Infrastructure Deployment in Vehicular Communication Networks Using a Parallel Multiobjective Evolutionary Algorithm

TL;DR

The paper tackles the RSU deployment problem in VANETs by formulating a multiobjective optimization to maximize QoS and minimize deployment cost, solved with a parallel NSGA-II MOEA. It introduces a problem-specific encoding, operators, and a master-slave parallel model, and validates the approach on realistic Málaga data across multiple traffic patterns and applications. Compared to two baselines (Constructive PageRank and Randomized Knapsack), NSGA-II yields accurate Pareto fronts and substantial improvements in both QoS and cost, demonstrating the practicality of MOEAs for smart-city infrastructure planning. The work emphasizes real-world applicability, offering insights for scalable city deployments and directions for future enhancements such as other urban areas and dynamic event modeling.

Abstract

This article describes the application of a multiobjective evolutionary algorithm for locating roadside infrastructure for vehicular communication networks over realistic urban areas. A multiobjective formulation of the problem is introduced, considering quality-of-service and cost objectives. The experimental analysis is performed over a real map of Málaga, using real traffic information and antennas, and scenarios that model different combinations of traffic patterns and applications (text/audio/video) in the communications. The proposed multiobjective evolutionary algorithm computes accurate trade-off solutions, significantly improving over state-of-the-art algorithms previously applied to the problem.

Figures (7)

• Figure 1: Global VANET architecture.
• Figure 2: Encoding for RSU-DP solutions.
• Figure 3: Variations applied by the mutation operator.
• Figure 4: Calculation of the vehicles attended by a RSU.
• Figure 5: Segments defined over the real map of Málaga.