NetSE: Small: Collaborative Research: Dynamic Flow Equilibria in Vehicular Traffic and Data Communication Networks

Purdue University, West Lafayette IN

Investigators

Abstract

NetSE: Small: Collaborative Research: Dynamic Flow Equilibria in Vehicular Traffic and Data Communication Networks Project Summary This project will develop a new methodology, algorithms, and models for dynamic flow equilibria in networked systems, with emphasis on two specific real-world networking applications ? vehicular traffic network modeling and efficiency analysis, and routing and flow control in data communication networks. The first research goal of this project involves appropriate modeling of dynamic flow problems in networks with self-interested agents, and understanding the properties of the equilibria for these problems. The other major research goal is to specifically study the application of dynamic flows to vehicular traffic networks and data communication networks. The research goals of this project will be achieved by an interdisciplinary but closely-integrated effort bringing together techniques from game theory, algorithm design, optimization, and real-world network simulation. Intellectual Merit: The novelty of this project is that it provides a holistic understanding of the theoretical, algorithmic, and implementation issues of dynamic flow equilibria in networked systems, particularly in the context of vehicular traffic and data communication networks. This research explicitly takes into account the non-negligible travel time of flows in a network, and its variation as a function of the time-varying congestion in the network ? this results in new notions of flow equilibria that are fundamentally different and significantly more complex than those for non time-varying flows. The models used in the project are based on emerging concepts in game theory, optimization theory, and vehicular traffic modeling, and the project is expected to result in the development of a new algorithmic framework for studying dynamic flow equilibria in networks. In addition, this research will contribute techniques for dynamic flow equilibria study to many interested research communities, including transportation, algorithmic game theory, economics, sociology, and multi-agent systems, and facilitate the transfer of techniques between disciplines. Broader Impact: This project will facilitate better network engineering, particularly in the context of transportation and communication networks. The implementation of the models developed by this research in specific disciplines (vehicular transportation and data communication networks) will have significant societal impacts, such as (i) enabling more efficient analysis and provisioning of transportation networks, leading to lower travel delays, and (ii) faster Internet access and download speeds due to more efficient routing and flow control protocols. This research will help rethink and possibly redesign important core components of the Internet through better flow control and routing choices. The findings of this research will be integrated into several different courses, and in Interactive Learning Modules aimed at attracting undergraduate and high-school students to research careers. Special efforts will be made to encourage participation of undergraduate and minority students.