Satellite navigation systems like GPS, Galileo, GLONASS and BeiDou provide global positioning and timing services that require high accuracy and reliability. Monitoring the performance of these systems is crucial for ensuring precise and robust navigation capabilities. Traditional monitoring techniques analyze satellite data to assess navigation accuracy and service quality. Key parameters like dilution of precision (DOP) and positioning errors are evaluated based on satellite geometry and measurements. However, these methods have limitations in modeling complex performance factors over time. Recent advances in deep learning provide new opportunities to enhance satellite navigation monitoring. Deep neural networks can uncover hidden patterns and dynamics in large volumes of satellite data. Models like convolutional neural networks (CNN) and long short-term memory (LSTM) are well-suited for satellite image processing and time series forecasting tasks respectively. A deep learning approach can fuse multi-source data from global sensor networks for a comprehensive view of system performance. The neural networks can learn complex mappings between raw satellite observations and quality metrics like DOP, timing drifts and positioning accuracy. Recurrent models can also estimate future degradations based on detected trends. Key research priorities include assembling high-quality training data, selecting network architectures, and interpreting model outputs. Challenges include generalizing to unseen defects and new satellite constellations. With sufficient data and validation, deep learning can significantly improve satellite monitoring to enable robust navigation as systems scale and evolve. In summary, deep learning holds substantial promise for enhancing satellite navigation quality assessment and prediction. With rigorous research, automated deep learning could become an integral technology for reliable high-precision positioning across the globe. The outlook is positive for these techniques to mature and integrate with existing monitoring infrastructure.
 Karolina Krzykowska and Michal Krzykowski. “Forecasting parameters of satellite navigation signal through artificial neural networks for the purpose of civil aviation”. International Journal of Aerospace Engineering, 2019.
 Shizhuang Wang, et al. “Enhancing navigation integrity for Urban Air Mobility with redundant inertial sensors”. Aerospace Science and Technology, 126 (2022), p. 107631.
 Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. “Deep learning”. Nature, 521.7553 (2015), pp. 436-444.
 Sepp Hochreiter and Jürgen Schmidhuber. “Long short-term memory”. Neural computation, 9.8 (1997), pp. 1735-1780.
 Li, B., Zhao, K., & Shen, X. Dilution of precision in positioning systems using both angle of arrival and time of arrival measurements. IEEE Access, 8, 2020, 192506-192517.
 B.Pattanayak and L. Moharana. Analyzing the Effect of Dilution of Precision on the Performance of GPS System. Retrieved from: https://ieeexplore.ieee.org/abstract/document/9428982, 2021.
 Zhong Zheng, Xiaoting Wang, Shan Luan, Hanyu Zheng, Minglong Pu, and Wei Zhang. "Error analysis of received signal strength-based visible-light positioning using dilution of precision." Optical Engineering, 60(10), 106102 (5 October 2021). https://doi.org/10.1117/1.OE.60.10.106102.
How to cite this paper
Civil Aviation Satellite Navigation Integrity Monitoring with Deep Learning
How to cite this paper: Maosen Lin. (2023) Civil Aviation Satellite Navigation Integrity Monitoring with Deep Learning. Advances in Computer and Communication, 4(4), 260-264.