Dynamic load balancing of traffic in the IoT edge computing environment using a clustering approach based on deep learning and genetic algorithms

Internet of Things (IoT) networks generate significant traffic, requiring careful monitoring due to the large number of connected devices and their continuous data communication. Edge servers can provide effective monitoring. However, effectively managing this traffic represents a major challenge due to the diversity of devices, unpredictable data fluctuations, and uneven server utilization. This paper proposes an innovative method to optimize load balancing across servers to ensure uniform traffic monitoring. We divide the traffic load by intelligently grouping machines so that servers have to monitor the same amount of traffic. Our approach uses a deep learning technique to anticipate future traffic variations and a genetic algorithm to intelligently distribute the load between servers according to the predicted variations. Simulation results demonstrate the effectiveness of our approach to adaptive traffic management in IoT networks.