Abstract:
The study of Joint Source-Channel Coding (JSCC) systems faces one majorchallenge in obtaining an analytical expression for the function thatlinks end-to-end distortion with channel signal-to-noise ratio, theD-SNR curve. A simple solution is to assume that the source is encodedoptimally and transmitted at a rate equal to the channel capacity. Otherapproaches rely on bounds developed by resorting to high and low SNRapproximations and asymptotically large source code dimension withinfinite complexity and delay. Unfortunately, these approaches do noteasily lend themselves well for applications of multimedia wirelesscommunications subject to strict delay constraints. In this talk, wewill discuss the properties of the D-SNR curve for multimedia systemsusing practical source and channel codecs. We will see that theseproperties can be used to obtain a simple closed-form expression for theD-SNR curve. This result will be applied to study issues arising fromusing practical source and channel codes, including the effects onperformance of channel codes of different strength or source codes withdifferent compression efficiency. In addition, the expression for theD-SNR curve will be used to study the performance of some cross-layersystems, such as those involving the use of user cooperation.
Biography:
Andres Kwasinski received in 1992 his diploma in Electrical Engineeringfrom the Buenos Aires Institute of Technology, Buenos Aires, Argentina,and the M.S. and Ph.D. degrees in Electrical and Computer Engineeringfrom the University of Maryland, College Park, Maryland, in 2000 and2004, respectively. He is currently an Assistant Professor at theDepartment of Computer Engineering, Rochester Institute of Technology,Rochester, New York. Prior to this he was with Texas Instruments Inc.,the Department of Electrical and Computer Engineering at the Universityof Maryland, and Lucent Technologies. His research interests are in thearea of multimedia wireless communications and networking, cross layerdesigns, multiple access to wireless networks, user cooperativecommunications, digital signal processing and speech, image and videoprocessing for signal compression and communication.