Research On Task Offloading Strategy Of Mobile Edge Computing In 5G Environment

In recent years,mobile devices(MDs)have become an indispensable part of human life.The rapid growth of sensing and communication technologies has opened up possibilities for numerous applications to achieve a truly digital era.With the increasing trend of smartphone usage,the availability of mobile devices is no longer limited to wired connections.Gaming,social media interaction,business,information entertainment,and even basic practical tools widely use built-in sensors,including cameras and microphones,to provide customized services.Given the limited resources of mobile devices,processing and storing these large amounts of multimodal sensory data are not feasible.Therefore,resource-intensive mobile applications such as real-time online gaming,virtual reality(VR)and augmented reality,image processing,speech,gesture or facial recognition or other similar user-adaptive services,pose significant design challenges in achieving the required performance for resourceconstrained MDs.Mobile Edge Computing(MEC)is a new computing paradigm that has been introduced in practice in recent years.The basic principle of MEC is to bring the computing capabilities of mobile cloud computing(MCC)closer to the network edge near the mobile device.Now,applications running on mobile devices can offload computationally intensive tasks to nearby MEC servers in a low-cost way,which can significantly alleviate network congestion issues and improve the response time of running applications.Although researchers use this technology to address high computing demands for intensive tasks of MDs in Small-Cell Networks(SCNs),mobile edge computing still faces problems such as heavy load on central computing node and low network coverage density.Based on the above problems,to alleviate the pressure on the centralized computing offloading center node,a distributed computing offloading model is proposed to improve the reliability and stability of the mobile edge computing system.The specific contents are summarized as follows:(1)Task offloading is considered in the small-cell network architecture under 5G environment.In order to save energy,this paper models the energy consumption of offloading from both the task computation and communication aspects,considering the front-end link and back-end link in the process.First,an energy optimization problem is formulated to minimize the total energy consumption of all system entities while considering the constraints of computing power and service latency requirements.Then,we propose a computing offloading algorithm based on Artificial Fish Swarm Algorithm to solve the energy optimization problem.In addition,this paper proves that our scheme has good global convergence.Finally,various simulation results prove the effectiveness of our scheme.(2)A distributed computing offloading strategy based on Orthogonal Frequency Division Multiple Access(OFDMA)is studied for small-cell networks in multi-device and multiserver systems.First,in order to meet the interests of different mobile devices(MDs)and analyze the interaction between multiple small cells,we formulate a distributed cost minimization problem aimed at jointly optimizing the energy consumption and delay of each mobile device.Secondly,in order to ensure the individual demands of different mobile devices,we formulate the proposed cost minimization problem as a strategic game.Then,the strategic game is proved to be a potential game by the characteristics of potential game theory.In addition,a distributed computing offloading algorithm based on Nash Equilibrium is proposed to solve the strategic game.Finally,simulation results show the effectiveness of the proposed distributed computing offloading strategy.