluation data,which protects the privacy of users.Experimental results and case studies show that the proposed framework is effective.(2)This thesis proposes a trust evaluation framework for cloud service providers based on blockchain.Existing trust evaluation frameworks use a third party to collect and process users' privacy data.However,in actual application environments,the third parties are often untrusted,which will bring a series of security issues.Our framework fundamentally removes third party by introducing a blockchain network,and the framework ensures that the users' anonymity and malicious users' traceability through group signature technology.In addition,compared with the existing trust evaluation framework with privacy protection,the evidence data of this framework is in the form of plaintext,so the efficiency of trust evaluation is higher.Finally,the safety analysis and performance analysis proved the effectiveness of the framework.