The BeiDou Navigation Satellite System has seen global deployment and large-scale application. Consequently, its distinctive short message communication service is experiencing explosive growth. However, actual communication processes involve many non-cooperative users. These users are often characterized by excessive transmission frequency and power, alongside a lack of effective state management. They cannot actively respond to outbound beams and frame numbers. As a result, traditional positioning methods struggle to locate them accurately. To address the above challenge, this paper proposes a positioning method for BeiDou RDSS non-cooperative users. The method integrates a spatiotemporal grid with deep learning. First, a multi-epoch spatiotemporal grid feature sequence is constructed through discretization. This transforms the non-convex positioning problem into a classification–regression problem on a discrete grid. Second, based on multi-epoch raw observation features, a physically derived feature set is further constructed. It includes pseudorange differences, the rate of change of time differences, and the mean of multiple inbound time differences. This set enhances the model’s ability to perceive satellite geometry and the spatial–temporal characteristics of signals. Finally, a deep residual network architecture is designed. It mines the nonlinear mapping between observations and geographic locations, thereby achieving high-precision positioning. Simulation experiments are conducted using real ephemeris data and user samples covering the entire territory of China. Simulation results demonstrate that the proposed method substantially reduces positioning errors compared to conventional approaches, providing a new technical pathway for high-precision localization of large-scale non-cooperative targets .
[1] YANG Y X, LIU L, LI J L, et al. Featured services and performance of BDS-3[J]. Science Bulletin, 2021, 66(20): 2135-2143.
[2] 杨元喜，姚铮，毛悦，等。星基弹性 PNT 体系设计及其关键技术 [J]. 中国科学：地球科学，2025, 55 (3): 686-698.
YANG Y X, YAO Z, MAO Y, et al. Resilient satellite-based PNT system design and key technologies [J]. Science China Earth Sciences, 2025, 55 (3): 686-698 (in Chinese).
[3] 谢军，庄建楼，康成斌。基于北斗系统的物联网技术与应用 [J]. 南京航空航天大学学报，2021, 53 (3): 329-337.
XIE J, ZHUANG J L, KANG C B. Internet of Things technology and application based on BeiDou system [J]. Transactions of Nanjing University of Aeronautics and Astronautics, 2021, 53 (3): 329-337 (in Chinese).
[4] 孙光才，王裕旗，高昭昭，等。一种基于短合成孔径的双星干涉精确定位方法 [J]. 电子与信息学报，2020, 42 (2): 472-479.
SUN G C, WANG Y Q, GAO Z Z, et al. A dual-satellite interferometric precise positioning method based on short synthetic aperture [J]. Journal of Electronics & Information Technology, 2020, 42 (2): 472-479 (in Chinese).
[5] YU Z, CHEN X, WEI Q, et al. Bandwidth-efficient TDOA estimation in generalized differential positioning with OFDM satellite signals of opportunity[J]. IEEE Transactions on Vehicular Technology, 2025.
[6] 王鼎，尹洁昕，张欣光，等。一种基于加权多维标度分析的多个非相关源 TDOA/FDOA 协同定位方法 [J]. 航空学报，2023, 44 (7): 227-250.
WANG D, YIN J X, ZHANG X G, et al. A weighted multidimensional scaling analysis based cooperative localization method for multiple uncorrelated sources using TDOA and FDOA measurements [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44 (7): 227-250 (in Chinese).
[7] 姚山峰，贺青，欧阳鑫信，等。一种低轨双星高脉冲重复频率雷达信号的定位模糊消除算法 [J]. 电子与信息学报，2021, 43 (3): 598-605.
YAO S F, HE Q, OUYANG X X, et al. An ambiguity resolution algorithm for LEO dual-satellite geolocation with high PRF radar signals [J]. Journal of Electronics & Information Technology, 2021, 43 (3): 598-605 (in Chinese).
[8] TING S, YONG G. TDOA estimation of dual-satellites interference localization based on blind separation[J]. Journal of Systems Engineering and Electronics, 2019, 30(4): 696-702.
[9] ZHAO L, ZHU X, ZHANG G, WANG Z. A positioning method and realization on single satellites in different orbits using TDOA[J]. China Communications, 2023, 20(7): 108-121.
[10] 邢楠，唐成盼，李晓杰，等。北斗三号 GEO 卫星的 RDSS 定轨精度分析 [J]. 中国科学：物理学 力学 天文学，2021, 51 (1): 96-102.
XING N, TANG C P, LI X J, et al. RDSS orbit determination accuracy analysis of BDS-3 GEO satellites [J]. Science China Physics, Mechanics & Astronomy, 2021, 51 (1): 96-102 (in Chinese).
[11] 孙霆，董春曦。传感器参数误差下的运动目标 TDOA/FDOA 无源定位算法 [J]. 航空学报，2020, 41 (2): 262-271.
SUN T, DONG C X. TDOA/FDOA passive localization algorithm for moving targets with sensor parameter errors [J]. Acta Aeronautica et Astronautica Sinica, 2020, 41 (2): 262-271 (in Chinese).
[12] ZHAN D, WANG S, CAI S, ZHENG H, XU W. Acoustic localization with multi-layer isogradient sound speed profile using TDOA and FDOA[J]. Frontiers of Information Technology & Electronic Engineering, 2023, 24(1): 164-175.
[13] GEERAERT J L, MCMAHON J W. Dual-satellite geolocation with ephemeris correction and uncertainty mapping[J]. IEEE Transactions on Aerospace and Electronic Systems, 2020, 56(1): 723-735.
[14] 彭勃，白吉康，陈伟文，等。基于深度学习的无人机搜救方法研究进展 [J]. 航空学报，2025, 46 (23): 7-24.
PENG B, BAI J K, CHEN W W, et al. Research progress on UAV search and rescue methods based on deep learning [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46 (23): 7-24 (in Chinese).
[15] WANG Z, HU D, ZHAO Y, HU Z, LIU Z. Real-time passive localization of TDOA via neural networks[J]. IEEE Communications Letters, 2021, 25(10): 3320-3324.
[16] PAWASE C J, REHMAN A U, CHANG K. Enhanced 6G non-terrestrial network link performance using deep learning-based channel estimation and Doppler compensation techniques[J]. Journal of Communications and Networks, 2026, 28(1): 73-85.
[17] KHICHAR S, MENG Y, SHARMA A, et al. Residual learning based convolution neural network for improved channel estimation for VehA channel[J]. Scientific Reports, 2025, 15(1): 23475.
[18] JABEUR R, ALAERJAN A, CHIKHA H B. Deep residual network enhanced with multilevel residual-of-residual for automatic classification of radio signals for 5G and beyond systems[J]. Scientific Reports, 2026.
[19] LIU J, AHMAD F A, SAMSUDIN K, et al. Deep residual networks with convolutional feature extraction for short-term load forecasting[J]. Scientific Reports, 2026.
[20] XIN X, et al. Gaussian mixture filter-incorporated self-attention residual neural network for UAV joint error identification and target localization[J]. IEEE Transactions on Instrumentation and Measurement, 2025, 74: 1-14, Art no. 6509914.
[21] DENG S, ZHANG W, XU L, YANG J. RRIFLoc: radio robust image fingerprint indoor localization algorithm based on deep residual networks[J]. IEEE Sensors Journal, 2023, 23(3): 3233-3242.
[22] 宋伟伟，林巍，楼益栋，等。北斗 + 5G 时空基准的 TDOA 定位方法研究 [J]. 武汉大学学报 (信息科学版), 2023, 48 (7): 1170-1179.
SONG W W, LIN W, LOU Y D, et al. TDOA positioning method based on BDS+5G spatiotemporal reference [J]. Geomatics and Information Science of Wuhan University, 2023, 48 (7): 1170-1179 (in Chinese).
[23] 康玉祥，陈果，尉询楷，等。深度残差对冲网络及其在滚动轴承故障诊断中的应用 [J]. 航空学报，2022, 43 (8): 63-74.
KANG Y X, CHEN G, WEI X K, et al. Deep residual hedge network and its application in rolling bearing fault diagnosis [J]. Acta Aeronautica et Astronautica Sinica, 2022, 43 (8): 63-74 (in Chinese).
[24] YING M, CHEN X, QI Q, GERSTACKER W. Deep learning-based joint channel prediction and multibeam precoding for LEO satellite Internet of Things[J]. IEEE Transactions on Wireless Communications, 2024, 23(10): 13946-13960.
[25] ZHAO M, ZHONG S, FU X, TANG B, DONG S, PECHT M. Deep residual networks with adaptively parametric rectifier linear units for fault diagnosis[J]. IEEE Transactions on Industrial Electronics, 2021, 68(3): 2587-2597.
CJA