Optimizing Routing Efficiency in MANETs: Integrating PSO-Optimized Neural Network Model for Improved Data Transmission

Because node mobility and network topology changes are dynamic, effective routing is still a major difficulty in the quickly developing field of Mobile Ad Hoc Networks (MANETs). This research presents a unique approach to improve routing efficiency in MANETs by utilizing Particle Swarm Optimization (PSO) in conjunction with neural network models. The core of the proposed approach involves optimizing a neural network using PSO to predict the most effective routing paths for data transmission. By integrating evolutionary optimization techniques with advanced neural network models, the system processes various network parameters-such as node distances, energy levels, and statistical metrics-to determine optimal routing paths. Comparative analysis of the PSO-optimized neural network (PSO-NN) and a traditional neural network (NN) model demonstrates that PSO-NN offers superior routing predictions, leading to improved network performance. The findings open the door for further developments in network optimization techniques and highlight how well PSO and neural networks work together to manage the complexity of dynamic network settings.