Adaptive Optimization and Resource Allocation (AORA) Model for IoT-Edge Computing Using Hybrid Newton-Raphson and Dolphin Echolocation Algorithm (HNR-DEA) Technique

The swift expansion of Internet of Things (IoT) applications demands the implementation of real-time data processing and low-latency communication, thereby posing challenges to conventional cloud computing paradigms. This study confronts these issues by amalgamating edge computing with hybrid optimization methodologies, and introduces the AORA (Adaptive Optimization and Resource Allocation) framework. The AORA framework capitalizes on intelligent task scheduling, dynamic resource allocation, and energy-efficient approaches to augment the performance of edge computing.Our research methodology encompasses the collection and preprocessing of data derived from IoT devices, feature extraction, prediction of resource demand, and the optimization of task offloading. It employs the Hybrid Newton Raphson and Dolphin Echolocation Algorithm (HNR-DEA) methodology, which synergizes Newton Raphson optimization technique for faster convergence rate around local optima with bio-inspired Dolphin Echolocation Algorithm (DEA) for comprehensive global exploration.The Experimental outcomes reveal considerable advancements, including a 45% reduction in average latency, a 38% decrease in overall energy consumption, and a 23% enhancement in resource utilization. Furthermore, the model attained a 97% task completion rate alongside consistently elevated network throughput. Simulations were executed in CloudSim with 100 IoT devices generating dynamic workloads and edge servers equipped with multi-core processors. The experimental framework utilized Google Cluster Data to replicate workload variations, thereby ensuring scalability and robustness. These results underscore the AORA model's potential as a resilient and scalable solution for real-time IoT applications.