Dynamic Offloading for Deadline-Aware Dependent Tasks in UAV-Empowered Multi-Access Edge Computing

View at publisher Summarise with VIRAT
Authors: J. Q. Wang, P. Y. Chen, Mengya Zhang, Quanjun Yin, Kaikai Chi, Yong Peng
Cited by: 0
References: 34

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Unmanned aerial vehicle (UAV)-empowered multi-access edge computing (MEC) networks (UEMN) have emerged as a promising solution for delivering flexible, scalable communication and computational services to mobile user equipments (UEs) in dynamic environments. However, existing research has yet to tackle the challenges of managing online-arriving, deadline-sensitive applications with complex task dependencies and unpredictable behavior in UEMN. In this paper, we propose a solution to tackle the problem of dynamic task offloading, aiming to maximize the success rate of applications while considering both deadline constraints and UAV energy consumption. We introduce a two-stage scheduling algorithm, called the Deadline-aware Scheduling scheme with Energy Management (DSEM), to formulate efficient offloading strategies in resource-constrained UAV systems. In the first stage, we prioritize applications and tasks based on the urgency and path-based ranking respectively, decoupling the temporal and data dependencies between tasks and ensuring optimal execution order. In the second stage, we optimize offloading decisions by jointly considering energy consumption and task execution time. Extensive simulation results demonstrate that the proposed DSEM algorithm significantly improves the success rate of applications compared to existing benchmarks, offering an effective solution for real-time task scheduling in UEMN.

Fields of Study & Subjects

UAV Applications and Optimization IoT and Edge/Fog Computing Satellite Communication Systems

Publication details

Journal / source
IEEE Transactions on Vehicular Technology
Volume
75
Issue
8
Pages
18192–18205
Publication date
23 March 2026
Language
EN