PROJECT TITLE :

Software-Defined Vehicular Networks With Trust Management A Deep Reinforcement Learning Approach

ABSTRACT:

The proper design of a vehicular ad hoc network, also known as a VANET, has become an essential component in the development of an effective smart transportation system. This system enables a variety of applications that are associated with improving both the safety of traffic and the efficiency of transportation. Because of the dynamic nature of VANETs and the absence of any underlying infrastructure, they are susceptible to the danger posed by malicious nodes, which can result in a decrease in performance. Recent developments in software-defined NetWorking (SDN) have made it possible to dynamically manage VANETs in a practical manner. In this article, we propose a novel software-defined trust based VANET architecture (SD-TDQL) in which the centralized SDN controller is served as a learning agent to get the optimal Communication link policy by making use of a deep Q -learning approach. Specifically, the SD-TDQL architecture is based on the trust that can be placed in the software. A joint optimization problem is modeled as a Markov decision process with state space, action space, and a reward function. The trust of each vehicle as well as the reverse delivery ratio are both taken into consideration within this problem. To be more specific, we evaluate the quality of the communication link between the connected vehicles by using a metric known as the expected transmission count (ETX). In addition, we design a trust model to protect against the potentially negative impact of malicious vehicles. The results of the simulation demonstrate that the proposed SD-TDQL framework improves the quality of the link.