Delay-Sensitive Task Offloading for Emergency Ambulance in Edge-MEC-Cloud Continuum using Deep Reinforcement Learning
Sonwani, BhaveshMarquering, Henk A.
Abstract
In pre-hospital emergency medicine, rapid decisions are critical. AI, 5G, and IoMT can enhance automation, but limited ambulance edge resources, network variability, and changing patient needs in real-time demand intelligent AI task offloading. In such scenarios, dynamic task offloading can optimize realtime computation by considering time sensitivity and resource availability across ambulance edge devices, MEC servers, and cloud servers. This paper presents a patient acuity-aware realtimeoffloading framework using Proximal Policy Optimization (PPO) across the edge-MEC-cloud continuum. The offloading problem is modeled as a Markov Decision Process considering patient acuity, deadlines, computation needs, battery, network, and MEC service availability. Actions select offloading locations to minimize delays and maximize task success. Validation with synthetic dataset shows the PPO model outperforms traditional algorithms in delay sensitivity and task completion under resourceconstraints.
Metrics
Topics
Related Documents
Delay-Sensitive Task Offloading for Emergency Ambulance in Edge-MEC-Cloud Continuum using Deep Reinforcement Learning
Sonwani, BhaveshMarquering, Henk A.
Reinforcement Learning for Optimizing Delay-Sensitive Task Offloading in Vehicular Edge–Cloud Computing
Ta Huu BinhDo Bao SonHiep VoBinh Minh NguyenHuỳnh Thị Thanh Bình
Delay-Sensitive Task Offloading With Edge Caching Through Martingale-Based Deep Reinforcement Learning
Chongwu DongWeidong LiZhi ZhouXu ChenZhihong TianWushao Wen
Deep Reinforcement Learning techniques for dynamic task offloading in the 5G edge-cloud continuum
Gorka NietoIdoia de la IglesiaUnai López-NovoaCristina Perfecto
HOODIE: Hybrid Computation Offloading via Distributed Deep Reinforcement Learning in Delay-Aware Cloud-Edge Continuum
Anastasios GiannopoulosIlias ParalikasSotirios SpantideasPanagiotis Trakadas