PROJECT TITLE :

An Evolutionary Algorithm-Based Vehicular Clustering Technique for VANETs

ABSTRACT:

Road accidents are responsible for the loss of a significant number of valuable lives all over the world. Vehicular ad hoc networks are the ultimate solution to this problem, which can be used to combat it (VANETs). It is of the utmost importance for there to be effective communication between the various vehicular nodes in a VANET because of the high mobility of the vehicles and the variable topology of the networks. Clustering is a well-known technique that is utilized in VANETs with the goal of improving the effectiveness of communication. As a result, a clustering algorithm that is based on Moth-Flame Optimization (MFO) and is given the name AMONET is being developed with the intention of being able to successfully function in the high mobility nodes scenario of VANETs. AMONET is built on a procedure that is bio-inspired, and it optimizes clusters so that communication can take place in a dependable and effective manner. Experimentally, our algorithm is evaluated with well-known algorithms like Ant Colony Optimization (ACO), Comprehensive Learning Particle Swarm Optimization (CLPSO), and Multi Objective Particle Swarm Optimization (MOPSO) (MOPSO). In order to determine which of these processes is the most effective overall, a number of experiments are carried out. The average cumulative results across all grid sizes come in at 27.1% for the AMONET grid, 36.3% for the ACO grid, 54.9% for the CLPSO grid, and 58.7% for the MOPSO grid. According to the findings, AMONET achieves results that are extremely close to perfect, completely covers the network, and produces the fewest possible clusters. It is an effective method to achieve vehicular clustering, which serves the dual purpose of enhancing the overall performance of the network and, as a direct result of this improvement, lowering the amount of money spent on the network's routing.