Abstract
Vehicular Ad Hoc Networks (VANETs) are characterized by high mobility and the absence of centralized infrastructure, posing significant challenges in maintaining stable and reliable communication. One of the key issues in VANETs is the frequent link breakage due to the dynamic topology, which affects routing stability and network performance. To address this challenge, a Hybrid Optimized VANET Routing Protocol Using RNN-LPC is proposed, leveraging Recurrent Neural Networks (RNNs) and Linear Predictive Coding (LPC) to predict the future distance between a node and its neighbors. By incorporating predictive distance estimation, the proposed protocol selects more stable nodes, thereby enhancing route longevity and communication reliability. The protocol is evaluated against a traditional VANET routing protocol using key performance metrics such as packet delivery ratio, routing overhead and end-to-end delay. The results demonstrate that the proposed approach reduces routing overhead while improving packet delivery ratio, making it a more efficient and scalable solution for highly dynamic VANET environments.
DOI: 10.61416/ceai.v27i2.9529