Queueing Network Model and Average Delay Analysis for Mobile Edge Computing

In the near future, mobile edge computing (MEC) services are expected with edge servers equipped in base stations whose current top priority is to provide the quality-of-service (QoS) of uplink and downlink transmission (UT & DT). The coexistence of MEC, UT & DT users may lead a big difference since MEC users compete with UT & DT users for wireless transmission resources. To investigate the impact caused by the admission of MEC users, this paper proposes a queueing network model for a base station which serves multi-class UT, DT and MEC users simultaneously. The average delay of each user class of each service type is derived. Simulations are done to validate all the analysis. Both simulation and analysis reveal that the average delay performance highly depends on the access rates of users and the resource deployment, especially for the performance of the low-priority users.