Predictive Resource Allocation for Mobility-Aware Task Offloading and Migration in Edge Environments

Mobility-enabled devices play a crucial role in the Industrial Internet of Things (IIoT). However, their performance in executing computationally heavy tasks is often hindered by limited battery capacity and processing power. Offloading computation to edge servers has emerged as a promising solution. However, conventional methods often struggle to adapt to the dynamic mobility patterns and fluctuating resource availability characteristic of mobile edge computing (MEC) environments. To address these limitations, we propose an enhanced MCOTM framework, which replaces the Lagrange interpolation method with Newton’s interpolation technique for more accurate trajectory prediction. Newton’s method offers improved numerical stability and computational efficiency, making it more suitable for real-time IIoT scenarios. Experimental evaluations demonstrate that our Newton-enhanced MCOTM approach achieves notable improvements: a 7.76% reduction in the mean task turnaround time, a 6.96% decrease in energy consumption, and a 4.55% reduction in migration rate. These results underscore the effectiveness of Newton interpolation in enhancing system efficiency and adaptability in dynamic IIoT environments.