Modeling and Performance Analysis of Stochastic Mobile Edge Computing Wireless Networks

Mobile edge computing (MEC) is an emerging architecture to enable variety of innovative applications and services with ultra low latency at the resource-limited mobile devices. In this paper, we investigate how the communication resources and the computing resources, including mobile users and MEC servers, interact with each other in multi-cell MEC-enabled stochastic wireless networks. To this end, the MEC-enabled network model including mobile users with limited storage capacity and computing capabilities is considered, which is characterized in random node distribution, dynamic traffic, orthogonal frequency division multiple access and task retransmission mechanism. Based on the model, the two-dimensional discrete Markov chain is employed to characterize the task execution process. We derive the stationary distribution of the buffer length and outage probability by combining the queuing theory and stochastic geometry, based on which the radio access network throughput is calculated to measure the network performance. Extensive simulations have been conducted to verify the effectiveness of the proposed offloading strategy and to provide valuable insight.