Joint Rate and Queue Based Routing for Vehicular Ad-Hoc Networks

Abstract

In recent years, Vehicular Ad-hoc Networks (VANETs) have become important and critical part of the Intelligent Transport System (ITS). VANETs have brought the potential of new services and various applications for emergency, comfort and commercial use. However, designing a practical VANET architecture is still hindered with lot of challenges. One such challenge is to route the data packets from source to destination vehicle as per application specific need without sacrificing the Quality of Service (QoS) requirements. In order to design a practical system that meets the requirements of new applications for VANETs, we propose a new cross-layer approach for efficient and reliable routing. Our proposed model considers the wireless transmission parameters from Physical (PHY) layer as well as the system buffer capacities from the Medium Access Control (MAC) layer to make informed and accurate routing decisions at the network layer. This model can be used to cater for specific application requirements in VANET by applying varying routing decisions depending on node specific attributes such as commodity data, queue backlog and rate allocation. The model exemplifies a multi-hop backpressure architecture in VANET where each vehicle is trying to select the next-hop vehicle on the basis of buffer size (queue backlogs), the data to be send (commodity) as well as the channel rate allocated. The key objective here is the optimal selection of next hop vehicle for information dissemination by integrating these metrics. We classify this as a rational Constraint Driven Problem (CDP) and we explore an optimal solution to this problem. Using a systematic mathematical model formulation and extensive simulations, we present the performance analysis of our exemplary model by deriving values of relevant key performance indicators applicable in such an environment.