中心博士生徐绍夫的工作——Adaptive deep learning algorithm for signal recovery of broadband microwave photonic receiving systems based on supervised training（用于宽带微波光子接收系统信号恢复的自适应深度学习算法）的相关成果近期被Journal of Optical Society of America B期刊接收发表，该工作得到了国家重点研发计划（2019YFB2203700）、国家自然科学基金（61822508）的部分资助。该工作提出一种基于深度学习的微波光子接收系统信号恢复算法，通过对数据集的学习，可以自动实现信号高保真重构，无需额外微波光子调控系统。该算法最大的特点为对于多种微波光子接收系统的自适应能力。通过改变训练数据集，该算法可以应用在不同的系统与场景下。实验结果表明，该算法对于微波光子接收系统的动态范围提升了约18dB，并且通过网络特征分析，强化验证了该算法对于更多微波光子接收系统的适应性，并且其噪声鲁棒性也得到了实验结果证实。
摘要: We show an adaptive deep learning algorithm that recovers the distorted broadband signals of defective microwave photonic (MWP) receiving systems. With data-driven supervised training, the adopted neural network automatically learns the end-to-end distortion effects of the photonic analog links and recovers the received signals in the digital domain. Through changing the training datasets and retraining the same neural network, this algorithm can be applied in various MWP receiving systems. Two MWP receiving systems are setup for experimentally demonstrating the capability of broadband signal recovery. Results evidently show that the neural network can reduce the signal distortion (measured with mean square error) by ~18 dB. Moreover, visualization analysis indicates that the proposed algorithm is potentially adaptive to more MWP receiving systems and applications. The noise robustness of this algorithm is also verified so that it is applicable in noisy situations. The proposed algorithm improves the performance of MWP receiving systems through appending a deep learning digital processer, of which the deployment is of low-cost.