The Interplay Between Error Control Coding and Iterative Signal Cancelation

The application of error control coding (ECC) algorithms to iterative signal cancelation systems is examined. It is shown that in such "turbo" equalization, or cancelation systems, the role of ECC codes extends beyond the mere goal of minimizing error rates. In an iterative cancelation system, the soft-output ECC algorithm has to guarantee convergence of the iterative process, as well as to establish target BERs at that convergence point. Bounds for the output symbol error variance of a general ECC are presented, and ECC systems are examined which approach these bounds in the low SNIR regime and therefore can tolerate the highest level of interference. Guidelines on how to evaluate ECC codes for use in cancelation systems are developed, and near-optimal code choices are discussed. It is shown that such iterative "turbo" equalizers with suitable ECC codes can achieve the capacity of the channel in the case of isotropic random interference, such as arises in certain multiple access and multiple-input multiple-output systems.