Complexity reduced soft bit estimation with iterative MMSE-SIC based MIMO detection

Abstract

In this paper, we propose a complexity reduced soft bit estimation for iterative multiple-input multiple-output (MIMO) system with linear order of complexity. A maximum likelihood (ML) detection achieves the maximum performance, but its complexity increases exponentially by the number of antennas and modulation orders. In this paper, we propose an efficient soft bit estimation method for iterative MIMO detection. In the proposed method, symbol-wise detection is first made by using a linear detection scheme, and then soft bit information is extracted from each detected symbol. Simulation results show that the proposed method provides substantial complexity reduction while achieving comparable BER performance to the conventional iterative detection methods.