(NSF Grant CNS-1116209)
PI: Eytan Modiano

This project will develop mechanism for reliable communications over wireless mesh networks, which have recently emerged as a promising solution for providing last-mile Internet access.  These networks will be tightly coupled with the wired Internet to provide Internet services to end-users. While users of wired networks are accustomed to highly dependable network services, wireless networks have to contend with interference, obstructions, jamming, and mobility, and are notoriously unreliable.  The focus on this project is on providing reliable communications in this challenging environment. In particular, we will adapt, and augment, some of the techniques that have been very successful for providing reliability in wired networks to the wireless environment.

Due to the challenging wireless environment, restoration in wireless networks has been limited to dynamic re-routing of failed paths, with little, if any, protection guarantees. Implicit in wireless networks, is the idea that should a route fail, a new route will be found that can serve as a backup. However, this ”reactive” approach, while simple, cannot be depended upon to provide a high-level of reliability. A major goal of our work, which differentiates us from previous efforts in this field, is to develop “proactive” mechanisms for providing reliable communications in wireless networks.  An important aspect of the proposed research is to design network restoration mechanisms that are tightly integrated with the wireless transmission medium, and take into account interference and the need for scheduling.

These mechanisms will include protection along preplanned backup paths; diversity routing along multiple paths; and spare capacity allocation for providing reliability guarantees. Moreover, many of the diversity concepts, that have been used to provide reliable communications at the physical layer, will be explored at the network layer for route diversity. Methods that utilize redundant transmissions and routing along multiple paths will be explored in order to increase network reliability; and techniques from the field of robust optimization will be used to design a novel backup network architecture that is robust to multiple random failures. Our research goals include:

• Develop new metrics for reliability in wireless networks that go beyond connectivity metrics and take into account throughput requirements.
• Investigate the impact of interference and scheduling on wireless protection and restoration.
• Develop protection mechanisms that reroute failed traffic along preplanned backup paths.
• Develop reactive restoration mechanisms using “diffusion routing” and redundant transmissions.
• Develop reliable network architectures using techniques from the field of robust optimization.

Participants:
Prof. Eytan Modiano, PI
Greg Kuperman, graduate student
Qingkai Liang, graduate student

Related Publications:

1. Matt Johnston, H.W. Lee, E. Modiano, “Robust Network Design for Stochastic Traffic Demands," IEEE Journal of Lightwave Technology, 2013.
2. Chih-Ping Li and Eytan Modiano, “Receiver-Based Flow Control for Networks in Overload," IEEE/ACM Transactions on Networking, 2014.
3. Greg Kuperman and Eytan Modiano, “Providing Protection in Multi-Hop Wireless Networks,” IEEE Infocom, Turin, Italy, April 2013.