Iteratively Decoded Bit Interleaved Coded Modulation: A Robust Reduced Complexity Design For Fading Channels

University Of Washington, Seattle WA

Investigators

Abstract

This research involves coded modulation, the combination of a convolutional code and a higher order digital modulation. The typical application is for a wireless communications system operating over fading channels. The design typically trades off performance versus complexity. More complex systems are larger in size and consume more power. High performance means high spectral efficiency (in bits per second per Hz), low error rate (probability of bit error) at minimum transmitted power. Modern systems basically achieve this with sophisticated receivers; decoders that are now feasible due to the advances in microelectronics. In this research, the goal is to improve the performance of bit interleaved coded modulation (BICM) by incorporating iterative decoding (BICM-ID). Systems using BICM-ID achieve high performance but are less complex than those incorporating turbo codes. More specifically, the investigator studies fundamental advances both the design and the performance evaluation of BICM-ID systems. This is done is several ways. The research optimize the codes, signal labels, bit interleaver in the encoder. Iterative decoding is improved by better matching with the specific channel type, and quality of the channel state information (CSI). Performance evaluation will be based on a more fundamental understanding of the error-free feedback bound for iterative decoding. The performance for channels with imperfect CSI and more general fading statistics is studied. Diversity will be improved in several ways: through explicit use of multiple channels and through constellation rotation. In all of these situations, design optimization and performance evaluation is investigated.