Enhancing urban vehicular communication and safety through HMM-OCR

Abstract Vehicular Ad Hoc Networks (VANETs) work in urban environments where the topology changes due to high mobility, limited communication and dense traffic conditions. These factors lead to increase in end-to-end delay, energy consumption, and significant packet loss such challenges highlight the need for robust and adaptive routing mechanisms that can maintain reliable communication under dynamic and dense traffic scenarios. To overcome these issues, this study proposes a Hybrid Meta-Heuristic and Machine Learning-based Optimised Cluster-Based Routing (HMM-OCR) aimed at enhancing communication reliability and routing efficiency in urban VANETs. The proposed method integrates Modified Golden Eagle Optimisation (MGEO) for energy efficient clustering and Improved Jackal Optimisation (IJO) for optimal cluster head selection. Additionally, a Multivariable Output Neural Network (MONN) is employed to ensure efficient data forwarding and path establishment. Simulation results obtained using NS2 shows that HMM-OCR outperforms across key performance metrics. Specifically, HMM-OCR enhances throughput by 4.83–8.55%, reduces packet drop by 4.53–22.24%, improves packet delivery by 3.78–21.9%, reduces delay by 0.56–2.24 s, energy consumption by 20.38–62.43%, and routing overhead by 5.87–22.87%. These results clearly demonstrate that HMM-OCR method which significantly, enhances communication efficiency and reliability in urban environments, making it suitable for intelligent transportation systems.