Prof. Brian L. Evans
Dept. of Electrical and Computer Engineering
Wireless Networking and
Communications Group
The University of Texas at Austin, Austin, Texas
Lead graduate students: Aditya Chopra, Kapil Gulati and Marcel Nassar
Research performed in collaboration with Keith R. Tinsley and Chaitanya Sreerama at Intel Labs
Thursday, October 9, 2008
Laptops and other computer platforms have many sources of radio frequency interference (RFI) that interfere with wireless reception. These sources include not only clocks and busses, but also power saving subsystems. We have developed methods for wireless receivers to sense and reduce platform RFI. These methods are complementary to the static circuit-level and board-level design methods used to mitigate the emission of platform RFI.
We model platform RFI as impulsive noise. In particular, we use Middleton Class A and Symmetric Alpha Stable impulsive noise models, which appear to fit platform RFI measurements obtained from Intel well. By passively listening to computing platform environment, we sense the platform RFI by estimating the parameters of the impulsive noise models. Given the model parameters, we employ a variety of filtering and detection methods to reduce the rate of bit errors by a factor of 10-100 for the same transmission rate. We evaluate design tradeoffs in the platform RFI sensing and reduction stages.