Mobility-aware task offloading in MEC with task migration and result caching

Abstract

Numerous computation-intensive applications have emerged, demanding lower delay and energy consumption. Multi-access Edge Computing (MEC) provides task offloading services for mobile users (MUs) through proximate MUs, effectively minimizing service delay and energy consumption. However, the challenges of supporting MUs’ mobility persist, and ensuring uninterrupted task offloading services from MEC servers still needs to be solved. Additionally, applications like autonomous driving generate repetitive tasks, introducing unnecessary computational overhead. This paper adopts a task migration strategy to tackle these issues, specifically addressing task offloading failure due to MU mobility. Simultaneously, a computation-result cache-assisted technique is leveraged to alleviate the burden of repetitive computation. The objective is to minimize the system cost, which consists of the processing delay and energy consumption of tasks for all MUs. The problem is modeled as an integer linear programming problem, and the optimal solution is meticulously derived. In response to the high computational complexity inherent in obtaining the optimal solution, a low-complexity heuristic algorithm (CMHA) is introduced. Simulation results demonstrate that optimal and heuristic schemes significantly reduce the overall system delay and energy consumption.