For wireless multihop networks to be used by thousands, the network has to be able to self-organize, which is what University of California, Riverside researchers are developing at the Bourns College of Engineering.
Self organization means that each wireless node is aware of its neighborhood and can make intelligent decisions about whom to communicate with. Multihop means the network allows a single connection to let many other users “hop onto” the network using the most efficient wireless routes. Creating essentially a wireless web of wireless networks is especially useful where traditional hardwired systems are limited in reach, such as in developing countries or in sparsely populated areas.
Computer Science and Engineering faculty members Srikanth Krishnamurthy, Michalis Faloutsos and Neal Young are working to develop a smart wireless network that reconfigures itself with each connection to optimize its quality and effectiveness. The project has received a three-year, $388,000 grant from the National Science Foundation.
Using the third floor of the Engineering II Building at UCR as their test network, they will determine what the realistic footprint of the wireless signal coming from each node is and how to best design the network that will constantly reconfigure itself to maximize the quality of signals between neighboring transmitters and receivers.
“When you see representations of the reach of a wireless signal, they usually show you a circle radiating from the antenna, but with walls, poles and other interfering devices, you rarely have a circular footprint,” said Krishnamurthy, one of the principal investigators. A goal of the research team is to use realistic assumptions and models.
The work will examine emerging physical layer technologies such as the use of smart antennas while facilitating this reconfiguring of neighboring nodes. The investigators will also develop a wireless teaching laboratory at UCR for both graduate and undergraduate students to perform experiments and understand the practical issues that arise with the network’s implementation.
Source: University of California, Riverside
Explore further: Machine-learning breakthrough paves way for medical screening, prevention and treatment