A dynamic layer ordering M-paths MIMO detection algorithm and its implementation

During the past decade, multiple-input-multiple-output (MIMO) systems have received a significant attention because of its promising capacity increase over single-input-single-output (SISO) systems. In this paper, as a key technology, we present a novel M-paths MIMO detection algorithm with dynamic-layer-ordering for spatial-multiplexing MIMO systems over rich scattering wireless environment. This algorithm obtains a near-ML detection performance (e.g., only 0.2 dB performance degradation for 4-TX and 4-RX MIMO with 16 QAM) with a much lower computational complexity compared to the ML detection. We also develop an recursive procedure for dynamic-layer-ordering and reduce the ordering complexity from O(N_T ⁴) to O(N_T ³), where N_T is the number of TX antennas. In order to verify the feasibility and performance, a pipelined structure is designed and implemented in FPGA with limited hardware resources and a 200 Mbps throughput.