Smart placement of RSU for vehicular networks using multiobjective evolutionary algorithms

This article describes the application of a computational intelligence technique for locating roadside infrastructure for vehicular networks over realistic urban areas. A multi-objective formulation of the problem is introduced, considering the deployment cost, and a new model for quality-of-service is defined, accounting for both the traffic (number of vehicles and speed) and the coverage of street segments in the city. The related optimization problem is solved using a multiobjective evolutionary algorithm. The experimental analysis considers real scenarios in the city of Malaga, using realistic traffic information and real antennas. The reported results indicate that the proposed multiobjective evolutionary algorithm is able to compute accurate trade-off solutions for the problem, significantly outperforming a greedy approach based on the ideas previously applied in the related literature.