Energy-aware Task Scheduling and Offloading in SDN-Enabled IoT Networks Using Fuzzy Clustering Neural Model and Secretary Bird Optimization

Abstract

The rapid expansion of Internet of Things (IoT) networks necessitates efficient task scheduling and offloading mechanisms to improve energy efficiency, reduce latency, and optimize resource utilization. However, conventional scheduling approaches often suffer from imbalanced workload distribution, high energy consumption, and increased processing delays, resulting in suboptimal system performance. To overcome these issues, this study proposes a Double Fuzzy Clustering-Driven Context Neural Network (DFC-CNN) integrated with the Secretary Bird Optimization Algorithm (SBOA) for energy-aware task scheduling and offloading in Software-Defined Networking (SDN)-enabled IoT environments. The DFC-CNN model dynamically clusters IoT tasks based on contextual attributes, enabling adaptive scheduling, efficient load balancing, and real-time task prioritization. Simultaneously, SBOA optimizes task scheduling and offloading decisions, improving fog and cloud resource utilization while reducing energy consumption and execution delay. Extensive experimental evaluations conducted on benchmark datasets demonstrate that the proposed framework achieves up to 35% lower energy consumption, 28% shorter task completion time, and 40% higher system throughput compared with state-of-the-art methods, including PSO, FA, SSA, HHO, MOMFO, and ABC. By integrating context-aware task clustering with nature-inspired optimization, the proposed framework enhances scalability, improves resource utilization, and supports sustainable and energy-efficient computing in large-scale IoT environments.