Optimized Resource Allocation for IoT Networks Using Genetic Particle Swarm Optimization and Enhanced K-Nearest Neighbor Algorithms

The proliferation of Internet of Things (IoT) devices has surged, with projections estimating their numbers to reach fifty billion.This growth underscores the imperative need for technologies that prioritize minimal power consumption and computational costs.However, the heightened latency inherent in cloud-based services presents significant challenges.This paper proposes a novel resource allocation strategy employing Genetic Particle Swarm Optimization (GPSO) and an optimized K-Nearest Neighbor (OKNN) algorithm to address these challenges.The GPSO algorithm models each network node as a particle, drawing inspiration from natural behaviors observed in birds.Through iterative updates, it optimizes the position and velocity of these particles, enhancing resource allocation efficiency.Additionally, the OKNN algorithm facilitates data classification into confidential and non-confidential categories, refining data processing mechanisms further.We conduct a comprehensive analysis, evaluating minimum, maximum, and average response times alongside data center processing times.Comparative assessments of CPU usage and energy consumption are performed against established algorithms like round-robin and genetic algorithms.Our findings reveal significant enhancements in both efficiency and performance across various IoT applications.This research advances IoT device management and operation, with implications for improving overall data processing mechanisms.The proposed resource allocation strategy offers a promising avenue for addressing the unique challenges posed by the evolving landscape of IoT technologies.