Research On Privacy-preserving Task Offloading Mechanism In Mobile Edge Computing

Along with the tremendous progress of mobile communication technology,mobile edge computing has been increasingly studied,which provides technical support for the application of ultra-low latency and ultra-large computing requirements in the future.Especially with the development of task offloading technology in mobile edge,i.e.,the device resource is limited and the edge server has enough resources,the task offloading decision and resource allocation are optimized.As well as the development of mobile edge computing privacy protection technology,it can provide a guarantee for future users to experience a more colorful network digital world.However,there are still many defects in the current research of mobile edge computing technology.Some existing task offloading algorithms will consume a lot of computing resources to allocate communication,storage and computing resources in complex mobile edge computing scenarios,which hinders the reduction of device latency in complex scenarios.And most of the current work does not consider the protection of user privacy.In order to solve these problems,this paper mainly does the following works:(1)Load balancing for energy-harvesting mobile edge computing.In the mobile edge computing scenario with energy harvesting devices,in order to achieve load balancing and reduce energy consumption delay on the task queue and task return queue between the server and the user.Firstly,the queue model,energy harvesting model and task processing model are optimized.Then,The load balancing problem is standardized as an optimization problem,minimizing energy consumption and queue redundancy.After that,the Lyapunov optimization algorithmic program is employed to solve this optimization problem,higher computing service capability of the MEC system is obtained.(2)Deep reinforcement learning-based privacy protection task offloading mechanism in multi-user MEC system.In view of this problem,current offloading algorithms consumes a lot of resources for task and resource allocation in complex mobile edge computing scenarios.First,this paper models computing model and privacy protection model.Computing model based on the allocation of system time and the privacy protection model is based on the task allocation scheme.Afterwards,system computing speed enhancement with user privacy protection,jointly solved as a joint optimization problem.Then,in this paper,the non-convex optimization problem is transformed into a convex optimization problem skillfully and can be solved by deep reinforcement learning method.Finally,the task unloading mechanism can improve the computing speed of the system and protect the user's location privacy and usage mode privacy.