Soft-Output MMSE MIMO Detector under Imperfect Channel Estimation

In reality, this is no such thing as perfect channel estimation. But unfortunately, channel estimation errors have never been taken into account by conventional soft-output minimum mean square error (MMSE) multiple-input multiple-output (MIMO) detector when calculating the log-likelihood ratio (LLR) for each coded bit, i.e., the soft information. As a result, its system performance can be significantly degraded. In this paper, we propose a novel soft-output MMSE MIMO detector by the random vector theorem, which takes channel estimation errors into account in the computation of the MMSE filter and LLR of each coded bit. When compared with conventional soft-output MMSE MIMO detector, our simulation results show that the proposed novel detector can remarkably reduce the residual error floor due to the channel estimation errors, while achieving significant performance gain at the cost of negligible increase of complexity. Furthermore, it is observed from our simulation results that the proposed detector is not sensitive to the inaccuracy of channel estimation error statistics. Therefore, it can be a promising candidate to be applied in practical systems.