星地量子通信已经验证了广域量子通信网络的可行性，面对未来量子通信网络多用户的特点，能够准确、快速预测成码率是高效利用星地量子网络资源的核心问题。提出了一种基于机器学习及恒星星像图像识别的信道预测新方法，并将此方法应用于北京地面站的观测中。实验结果表明，恒星星像的图像识别正确率可达88%，并给出是否开展星地实验的建议。在建议开展星地对接的信道情况下，预估该时段量子卫星北京地面站在仰角39.5°的筛选成码率约为8~9 kbit/s，实际星地量子通信实验的筛选成码率为8.8 kbit/s。实验结果可用于合理安排多颗卫星、多个地面站的星地对接任务，提高星地量子通信的成功率，避免浪费卫星和地面站资源，推动量子通信卫星组网的实用化研究。

Single-photon light detection and ranging (lidar) offers single-photon sensitivity and picosecond timing resolution, which is desirable for high-precision three-dimensional (3D) imaging over long distances. Despite important progress, further extending the imaging range presents enormous challenges because only a few echo photons return and are mixed with strong noise. Here, we tackled these challenges by constructing a high-efficiency, low-noise coaxial single-photon lidar system and developing a long-range-tailored computational algorithm that provides high photon efficiency and good noise tolerance. Using this technique, we experimentally demonstrated active single-photon 3D imaging at a distance of up to 45 km in an urban environment, with a low return-signal level of ～1 photon per pixel. Our system is feasible for imaging at a few hundreds of kilometers by refining the setup, and thus represents a step towards low-power and high-resolution lidar over extra-long ranges.

We report a quantum key distribution (QKD) system that uses light-emitting-diodes (LEDs) at 1310 nm as optical sources. Compared to the normally used laser diodes (LDs), LEDs are easier to manufacture and integrate, and thus have the potential to reduce the costs of practical systems. To demonstrate the feasibility of a low-cost, integratable QKD system that aims at meeting the demand of the last-mile secure communication, we utilize LEDs at 1310 nm as the optical sources, while using only passive optical components and only one single photon detector at the receiver’s side. With a repetition rate of 10 MHz, we obtain secure key rates of 10.9 kbps within the experimental time of 1000 s over a fiber length of 1 km.