Iterative Decision Feedback Equalization and Decoding for Rotated Multidimensional Constellations in Block Fading Channels

It is known that rotated multidimensional constellations can be used effectively to achieve full-rate and full-diversity transmission in block fading channels. However, optimal decoding complexity is exponential with the number of fading blocks (or degrees of freedom). In this paper, we propose a reduced-complexity iterative receiver structure operating on a block basis for coded modulation schemes with rotated constellations. The proposed detector is based on iterative forward and backward filtering followed by a channel decoder that uses a priori log- likelihood ratios (LLR) of coded symbols. Forward and feedback filters are jointly optimized according to the minimum mean square error (MMSE) criterion to minimize the spatial interference induced by rotation. It is observed that the proposed structure achieves full diversity and performance close to outage probability for rotated inputs even with simple Discrete Fourier Transform (DFT) rotations.