This paper studies the lightweight deep learning object detection algorithm to detect ship targets in SAR images that can be deployed on-orbit and accessed in the space-based IoT. Traditionally, remote sensing data must be transferred to the ground for processing. With the vigorous development of the commercial aerospace industry, computing, and high-speed laser inter-satellite link technologies, the interconnection of everything in the intelligent world has become an irreversible trend. Satellite remote sensing has entered the era of a big data link with IoT. On-orbit interpretation gives remote sensing images expanse application space. However, implementing on-orbit high-performance computing (HPC) is difficult; it is limited by the power and computer resource consumption of the satellite platform. Facing this challenge, building a processing algorithm with less computational complexity, less parameter quantity, high precision, and low computational power consumption is a key issue. In this paper, we propose a lightweight end-to-end SAR ship detector fused with the vision transformer encoder: YOLO−ViTSS. The experiment shows that YOLO−ViTSS has lightweight features, the model size is only 1.31 MB; it has anti-noise capability is suitable for processing SAR remote sensing images with native noise, and it also has high performance and low training energy consumption with 96.6 mAP on the SSDD dataset. These characteristics make YOLO−ViTSS suitable for porting to satellites for on-orbit processing and online learning. Furthermore, the ideas proposed in this paper help to build a cleaner and a more efficient new paradigm for remote sensing image interpretation. Migrating HPC tasks performed on the ground to on-orbit satellites and using solar energy to complete computing tasks is a more environmentally friendly option. This environmental advantage will gradually increase with the current construction of large-scale satellite constellations. The scheme proposed in this paper helps to build a novel real-time, eco-friendly, and sustainable SAR image interpretation mode.

The high-accuracy and high-stability space-based time system is necessary for satellite navigation systems to achieve high quality of service (QoS) on navigation and positioning in smart city applications. This paper proposes a precise and high-stability space-based time system established under the autonomous time scale of navigation satellites. The generation, maintenance, and transfer of high-precision space-based time references are researched. A centralized time comparison method based on the ALGOS algorithm conducts the two-way time comparison of the inter-satellite link. Specifically, using the relative clock difference observations of all links between satellites for a certain period of time, the clock difference, clock speed, and clock drift parameters of $n-1$ stars in a constellation of n stars relative to the same reference can be estimated simultaneously. Simulations are conducted on real collected data from the Beidou navigation systems when providing services to smart cities around the world. The simulation results show the high accuracy and stability of the proposed space-based time system under the autonomous time scale reference. Moreover, the clock offset monitoring arc coverage is much higher than the satellite clock offset obtained by the direct observation of the satellite and the anchor station. It proves the efficiency of the proposed space-based time system to be used for satellite clock offset modeling and prediction.

Faced with the complex space electromagnetic environment, the long solar illumination periods of medium-Earth-orbit (MEO) satellites, and the unique operating limitation under which the batteries operate under shallow charge–discharge cycles, this paper proposes a dedicated life evaluation method suitable for MEO satellite batteries. The proposed method quantifies battery performance degradation by analyzing the statistical correlation between the discharge capacity and discharge cut-off voltage during eclipse phases. Using seven years of on-orbit-measured data covering 14 eclipse seasons, we build prediction models for the cut-off voltage via Random Forest and XGBoost, with the eclipse season index and discharge capacity set as model input features. The mean squared error (MSE) of both models is lower than 0.003, demonstrating the high reliability of the modeling framework. Under the condition of a fixed discharge capacity, no obvious battery degradation exceeding the measurement uncertainty is detected within the observation period. The magnitude of potential voltage degradation is comparable to or even smaller than the 2 mV acquisition uncertainty, such that no measurable degradation phenomenon can be identified in the available dataset. At present, long-term on-orbit-measured data of MEO satellite batteries are scarce in engineering research. This study provides valuable full-measurement data support for the optimized design, autonomous on-orbit management, and life evaluation of NCA lithium-ion batteries and fills the gap in long-term on-orbit datasets for MEO satellite power systems.