Service-dependent task offloading for multiuser mobile edge computing system

摘要

Mobile edge computing (MEC) has been deemed as one of the key technologies for pushing powerful computing ability to the radio access network. Different scenarios require different services, and tasks usually require computing in a specific runtime environment. However, few studies have taken this into consideration. To this end, the authors propose a novel model of service-dependent task offloading for the multiuser MEC system with resource constraints. Through determining which services should be deployed at the network edge and how many tasks should be offloaded, a revenue maximisation problem is constructed. Accordingly, they propose an iterative optimisation algorithm with high performance for the service deploying and task offloading problem. Finally, numerical simulations reveal that their proposed algorithm outperforms other schemes.