Adaptive candidate selection scheme in QRM-MLD algorithm for MIMO detection

This paper proposes an adaptive candidate selection scheme in the QRM-MLD algorithm for MIMO detection. The QRM-MLD is a near-ML detection algorithm which can achieve a tradeoff between the BER performance and the computational complexity in the MIMO systems. In this paper, we adopt an adaptive candidate selection scheme into the QRM-MLD. First, similarly to the conventional QRM-MLD, the proposed detection applies a fixed number of the survived branches to achieve a near-ML performance. Next, in order to evaluate the reliability of the survived branches in each detection layer, we introduce a ratio function of the path metric to the smallest path metric among the survived branches. The survived branch with lower reliability has less children nodes as the candidates in the next detection layer, which can avoid a large amount of the path metric evaluations and sorting. Hence, the complexity of the proposed detection should be low. Numerical results exhibit that the proposed scheme achieves the near-ML performance with lower complexity compared to the conventional QRM-MLD.