Multi-user Service Migration for Mobile Edge Computing Empowered Connected and Autonomous Vehicles

Abstract

Connected and autonomous vehicles (CAVs) are promising in improving driving safety and efficiency, which are usually empowered by mobile edge computing (MEC) to push computing and storage resources to the edge networks. By deploying vehicular services at the edge servers in close proximity to vehicles, the service latency can be greatly reduced. Due to the high mobility of vehicles, the services have to be migrated to follow the vehicles to achieve a balance between the service latency and the migration cost. Making service migration decisions for each vehicle independently will suffer from the interference among the vehicles. Moreover, trajectory prediction, which is crucial for service migration decisions, becomes intractable when the number of vehicles is large. In this paper, we investigate the multi-user service migration problem in MEC empowered CAVs, and formulate the service migration of all the vehicles as an optimization problem with the aim of minimizing the average latency, where the interference among different vehicles is taken into account. We then develop an efficient multi-user service migration scheme based on Lyapunov optimization, called ING, to solve the optimization problem in an online fashion without predicting the trajectories of the vehicles. Finally, a series of simulations based on real-world mobility traces of Rome taxis are conducted to verify the superior performance of the proposed ING algorithm as compared with the state-of-the-art solutions.