A close-to-capacity dirty paper coding scheme

An information theoretic framework for the study of efficient known interference cancellation technique is presented in this paper. The dirty paper channel model is given where the arbitrary interference is known at the transmitter is a statistically independent Gaussian random variable with variance. The link to precoding was made and it was shown that the full capacity might be achieved using lattices and MMSE scaling for arbitrary interference. A BCJR detector computes the a posteriori probability (APP) values of the channel code bits, summing over the corresponding coset. Extrinsic information is iteratively passed between BCJR detector and channel decoder. Code design is done using the EXIT chart, achieving an improvement over optimal scalar quantization (SQ).