ECE: Electrical & Computer Engineering
ECE News


FINS Framework

Visit the FINS framework website.

MANIAC Challenge

The MANIAC Challenge was funded under a prior NSF NeTS grant.

FINS to accelerate deployment of seamless mobile networks

Virginia Tech ECEs are developing a framework that is expected to bring closer the day when iPhones, Blackberries, laptops and other mobile devices can talk to each other effortlessly, anywhere in the world.

This communications ability will involve mobile devices sending data through each other, without relying on fixed transmission points like cell towers, according to Allen MacKenzie.

Unfortunately for mobile, ad hoc network (MANET) technology, existing network frameworks are based on the assumption that there will be stable routes and established nodes, like cell towers. “Mobile networks invalidate this assumption and throw traditional routing techniques into disarray,” MacKenzie explains.

One of the biggest hassles for researchers is making major modifications to networking software: the software protocols controlling networking (the “stack”) are layered deep within computer operating systems. Even tinkering with a single setting can require a complete rebuild of the computer’s master program – the kernel. At present, no replacement for the traditional, fixed network stack is available, according to MacKenzie.

MacKenzie and Luiz DaSilva have been awarded a $350,000 NeTS grant from the National Science Foundation (NSF) to develop a framework for modular, extensible, experimental network technology.

Their Flexible Internetwork Stack (FINS) Framework will leverage existing protocols, and give researchers real-time control and easy access to data.

FINS will be implemented on hand-held devices, which have the advantage of longer battery life. The team is building the improved hand-held communications technology based on their experiences in deploying large, laptop-based MANETs as part of the MANIAC Challenge.

DaSilva and MacKenzie organized the MANIAC Challenge under a prior NSF NeTS grant. In MANIAC, teams formed an ad-hoc network with each team representing a node in a MANET. Organizers generated data packets destined for each team. Teams were judged according to how many of the packets destined for them they received. To receive a packet, they had to balance selfish interests with cooperation and get other teams to forward their packets from node to node by offering to forward other packets in return.

For the competition, organizers had to modify routing behavior in the networking stack to allow participants to view packet headers and data and to make per-packet forwarding decisions based on their strategy. It was difficult to obtain even partial data without access to all the network layers.

FINS will resolve these issues. DaSilva and MacKenzie plan to spur adoption of the framework by organizing a new design competition using FINS. They will also release networking course modules based on the FINS Framework.

The FINS framework will be openly shared, allowing researchers throughout the world to expand their experimental work and build more robust network models.

FINS will also be understandable to researchers with less experience in the field. This will dramatically improve the ability of undergraduate students, novice graduate students and students enrolled in networking courses to “play” with the network stack. “We want undergraduates to be able to play with the technology: play is critical for developing understanding and innovation,” MacKenzie says.