Low-complexity detection for large MIMO systems using partial ML detection and genetic programming

We propose a low-complexity detector for multiple-input multiple-output (MIMO) systems using BPSK or QAM constellations. The detector operates at the bit level and is especially advantageous for large MIMO systems. It consists of three stages performing partial ML detection, generation of soft values, and soft-input genetic optimization. For the last stage, we present a genetic programming algorithm that uses the soft values computed by the second stage. Simulation results demonstrate that for large systems, our detector can outperform state-of-the-art methods, and its complexity scales roughly cubically with the system dimension.