MIMO-OFDM系统迭代球形译码检测算法

An iterative sphere decoding detection algorithm in a MIMO-OFDM system

在MIMO-OFDM信号检测中,应用迭代译码的比特交织编码调制(BICM-ID)可有效地在获得分集增益和编码增益的基础上提高检测精度.但检测复杂度会随星座点数与发射天线数指数倍增加,代价较高.为了避免迭代检测所增加的复杂度,用较小复杂度的MAX-LOG-MAP译码器来代替MAP译码器来进行卷积译码,且将MAP解调过程用改进的球形列表译码算法完成以进一步降低复杂度.仿真结果表明,与其他迭代译码方式相比,应用此迭代球形检测译码算法可以使迭代译码的复杂度显著降低,且可以随星座大小及发射天线数自适应地改变,具有较强的鲁棒性.

In a MIMO-OFDM system,the coding and the diversity gains can be obtained easily by using the BICM-ID method.However,the increase in complexity is especially significant for higher order constellations and transmitting antennas.The complexity of detection will increase exponentially with the numbers of constellations and transmitting antennas.In the following research,in order to reduce the complexity due to the iteration,the decoding process was replaced with MAX-LOG-MAP to approximate the original MAP decoding performance,and posteriori information was applied from the decoder to modify a list of candidate points in the constellation.The simulation result shows that,compared to the conventional algorithm,the iterative sphere decoding detection algorithm can obviously reduce the complexity of the iterative decoding algorithm,the spherical radius of the proposed algorithm can adaptively change with the channel parameters and the size of the modulation order;it can efficiently reduce the complexity of decoding with only a slight loss in performance.