Hybrid Feed Forward Neural Networks and Particle Swarm Optimization for Intelligent Self-Organization in the Industrial Communication Networks

Abstract

In recent times, the Industrial Communication Networks (ICNets) have been playing a vital role in advancing mobile generation networks, especially in the evolution of 6G networks. This research proposes a novel technique for self-organization that integrates Feed Forward Neural Network (FFNN) and Particle Swarm Optimization (PSO) to enhance the network management, optimization and adaptive learning in 6G. The traditional self-organization models in ICNets and 6G depends on rule-based heuristic, reinforcement learning and classical optimization techniques, which often struggle with high computational complexity, slow convergence rates, and suboptimal decision making. In contrast, FFNN+PSO fusion model leverages the predictive learning capability of FFNN and the global optimization strength of PSO to ensure intelligent self-optimization, real-time adaptability, and ultra-low-latency in the dynamically changing 6G environments. The experimental results demonstrate that the proposed method achieves a significantly higher accuracy of 98.25% by outperforming the existing models such as Random Forest (80%), Reinforcement learning (90%), Max Overlapping (88%), and Ant Colony Optimization (92%), Further, the proposed method enhances the energy efficiency, complex network function approximation, and collaborative optimization which make it an ideal choice for scalable and self-organization model in the 6G and ICNets. This study provides a transformative contribution to self-organization in the 6G networks and it offers robust, high-performance alternative to the conventional models as well it ensures massive device connectivity with intelligent network adaptation.