Algorithms For D2D Task Allocation In Mobile Edge Computing

With the rapid development of wireless communication technology and mobile internet,the number of mobile applications is continuing to increase,more and more computationally intensive tasks are being performed on mobile devices,such as augmented reality,face recognition,and interactive games.Usually performing these computationally intensive tasks requires a large amount of computing resources and high energy consumption.Existing terminal devices are difficult to meet the requirements of low latency and high reliability mobile application due to limited computing power and battery capacity.In order to overcome the limitations of resource-constrained mobile devices and improve the efficiency of mobile application,computing offloading is a possible solution.Mobile Edge Computing is designed to provide cloud computing capabilities at the edge of the mobile network.Users can offload some or all of their tasks from end devices to edge cloud computing,reducing application runtime and end device power consumption.The use of D2D communication technology on the basis of traditional cellular systems can generate a variety of performance gains.Firstly,the physical distance between end users based on D2D communication is short,and the communication link quality is high,so low delay,low power and high data transmission rate can be provided;secondly,compared with the uplink and downlink communication modes adopted by the traditional cellular network,the communication mode through the D2D link is more convenient and quick,and the resources are saved;finally,the user can also reuse the cellular network resources by underlay D2D communication.The cellular user spectrum resources enable the D2D communication link to coexist with the cellular network,thereby further improving the utilization of the spectrum resources of the entire system.In addition,D2D communication technology in mobile edge computing can utilize both traditional cellular links and direct D2D links,while increasing the capacity of the network and reducing the computational complexity of the base station.In order to minimize the energy consumption of mobile devices,DVFS technology is also used to reduce the clock frequency and working voltage of the terminal devices at the same time,thereby reducing the total energy consumption of the system.In this paper,the optimization of the total cost of the application,namely energy consumption and time,is carried out.A mobile edge computing model based on D2D communication is proposed.Two fast approximation algorithms are proposed: 1)the fast algorithm of greedy strategy(HGA),2)heuristic particle swarm(HPSO)algorithm based on greedy strategy to further optimize the solution of HGA.The experimental results show that compared with the traditional strategy of performing all tasks on only one device and uploading the cloud as much as possible,the total cost of the HGA algorithm proposed in this paper is optimized by 28.5% and 9.1% respectively.Compared with the HGA algorithm,the total cost of HPSO is reduced by 12.3%.that is,the proposed algorithm can effectively reduce the total cost of the system and meet the needs of end users.