Combined-Jelly-Snake optimization with Deep Learning Architecture for Task offloading in Edge Computing

Edge computing allows devices with limited processing power to transfer their computational tasks to nearby edge servers for efficient processing. However, effectively managing offloading decisions to meet latency requirements and optimizing resource allocation during peak loads while ensuring consistent service quality remains a significant challenge in edge computing. To tackle this issue, a CNN-based task offloading method is utilized. The approach incorporates essential components such as a wireless body area network (WBAN), edge servers (ES), and a medical center integrated into edge computing for the offloading structure. Various sensors are distributed across the patient’s body to form a patient-centered WBAN, which records medical monitoring data and transfers it to the appropriate gateway user node (GUN) responsible for task computation decisions. The GUN handles task-related data scaling, which serves as input for the CNN. The CNN produces a result ranging from zero to one, where "0" indicates local task execution and "1" indicates task offloading to an edge server (ES). An optimal resource allocation based on a hybrid algorithm is proposed to effectively manage workload demands while maintaining efficient performance.