Electrical and Computer Engineering

NSF Research Focuses on High Performance Wireless Communications, continued

In a $1 million project funded by the National Science Foundation's Advanced Networking Infrastructure program, ECE faculty Timothy Brown, Dragan Maksimovic, Zoya Popovic and Mahesh Varanasi are using an integrated approach to minimizing power consumption aimed at achieving more than a ten-fold reduction in power consumption compared to current systems.

These savings will open the door for smaller, lighter, more reliable, and more capable laptop computers, personal digital assistants, cell phones and other battery-powered devices.

Varanasi is working to develop better modulation and multiple-access methods for multi-antenna communications, along with better detection and decoding techniques. This involves looking at space-time communication, in which messages are smartly encoded in signals that are related both spatially (across antennas) and sequentially (in time) to make significantly better use of power and the spectrum for simultaneous wireless transmissions on multiple antennas.

"Our results to date include a powerful tool to analyze coherent, differentially coherent, and non-coherent space-time receivers in a unified way, which gives asymptotically tight expressions for pair-wise error probabilities," Varanasi says. "We've also developed a computationally efficient recursive technique, the Successive Update Method, for designing large-space-time signal constellations for non-coherent communication."

These signal constellations ensure a "full order of diversity," and can be used to improve wireless communication for people riding on high-speed trains or in automobiles, where channels associated with the transmit-receive antenna pairs change too rapidly to be measured at the receiver.

Brown is investigating the role communication protocols play in energy consumption and has developed algorithms for adjusting when and how data is sent in order to significantly reduce the power needed to communicate.

Implementation aspects of the project are tackled by Dragan Maksimovic and Zoya Popovic. Maksimovic's research group has developed an adaptive power management architecture that enables power savings in signal processing electronics for wireless communication.

Results achieved by Popovic and her graduate students include high-efficiency power amplifiers and active antenna arrays to reduce power consumption and provide the components necessary for efficient implementation of spatial diversity techniques.