Real-Time Machine Learning for Multi-User Massive MIMO: Symbol Detection Using Multi-Mode StructNet
Real-Time Machine Learning for Multi-User Massive MIMO: Symbol Detection Using Multi-Mode StructNet
复制标题
DOI：
10.1109/twc.2023.3268945
复制
发表时间：
2023-12
影响因子：
10.4
通讯作者：
Lianjun Li;Jiarui Xu;Lizhong Zheng;Lingjia Liu
中科院分区：
文献类型：
--
作者：
Lianjun Li;Jiarui Xu;Lizhong Zheng;Lingjia Liu
本文针对大规模MIMO-OFDM系统，开发了一种基于学习的符号检测算法。为了利用海量天线阵列接收信号中所继承的结构信息，采用多模库计算作为构件，便于在时域进行空中训练。此外，我们的算法采用交替递归最小二乘优化方法和决策反馈机制来实现实时学习能力。也就是说，神经网络是纯在线训练的，其权值以OFDM符号为基础更新，能够及时自适应地跟踪动态环境。此外，设计了基于在线学习的模块来补偿射频电路元件引起的非线性失真。在此基础上，在频域引入StructNet分类器，利用QAM星座结构模式进一步提高符号检测性能。评估结果表明，在动态信道环境和射频电路非线性失真情况下，我们的算法比传统的基于模型的方法和最先进的基于学习的技术取得了显著的增益。此外，经验结果表明，我们的神经网络模型对训练标签误差具有鲁棒性，这有利于决策反馈机制。
In this paper, we develop a learning-based symbol detection algorithm for massive MIMO-OFDM systems. To exploit the structure information inherited in the received signals from massive antenna array, multi-mode reservoir computing is adopted as the building block to facilitate over-the-air training in time domain. In addition, alternating recursive least square optimization method, and decision feedback mechanism are utilized in our algorithm to achieve the real-time learning capability. That is, the neural network is trained purely online with its weights updated on an OFDM symbol basis to promptly and adaptively track the dynamic environment. Furthermore, an online learning-based module is devised to compensate the nonlinear distortion caused by RF circuit components. On top of that, a learning-efficient classifier named StructNet is introduced in frequency domain to further improve the symbol detection performance by utilizing the QAM constellation structural pattern. Evaluation results demonstrate that our algorithm achieves substantial gain over traditional model-based approach and state-of-the-art learning-based techniques under dynamic channel environment and RF circuit nonlinear distortion. Moreover, empirical result reveals our NN model is robust to training label error, which benefits the decision feedback mechanism.