Energy-Efficient Computation Offloading in Edge Computing for Multi-UAV Networks
Kimia Ghasemi, Zeinab Movahedi · 2024
With the pervasive advancement of communication networks, the utilization of IoT permeates various spheres of societal life. To address the inherent limitations of IoT devices concerning computational power and energy, coupled with the imperative of minimizing access delays, computing offloading to edge computing emerges as a pragmatic solution. Moreover, propelled by the advent of 5G networks, drones have gained significant traction in augmenting network coverage across diverse environments. This paper delves into the intricacies of computation offloading and UAV path planning within an edge computing framework, particularly focusing on multi-UAV scenarios. We formulate the problem with the overarching objective of maximizing the energy efficiency across both UAV and IoT device layers, while accounting for the costs associated with locally executed tasks alongside offloaded ones. Given the formidable computational complexity entailed, we propose a solution algorithm predicated on the functional coverage set paradigm, aimed at enhancing the system's energy efficiency holistically. Evaluation results corroborate the efficacy of the proposed algorithm, demonstrating marked enhancements in energy efficiency and consumption by 128/% and 25%, respectively, compared to prior endeavors.