Joint FEC encoder and linear precoder design for MIMO systems with antenna correlation

We study transmissions in multiple-input multiple-output (MIMO) systems with antenna correlation. We focus on joint forward error correction (FEC) encoder and linear precoder design based on channel covariance information at the transmitter (CCIT). We aim to optimize the system performance using the extrinsic information transfer (EXIT) chart type curve matching principle. By adopting a Hadamard precoding technique, we show that, the EXIT chart type curve of the precoded system is asymptotically determined by the channel correlation matrix at the transmitter as the number of receive antennas tends to infinity. The encoder-precoder curve matching can be made asymptotically accurate even in the lack of full channel state information at the transmitter (CSIT). Excellent performance based on this strategy is demonstrated by simulation results in systems with only a moderately large number of receive antennas.