Rate adaptation for interference cancelation receivers in slowly time-variant MIMO channels

In this work we investigate rate adaptation for successive interference cancelation (SIC) receivers in per-spatial layer coded MIMO systems using outdated channel state information (CSI). Outdated CSI may cause the rate adaptation to be too optimistic, which results in transmission errors and, in case of SIC receivers, error propagation. By carefully accounting for the error propagation problem it is shown that the SIC gain as compared to linear receivers can be maintained. We show how approximating the log-transformed post equalization SINR statistics by a Gaussian distribution can be used to implement low-complexity rate adaptation for a wide range of velocities.