Hybrid Immune Whale Differential Evolution Optimization (HIWDEO) Based Computation Offloading in MEC for IoT

Abstract

The adoption of User Equipment (UE) is on the rise, driven by advancements in Mobile Cloud Computing (MCC), Mobile Edge Computing (MEC), the Internet of Things (IoT), and Artificial Intelligence (AI). Among these, MEC stands out as a pivotal aspect of the 5G network. A critical challenge within the realm of MEC is task offloading. This involves optimizing conflicting factors like execution time, energy usage, and computation duration. Additionally, addressing the offloading of interdependent tasks poses another significant hurdle that requires attention. The developed models are single objective, task dependency, and computationally expensive. As a result, the Immune whale differential evolution optimization algorithm is proposed to offload the dependent tasks to the MEC with three objectives: minimizing the execution delay and reducing the energy and cost of MEC resources. The standard Whale optimization is incorporated with DE with customized mutation operations and immune system to enhance the searching strategy of Whale optimization. The proposed HIWDEO secured reduced energy and overhead of UE to execute its tasks. The comparison between the developed model and other optimization approaches shows the superiority of HIWDEO.