Análise do impacto do algoritmo de escalonamento de recursos no desempenho de redes veiculares utilizando o LTE como tecnologia de acesso

Abstract

Intelligent Transport Systems are several information and communication technologies applied to vehicular transportation. It is called intelligent due to its memory capacity, sensing, communication, information processing and adaptive behavior. Within this universe, the Vehicular Ad-hoc Networks (VANET) provide communication between vehicles, as well as the telecommunications infrastructure over the roads. In order to achieve connectivity continues with high data rate and low latency, the integration of vehicular networks with LTE system is pointed as a potential access technology solution. There are many challenges found to reach the QoS requirements characteristic of entertainment applications, road safety and traffic efficiency. Therefore, this dissertation proposes a methodology for the study of the impact of radio resource scheduling algorithms for vehicular networks using LTE as the access technology. This study uses a discrete simulation approach, carried out with NS-3 and SUMO tools, to evaluate five scheduling algorithms: RR, PF, TTA, CQA and PSS. The criteria used in the analysis are throughput, delay, packet loss, spectral efficiency, fairness and qualitative analysis of computational complexity. The results indicate for vehicular networks is more appropriate to apply strategies that guarantee high level of fairness between connected vehicles. The evaluation also confirm the need to use algorithms more specific, taking into account the data generation pattern of these kind of application, with the aim to ensure the best performance for the network.