Optimization Algorithm for Data Transmission in the Vehicular Networking Based on Federated Edge Learning

In the context of Vehicle-to-Everything (V2X) communications, optimizing data transmission is crucial. This study uses evolutionary game theory to build a profit model for channel data transmission, focusing on data security requirements and privacy encryption costs. It aims to maximize overall benefits while ensuring data security and accuracy. A privacy-preserving federated learning algorithm is integrated, enabling effective information amalgamation across multiple data nodes while protecting privacy. The research also applies a profit function from federated edge learning and replicator dynamics to simulate competitive strategies, finding equilibrium in strategy selection. Simulation results demonstrate the strategy's effectiveness in V2X environments, offering theoretical and practical insights for enhancing network data transmission security and efficiency.