Optimizing Roadside Unit Deployment in VANETs: A Study on Consideration of Failure
Bingjie Liang, Fujun Wang, Bin Ran
Abstract
With the rapid development of Internet of Things and communication technologies, the connected vehicle technology with vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication is seen as the most promising solution for reducing traffic accidents and alleviating traffic congestion. Roadside Units (RSUs) play a crucial role in enabling V2I communication, as they can gather and broadcast traffic event information in the road network. Implementing a well-thought-out RSU strategy can significantly improve the stability and efficiency of traffic event information transmission in the road network. However, there is a lack of effective strategies for RSU deployment in stochastic scenarios. The presence of unpredictable factors like equipment failures and network attacks may disrupt the normal functioning of the RSUs. To guide the deployment of roadside facilities, we analyze the transmission time of traffic events in the road network and construct a multi-objective RSU deployment model considering the possibility of RSU failures. This model aims to maximize the number of connected vehicles served and minimize the expected transmission time of events. Through fuzzy programming theory and different decision preferences, we transform the two objectives into a single objective function. A random-key genetic algorithm (RKGA) is designed to solve the transformed model. Moreover, several simulation examples are presented to verify the feasibility and effectiveness of the proposed model and algorithm. The results indicate that decentralized deployment is preferred for low RSU failure probabilities, while centralized deployment is preferred for high failure probabilities.