We propose a neural network equalization delta-sigma modulation (DSM) technique. After performing DSM on the multi-order quadrature amplitude modulation (QAM) orthogonal frequency division multiplexing (OFDM) signal at the transmitting end, neural network equalizer technology is used in the digital signal processing at receiving end. Applying this technology to a 4.6 km W-band millimeter wave system, it is possible to achieve a 1 Gbaud 8192-QAM OFDM signal transmission. The data rate reached 23.4 Gbit/s with the bit error rate at 3.8 × 10^-2, lower than soft-decision forward-error correction threshold (4 × 10^-2).

随着社会信息化程度的逐步提升，人们对光纤通信系统容量的需求持续增长。信道损伤一直是限制光纤通信容量提升的关键因素，信道损伤补偿作为光纤通信系统的基础性功能和关键技术也在不断改进。近年来，将深度学习应用到光纤通信信道损伤补偿中更是吸引了不少研究人员的关注。文章介绍了近年来应用于光纤通信中的常见深度学习模型，阐述了深度学习技术在光纤通信系统信号损伤补偿方面的发展情况。

Recently, deep learning has been used to establish the nonlinear and nonintuitive mapping between physical structures and electromagnetic responses of meta-atoms for higher computational efficiency. However, to obtain sufficiently accurate predictions, the conventional deep-learning-based method consumes excessive time to collect the data set, thus hindering its wide application in this interdisciplinary field. We introduce a spectral transfer-learning-based metasurface design method to achieve excellent performance on a small data set with only 1000 samples in the target waveband by utilizing open-source data from another spectral range. We demonstrate three transfer strategies and experimentally quantify their performance, among which the “frozen-none” robustly improves the prediction accuracy by ∼26 % compared to direct learning. We propose to use a complex-valued deep neural network during the training process to further improve the spectral predicting precision by ∼30 % compared to its real-valued counterparts. We design several typical teraherz metadevices by employing a hybrid inverse model consolidating this trained target network and a global optimization algorithm. The simulated results successfully validate the capability of our approach. Our work provides a universal methodology for efficient and accurate metasurface design in arbitrary wavebands, which will pave the way toward the automated and mass production of metasurfaces.

为解决基于深度学习的立体匹配方法面临着网络规模大、网络结构复杂等问题，提出了一个网络规模较小、精度较高的网络结构。该网络在特征提取模块删减修改了复杂冗余的残差层并引入了空洞卷积金字塔池化模块来扩大视野范围，提取更多有用的上下文信息；在代价计算模块中使用了三维卷积层以成本聚合提升立体匹配的精度；最后，在代价聚合模块引用了双边格网模块以较低分辨率的成本量来获取精度较高的视差图。将该网络在KITTI 2015数据集和Scene Flow数据集等主流数据集上进行实验，结果显示，相较于其他主流优秀网络类如金字塔立体匹配网络（Pyramid Stereo Matching Network ，PSM-Net），网络规模参数量减少了约38%，并取得了较高的实验精度，其中Scene Flow数据集的终点误差（End-point Error，EPE）为0.86，是一个同时兼顾速度与精度的立体匹配网络。