Michael P. Fitz and Urbashi Mitra Proposal 9980616, The Ohio State University
This research will extend the state of the art in channel coding for multi-antenna communication systems. It has recently been shown that the capacity of a wireless communications system can be greatly increased if multiple antennae are used both at the transmitter and the receiver. Furthermore, a method for achieving such gains is to introduce controlled redundancy (coding) in the transmitted data sequence spatially as well as temporally, which is the conventional scheme. This area of research is still quite new and several key problem areas need to be addressed. For example, computationally efficient methods for finding space-time codes are necessary. Typically, the wireless channel is time-varying and it is not always possible to achieve good estimates of the channel conditions; thus evaluating the robustness of space-time codes is of value. A related project is to consider the interplay of space-time coding and decoding methods with algorithms designed to estimate the channel and other key communication parameters. Turbo codes have recently emerged as powerful codes for achieving near Shannon capacity, thus space-time generalizations of such codes will be studied. Finally, much of the current work on space-time coding is for narrowband time division multiple access systems, thus we shall consider space-time code design and decoder design for wideband code-division multiple-access systems as well.
The first research thrust will focus on the development and analysis of improved search techniques for space-time trellis codes. Current methods for searching the distance spectrum of space-time trellis codes can be computationally expensive. Methods for reducing the size of the error state trellis coupled with reduced complexity trellis search schemes will be explored. The next focus is on evaluating the robustness of space-time codes under realistic channel conditions. In addition, space-time coding for parametric channels will be considered. Thus, constrained channel uncertainty will exist and code designs for such scenarios will be developed. The third research focus will be on space-time turbo codes as one method to achieve robustness to a variety of channel conditions. The random--like structure of turbo coded schemes hold some promise of mitigating the possible interactions between the code structure and the channel characteristics. Space-time code designs for CDMA channels will be investigated in the fourth research thrust. CDMA signals have a larger dimensionality and thus the space-time code design is different from that for single-user systems. In addition, CDMA signals in a multipath channel will result in different optimal code designs. The final research topic will consider receiver designs in the context of space-time coding. That is, channel estimation and synchronization for space-time coding and modulation will be investigated. Algorithms which exploit the properties of space-time codes will be designed. In addition, the performance dependence on accurate channel estimates and synchronization will be analyzed.
