A hybrid optimization with ensemble learning to ensure VANET network stability based on performance analysis

High vehicle mobility, changing vehicle density and dynamic inter-vehicle spacing are all important issues in the VANET environment. As a result, a better routing protocol improves VANET overall performance by permitting frequent service availability. Therefore, an ensemble-based machine-learning technique is used to forecast VANET mobility. Effective routing based on a hybrid metaheuristic algorithm combined with Ensemble Learning yields significantly improved results. Based on information collected from the Road Side Unit (RSU) or the Base Station, a hybrid metaheuristic (Seagull optimization and Artificial Fish Swarm Optimization) method is used to estimate (BS). The suggested approach incorporates an ensemble machine learning and hybrid metaheuristic method to reduce the latency. The current model's execution is calculated using a variety of Machine Learning techniques, including SVM, Nave Bayes, ANN, and Decision Tree. As a result, the performance of machine learning algorithms may be studied and used to achieve the best results. Comparative analysis between the proposed method (HFSA-VANET) and (CRSM-VANET was done on different performance parameters like throughput, delay, drop, network lifetime, and energy consumption to assess system performance on two factors Speed and Nodes. The HFSA-VANET method shows an overall drop in the delay of 33% and a decrease in the energy consumption of 81% and an increase of 8% in the throughput as compared with the CRSM-VANET method at 80 node. The proposed method that is HFSA-VANET has been implemented in the MATLAB and NS2 environment.