Multi-level SINR thresholding for reduced complexity MIMO detection

In previous research works, it was established that the post-processing signal-to-interference-and-noise ratio (SINR) distribution of MMSE (Minimum Mean Square Error) detector shows the potential of high reliability in the high-SINR detected symbols. In this paper, a new algorithm with reduced complexity is proposed exploiting the fact that reliability of MMSE estimated symbols increases with increase in post-processing SINR. For high SINR region, MMSE estimated symbols are retained as the final decisions and the remaining symbols are grouped into different regions based on the SINR. Based on the group in which the symbols lie, we restrict the search space and feed it into SD (Sphere Decoder). A SD-SE (Schnorr-Euchner) implementation of SD algorithm is used for comparison of results.