GENERAL DESCRIPTION:
High data rate communication systems can experience severe distortion when used over wireless channels. To make the system relatively immune to some of those distortions, a particular type of signal design referred to as orthogonal frequency division multiplexing (OFDM) has been suggested. In this research project, we propose to study the performance of OFDM when used in wireless systems that exhibit high degrees of mobility. This mobility makes it much more difficult to design an efficient receiver, and we propose to compare many receiver designs to one another. Further, we propose to build certain key components to an OFDM system.

TECHNICAL ABSTRACT:
This proposal concentrates on some of the more difficult aspects of orthogonal frequency division multiplexing (OFDM) as applied to wireless channels. As is well known, perhaps the key virtue of OFDM when used over a wireless channel is that it significantly relaxes the requirements for adaptive equalization to combat multipath-induced inter-symbol interference (ISI). However, it trades off decreased sensitivity to ISI for a large peak-to-average power ratio, and thus, increased sensitivity to nonlinear amplification; it also displays increased sensitivity to Doppler spreads.

Most of our proposed research will correspond to mobile channels, where the Doppler spreads can cause significant cross-channel interference, and where the problems involved with channel estimation become especially difficult, because as the Doppler spread increases, the coherence time of the channel decreases. Thus, the time available to make a measurement of, say, the state of the fade of a given subchannel, becomes progressively smaller as the relative velocity between the transmitter and the receiver increases.

Our research will have both a theoretical component and an experimental one. Among the theoretical topics that we will concentrate on are the use of new modulation and coding techniques for OFDM systems, a comparison of the performance of an OFDM-based CDMA system with that of a multicarrier direct sequence CDMA system, and the effect of channel estimation errors. On the experimental side, our research will focus on the design and testing of hardware for the implementation of fast FFTs and equalizers, frequency synthesizers, and digital-to-analog converters.

PARTICIPATING FACULTY:
The following CWC faculty are participating in this research project: Larry Milstein(lead PI), John Proakis, Jim Zeidler, and Bang-Sup Song.