Context Driven MEC Resource Allocation for Time-Critical AR Applications

Mobile Edge Computing (MEC) is a promising solution for delay-sensitive and computationally intensive applications mitigating the increased latency that cloud computing endures. This is achieved by delivering the computation resources in close proximity to the users' Mobile Devices (MDs) at the edge of the Radio Access Network (RAN). This network paradigm, has attracted the attention of researchers and several solutions for bandwidth and computation resources allocation have been proposed in order to achieve low latency and energy efficient models for offloading applications' tasks. In this paper we present an offloading optimization strategy, exploiting the visual information on each MD for allocating MEC resources. The significance of the several parameters in each video frame context is evaluated along with computation delay in a proper objective function. A game theory optimization algorithm is used for the objective function minimization leading to the optimal trade-off between latency and reliability.