The Centennial Mesh project aims to build a highly programmable, extensible, open testbed to support research and education on the design of wireless mesh networks, as well as IT systems and applications enabled by wireless mesh networks, using open source code built by NCSU students and faculty, and re-using other open source code.
The physical site of the testbed is in the Centennial Campus of NCSU. Centennial Campus is an initiative unique in academic environments, and represents a remarkable visionary coming together of academia, government, and private industry.
Students:
The goal of this project is to develop an inexpensive system for
monitoring environmental conditions in the Chesapeake Bay, the
sounds of North Carolina and other coastal waterways. One one hand,
existing commercial technology is costly to implement on a large
scale, and is not easy to use. On the other hand, traditional
wireless sensor networks are extremely customized and require
two PhDs
to program and use. We attempt to use inexpensive, off
the shelf embedded linux systems (a BeagleBoard with a WiFi card)
to create a new kind of wireless sensor network (WSN). The
networking part is by comparison with traditional WSNs much
simpler, as traditional TCP/IP is employed with all its advantages
(and few disadvantages). Regarding the power, we employ a very low
power microcontroller for reading the sensors and storing the data
(at rates from seconds to minutes), and use the embedded linux
computer only to (relatively infrequently - every few hours) move
the data ashore (and to the Internet) via long WiFi links.
The measured data include Oyster gape opening (16 oysters), water salinity, water clarity (four color channels), water temperature, water depth (including tides and waves). We also monitor the state of charge of the main battery and solar panel. The goal is to correlate oyster activity and water conditions, thus allowing researchers to quantify which conditions are most conducive to growth.
The system has been designed and deployed on Chesapeake Bay (with a solar panel and battery supporting the system through up to a week of cloudy days).
Details about the system implementation are available at the development website.
The project is a collaboration between three NCSU PIs:
Alexander Dean (NCSU - ECE),
Mihail Sichitiu (NCSU - ECE),
Tom Wolcott (NCSU - MEAS),
and two PIs from the
Chesapeake Bay Environmental
Center: Judy Wink and
Vicki Paulas
.
Sponsors: NSF
The goal of this project is to to build a flexible, scalable distributed network emulator that can conveniently and accurately evaluate the performance of large network systems including a variety of network elements. The proposed emulator will combine the convenience of network simulators with the realism of the testbed results, while avoiding the disadvantages of both methods. The emulator will run on a cluster of PCs, efficiently allocating the computing resources to maximize the performance of the tool. The proposed tool will use virtual machines and real code to emulate network elements such as switches, routers and hosts, thus capturing many of the subtle effects affecting real networking systems.
Students:
The goal of this project is to use a distributed wireless packet capture in a large enterprise to analyze performance and security parameters of the system. The main goal is to establish fundamental limits on what information can be obtained by using the (always) incomplete traces.
Students: