Resources Optimization For Mobile Edge Caching,Computing And Wireless Communications

The rapid development of Mobile Internet and Internet of Things(IoT)has brought about the proliferation of a wide range of innovative mobile applications(e.g.,augment/virtual reality,online gaming and autonomous driving).These applications are typically not only latency and energy-sensitive but also computation-intensive,thereby imposing significant stress on the cellular networks,in terms of lower backhaul load,lower application processing latency and energy consumption,etc.Recently,the concept of mobile edge computing(MEC)has been proposed to provide the necessary computing capability at the wireless edge to support these application services.However,the design of an efficient MEC system must jointly utilize the storage,computation and communications resources at the wireless edge in a more sensible way.To this end,this thesis will conduct an in-depth study on the joint optimization of these three types of resources to achieve an improvement of network performance,i.e.,relieving the backhaul load,reducing the application processing latency and energy consumption.First,in order to make better use of the storage resources at the wireless edge for reducing the backhaul load,we focus on the analysis and optimization of a random caching strategy.Specifically,we propose two base station(BS)cooperation schemes to enhance the performance of random caching.By using tools from stochastic geometric,we derive a tractable expression for the successful transmission probability(that is,the probability that a user can successfully receive the requested file)under each scheme.We consider the successful transmission probability maximization by optimizing the caching distribution to obtain an optimal random caching strategy.However,the optimization problems in both schemes are non-convex.To solve these problems,by exploring optimality properties and using optimization techniques,under each scheme,we obtain a locally optimal solution in the general case and a globally optimal solution in some special cases.The experimental results show that compared with the BS cooperation scheme based on other caching strategies,our schemes can effectively relieve the backhaul load and provide the a better service of quality for mobile users.Next,we consider applying BS random discontinuous transmission(DTX)to enhance the performance of random caching.In addition,to capture the impact of user mobility on the design of random caching,two special mobility scenarios,namely high mobility and static scenarios,are considered.Specifically,by using tools from stochastic geometry,we derive a closed-form expression for the successful transmission probability under a given transmission delay.It is shown that a larger caching probability corresponds to a higher successful transmission probability in both scenarios;random DTX can improve the successful transmission probability in the static scenario and its benefit gradually diminishes when mobility increases.In each scenario,we consider the maximization of the successful transmission probability,which is a challenging non-convex optimization problem.In the high mobility scenario,we obtain a globally optimal solution.In the static scenario,we develop a low-complexity iterative algorithm to obtain a stationary point.Numerical results show that the proposed solutions achieve significant gains over existing baseline schemes and can well adapt to the changes of the system parameters to wisely utilize storage resources and BS transmission opportunity.Then,we study the joint allocation of computation and communications resources to design an energy-efficient MEC system.Specifically,we consider an MEC system consisting of one serving node and multiple users–each with an inelastic computation task of a non-negligible computation result size.We adopt the Time Division duplexing(TDD)mode and consider a orthogonal frequency division multiple access(OFDMA)system.A joint uplink/downlink sub-channel,bit and time allocation problem is investigated to minimize the energy consumption,which happens to be a very challenging non-convex mixed integer nonlinear programming(MINLP)problem.We equivalently convert it into a convex MINLP problem by using the McCormick envelope and obtain its optimal solution via the Branch-and-Bound method.Subsequently,two suboptimal solutions are designed to achieve a tradeoff between the energy consumption and the computational complexity.Simulation results confirm the advantages of our developed solutions.Finally,we study the joint allocation of storage,computation and communications resources to design an energy-efficient MEC system.Specifically,we consider an MEC system consisting of one serving node of caching and computing capabilities and multiple users–each with computing capability and having an inelastic computation task of a non-negligible computation result size.We adopt the TDD mode and consider a time division multiple access(TDMA)system.We propose a joint caching,computation and communications mechanism which involves software fetching,caching and multicasting,as well as task input data uploading,task executing(with non-negligible time duration)and computation result downloading.We optimize the joint caching,offloading and time allocation policy to minimize the energy consumption subject to the caching and delay constraints.The problem is a challenging twotimescale non-convex MINLP problem.We convert it into an equivalent convex MINLP problem and propose two low complexity of fast suboptimal algorithms.The experimental results confirm the advantages of the proposed suboptimal schemes in the joint utilization of network edge storage,computation and communications resources by comparing with other schemes.