Computation Offloading Strategy Based on Improved Auction Model in Mobile Edge Computing Network

Abstract

In order to solve the problem of computing power or low energy of users, and to make better use of the idle resources of surrounding mobile devices, the use of D2D direct communication to perform collaborative computing with surrounding mobile devices is an important way to improve resource utilization and reduce local execution energy consumption and delay. To this end, we propose a D2D computing task offloading strategy based on the auction model. Firstly, the requesting device detects the surrounding idle mobile device resources through D2D communication technology, and constructs a communicated topology diagram of the requesting device and the service device. Then, the auction model is used to solve the computing task migration problem of multi-user and multi-service nodes in D2D. At the same time, the offloading problem is converted into an auction model and the task price model is also reasonably defined through comprehensively considering task energy consumption and time delay. The simulation results show that compared with the traditional migration strategy based on communication rate and computing power, the offloading strategy based on the auction model has achieved good results in terms of energy consumption, delay and failure rate.