Research On Secure Job Scheduling Based On Fuzzy Threshold And STML In Edge Computing

With the development of the Internet of Things,the number of mobile terminal devices increases rapidly.These data generated by mobile terminal devices have characteristics of massive,real-time and heterogeneity,which poses a heavy burden on traditional cloud computing.Cloud computing is a centralized computing platform that provides service by the remote service center,which results in aggravating the network congestion,increasing the transmission delay and reducing the quality of the service.Edge computing extends the cloud computing to the edge of network,which is closer to the user,and provides low transmission delay and real-time services.Moreover,edge computing has advantages of location awareness,mobility support and wide geographical distribution,which can effectively cover the shortage of cloud computing in the background of the Internet of things.At present,there are two urgent problems in edge computing: the one is how to effectively protect user's location privacy when providing location based services.The other is that edge servers have limited resources,therefore,how to reduce system energy consumption while meeting the requirement of delay.Task scheduling affects system energy consumption and delay,which determines the service quality and user's satisfaction.Therefore,this paper chooses task scheduling in edge computing as the research topic.According to the architecture of edge computing,this paper considers the factors of user's privacy,system energy consumption and delay,and proposes a tasks scheduling scheme based on obfuscation areas and STML(Scheduling to Minimize Lateness)algorithm,aiming to protect user location privacy,reducing system energy consumption and delay.The specific research contents are as follows:(1)This paper designs an improved unchained cellular obfuscation areas generation method.Firstly,the potential threat of an existing unchained cellular obfuscation areas generation method is analyzed.Then,combined with the hash function,an improved unchained cellular obfuscation areas generation method is proposed,which increases the diversity of obfuscation areas shapes and enhances the randomness of obfuscation areas generation.The edge server is used to replace the traditional fully trusted third party(TTP)to protect the user location privacy.Finally,the improved unchained cellular obfuscation areas generation method is used to protect the user's location in the edge computing by using obfuscation areas to replace the user's real position coordinates.Experimental results show that the improved unchained cellular obfuscation areas generation method we proposed has lower cost and better protection effect than the original obfuscation areas generation method.(2)This paper designs tasks scheduling algorithms in edge computing.Firstly,the mathematical model of the system is established by using queue theory,and the energy consumption function and delay function are established.Then,a scheduling strategy to balance the energy consumption and delay is proposed.STML algorithm is used to reduce the delay,and nonlinear programming is used to reduce the energy consumption.Finally,cloud-edge servers cooperation scheduling algorithm is designed to reduce the system energy consumption when the allocated device is overloaded.And the task offloading algorithm is designed to complete tasks when their nodes leave.The experimental results show that the energy consumption is reduced by approximately 22%,while the delay is 12.5% less than the first come first served(FCFS)approach.