Vanet efficiency optimization viagraph convolutional layers and migrating birds optimization with non-Linear analysis

Abstract

Vehicle-to-vehicle and vehicle-to-infrastructure communication is made feasible by vehicle ad hoc networks (VANETs), fundamental components of intelligent transportation systems. Still, the highly dynamic and non-linear structure of vehicle environments makes VANET efficiency increasingly challenging. Conventional methods of optimization might not be able to sufficiently represent the complex interactions in these networks, therefore sacrificing performance. VANETs can be more effective by means of improved communication dependability, reduced latency, and bettering of resource distribution. Conventional routing algorithms and machine learning techniques among other current approaches find it challenging to handle the non-linearity and dynamic topology of VANETs, thereby generating limited scalability and flexibility. Inspired by non-linear analysis techniques, we provide a new method to address these problems integrating Graph Convolutional Layers (GCLs) with the Migrating Birds Optimization (MBO) algorithm. While MBO maximizes routing decisions by replicating the group behavior of bird flocks, GCLs effectively capture the spatial dependencies and topological variations in VANETs. By use of non-linear analysis, the system's properties are tuned, therefore enhancing the adaptability and resilience of the proposed model. Experimental studies reveal that the proposed approach much outperforms traditional approaches in key performance metrics. The average communication latency went lowered by 23% and the packet delivery ratio was boosted by 17% when compared to state-of- the-art technologies. Apart from that, the method reduced energy usage by 9% and raised general network efficiency by 12%.