The rapid growth in voice and data communications services over satellite and wireless communication networks has resulted in the rapid incorporation of the state of the art in digital communication techniques into current systems. The next generation of these networks will demand even more sophisticated techniques at every point in the system. It is evident that channel coding will be ubiquitous in order to combat channel transmission effects, to more efficiently utilize system resources and to facilitate multiuser access. This project is concerned with the investigation of new classes of channel codes to meet these demands. In particular, new classes of time varying codes are being developed to reduce transmitter power requirements and to enable systems to operate close to theoretical limits with several levels of quality of service. In addition, new iterative decoding algorithms and soft information exchange paradigms are being developed to provide robust integrated solutions to a variety of system functions such as timing and synchronization and multiuser detection.
This project involves the design, analysis and performance of time-varying channel codes in the context of the basic theorems of information theory and the demands of the next generation of satellite and wireless communication systems. The classes of codes being studied include parallel and serial concatenation of block and trellis codes and new classes of periodically time varying convolutional codes. The performance of these coding schemes is studied for both the additive white Gaussian noise channel and a variety of multipath fading channel models. A particular effort is being made to develop codes, including asymmetric parallel concatenations, with short block lengths and near capacity performance at a variety of error rate for mobile wireless communications. The principle of iterative decoding and soft information exchange is also being exploited to enhance other system functions. This includes low signal to noise ratio timing and synchronization algorithms that interact with soft output channel decoding algorithms and novel joint source/channel coding structures. Iterative decoding of channel codes is also being investigated as a means of developing iterative multiuser detection algorithms that achieve near single user capacity on highly loaded multiuser systems.
