Research On Communication And Computing Resources Fusion Strategy Based On Information Timeliness

With the deep integration of the mobile Internet and the Internet of Things,a large number of new real-time application scenarios such as autonomous driving,industrial control,and smart home are emerging.In those scenarios,edge devices can sense the surrounding environment,generate tasks for processing,thus obtaining control decisions.However,the obtained information can only guide the behavior of devices for a limited period.Thus,freshness of information is critical to behavioral decision.Recently,age of information(AoI)is proposed as a metric to measure the freshness of information.The age of a piece of information is defined as the time elapsed since the last received packet is generated at the source.However,the limited computing power of edge devices restricts the real-time processing of tasks.Mobile edge computing can reduce the processing time of task by offloading compute-intensive tasks from edge devices to edge servers.Then the freshness of tasks can be improved.Therefore,it is worthy to research the joint scheduling strategy of communication and computing resources and make full use of ubiquitous resources to achieve high-efficiency task processing considering the dynamics of communication and computing resources.Aiming at the need for fresh information in new real-time applications,in this thesis we study the communication and computing resources allocation strategies in mobile edge computing systems,in order to guarantee the freshness of task results.First,for multi-application scenarios,considering the dynamic of communication resources,an age-based task scheduling and offloading strategy is proposed.Furthermore,based on the dual dynamics of communication and computing resources,an optimal task generation strategy is studied.The main contributions of this article are as follows:(1)Considering the critical importance of the freshness of information to the performance of systems and the dynamic of communication resources,a task scheduling and offloading strategy was designed for optimizing the information freshness of task results under the constraints of energy consumption.Simulation results show that the age-based strategy proposed in this article is superior to the delay-based strategy in guarantying the freshness of task results.Meanwhile,it is also very competitive in minimizing the delay of tasks processing.(2)Further more,considering the dual dynamics of communication and computing resources and the difference of time-sensitive for different applications,based on the concept of age of information,an evaluating freshness function for different applications is proposed.The communication and computing processes was abstracted into M/M/1:M/M/1 queue model.The closed-form solution of task-generate frequency with optimal freshness was derived based on queuing theory and probability theory.