Task Offloading Based On Mobile Cloud Computing

More computing tasks are executed on mobile devices in mobile applications,which need more resource-constrained mobile devices.To achieve the high efficiency of mobile applications under limited resources,the mobile cloud computing(MCC)is proposed to reduce the energy consumption and the running time of applications.However,the quick,energy-efficient and secure offload tasks is focused on the MCC,due to the diversity of the topology of mobile applications and the variability of network channels in wireless communication environments.In addition,the enhancement of capability of MCC using the existing auxiliary devices of the communication architecture is another important issue.Firstly,to use of the computation reusing resource and Base Station's storage capability adequately,an AC management architecture based on 5G network technology is proposed.It relieves the burden of base stations and reduces energy and time consumption for mobile applications.Hierarchical management of access controllers can achieve the balance and efficiency.Meanwhile,the utilization of hardware storage resources is improved.Furthermore,the reuse of the computational tasks' results can effectively reduce the running time and energy consumption of the mobile application.To address the problem of topology diversity of mobile applications,a task offloading model for applications with sequential and general task topologies is proposed.For sequential topology applications,a dynamic programming algorithm(DPA)that can produce optimal solution is proposed to solve the time-constraint offloading problem.For general topology applications,a critical-path based heuristic algorithm(GDPA)is proposed to quickly get solution.Then,a two-stage online algorithm is proposed based on the Markov model using the network state transition matrix.The purpose is to adapt the dynamic channel changes in the actual wireless network.In the off-line stage,the time-energy status table is obtained through the positive process of dynamic planning.In the online stage,an online algorithm based on the A* algorithm is used to obtain the task offloading strategy in real time.Finally,based on the reliability hierarchy,encryption and decryption process,a fault-tolerant task security offloading model and an online algorithm for secure task-offloading are proposed to minimize energy and time consumption.Then,a Pareto optimal online algorithm based on receding horizon is proposed to improve the robustness of the algorithm.