To address spectrum scarcity, complex airspace, and moving obstacles in large-scale low-altitude operations, a UAV decision framework that couples communication, control, surveillance, and trajectory are proposed. Two performance maps, Information Performance Map (IPM) and Surveillance Performance Map (SPM), are built to quantify control-link availability and radar reliability. A cumulative outage constraint ensures both flyability and controllability while three-dimensional point-cloud data are exploited to maximize the air-to-ground rate. A DQN (Deep Q-Network)-based algorithm is then introduced: point-cloud and CNN(Convolutional Neural Network) features are jointly processed to select the next waypoint and vehicle access in a discrete action space, with experience replay and a target network stabilizing training. After 8 000 training episodes, the UAV cruises efficiently through areas of robust control and radar coverage while avoiding blind spots, as rewards converge and losses stabilize. The proposed method offers a scalable solution that balances spectrum efficiency, safety, and regulation.

Given the vulnerability of satellite signals to interference, the research on anti-jamming algorithms based on array antennas becomes crucial to ensure the reliable positioning accuracy of GNSS receivers. However, the existing variable step-size power inversion algorithms rely on the single regulation mechanism of instantaneous energy, which have insufficient stability in the dynamic interference scenarios. This paper proposes a variable step-size anti-jamming algorithm modified by power change rate, and this method adds the power change rate to correct the step-size variation based on the original variable step-size. Through the dual adjustment mechanism of input power normalization and output power change rate correction, the proposed algorithm is promoted to restore quickly the stable convergence state after an abrupt power change. This method effectively mitigates the violent oscillation of the weights caused by rapid power changes, improving both the convergence speed and robustness of the algorithm in the dynamic interference environment. Simulation experiments show that the proposed algorithm achieves faster convergence speed and deeper null depths than a single adjustment mechanism relying solely on instantaneous energy. Moreover, it can effectively cope with the impact of abrupt changes in signal power on interference suppression, thereby reducing interference signal power.

Inertial/Global Navigation Satellite System (GNSS) integrated navigation has been widely applied in various mobile platforms such as unmanned aerial vehicles (UAVs). However, during GNSS signal outages, INS errors accumulate rapidly, severely degrading navigation accuracy. Existing research primarily focuses on horizontal two-dimensional error modeling while neglecting the dynamic characteristics in the vertical (altitude) direction, limiting its practical application in three-dimensional space. To address this issue, this paper proposes a dual-branch neural network model for three-dimensional navigation, which simultaneously models position increments in the longitude, latitude, and altitude directions to cater to the demands of dynamic navigation in 3D space. The model adopts a decoupled dual-branch structure built with LSTM and GRU networks, designing separate modeling paths for the horizontal and vertical components. A convolutional neural network (CNN) is further incorporated into the main branch to enhance temporal feature extraction. Experimental results demonstrate that the proposed network significantly improves three-dimensional navigation accuracy. Compared with conventional positioning methods, it reduces the root mean square error(RMSE) along the east, north, and up axes by 97.8 %, 97.9 %, and 26.2 %, respectively, demonstrating its strong potential for practical deployment.

In autonomous driving within the Internet of Vehicles (IoV), positioning accuracy is key to stable operation. However, standalone navigation systems such as satellite navigation and inertial navigation cannot fully ensure continuous high-precision positioning. Therefore, achieving high-precision positioning through information collaboration between vehicles has become the main approach. This paper proposes a neural network-based large-scale cooperative vehicle positioning method. Aiming at the characteristic of vehicles freely gathering and dispersing during driving, principal component analysis is introduced to process navigation information and reduce computational complexity. Furthermore, the Fireworks Neural Network method is used to rapidly fuse navigation information in the IoV, ensuring positioning accuracy and stability during vehicle operation. Compared with existing cooperative positioning methods, experimental results show that the proposed method has faster convergence and better positioning stability.

The current navigation system relies heavily on GNSS system, which makes the users unable to work in the denied environment. This paper analyzes the navigation and position technology by low orbit satellite, based on satellite receiver's Doppler measurement, this technology can be used to realize the positioning of user on earth. On the condition of orbit altitude 800 km, observing arc segment 8 min, measuring error of Doppler 0.01 Hz, the simulating result is that the user's positioning error is 1 000 m. In the Global Positioning System denied environment, the findings of this study would offer an alternative positioning approach for specific user groups.

At present, GNSS-IR sea surface height retrieval technology relies primarily on traditional geodetic GNSS receivers. However, due to their high cost and portability limitations, these devices struggle to meet the demands for low-cost and high-precision applications, which restricts the widespread adoption of this technology. To address this, our study utilizes smartphones for GNSS-IR sea surface height inversion experiments. By employing robust estimation techniques, we integrated multi-system and multi-frequency data to enhance the accuracy of the sea surface height inversion. The results indicate that the smartphone's overall retrieval accuracy is approximately 21 cm, with an average correlation coefficient of 0.96, performance comparable to that of traditional GNSS receivers. The multi-system multi-frequency data fusion further improved the sea surface height retrieval accuracy to 6.8 cm, with a correlation coefficient of 0.996. Compared to the initial retrieval results, this approach significantly enhanced both the temporal resolution and the accuracy of sea surface height retrieval. After smoothing, the accuracy of the fused retrieval results reached 6.1 cm. These findings demonstrate that smartphones can achieve centimeterlevel accuracy in sea surface height retrieval, making them a viable alternative to traditional GNSS receivers for GNSS-IR applications.

To counteract the influence of the Earth's non-spherical perturbation, Geostationary Earth Orbit (GEO) satellites must periodically execute longitude drift control to maintain their position in the east-west direction. East-west station-keeping is typically achieved via pulse ignition. During this process for geostationary orbit satellites, continuous telecommand must be sent, severely restricting payload applications closely coupled with satellite telecommand. This paper devises a decentralized control approach for the east-west station-keeping of GEO satellites, constructs a mathematical model, formulates control strategies and implementation details, and validates the control effectiveness using two distinct types of in-orbit satellites. This method efficiently exploits fragmented resources during payload task intervals, enhancing the availability of satellite tracking, telemetry, and telecontrol command resources.

With the rapid development of smart vehicles and autonomous driving technology, high-precision 3D navigation has emerged as a crucial supporting technology. However, the limitations of the BeiDou Navigation Satellite System (BDS) in terms of accurate elevation positioning and complex, variable highway traffic scenarios have posed constraints on autonomous and intelligent driving systems. This paper proposes a BDS+5G integrated positioning model based on adaptive Kalman filtering technology, aiming to address the decline in positioning accuracy caused by signal occlusion, signal loss, and multipath effects in complex highway traffic environments. By constructing an integrated positioning vector equation and introducing innovation vectors and robust factors, the model achieves adaptive suppression of measurement noise, thereby enhancing positioning accuracy. Field tests conducted at the Liuxia Hub on the Hangzhou Beltway Expressway, coupled with superimposed validation using 3D high-precision laser point cloud maps, demonstrate that this model can significantly improve positioning performance in complex traffic environments, with an elevation positioning error of less than 0.2 m, capable of supporting 3D lane-level navigation. This validates the application potential of BDS+5G hybrid positioning technology in the field of smart vehicles and autonomous driving.

Traditional inertial-satellite integrated navigation relies on satellite navigation to provide high-precision positioning results for position correction, which is difficult to adapt to the environment of strong suppression interference. To address this issue, this paper proposes an inertial navigation correction method based on interference direction-of-arrival (DOA) estimation, which can provide satellite-independent positioning correction information in strong jamming environments. The algorithm first estimates the interference direction-of-arrival (DOA) using compressed sensing-based direction finding, then combines this with the shortterm high-precision vehicle trajectory provided by inertial navigation to localize low-dynamic or stationary interference sources. Subsequently, it utilizes the estimated interference source positions and DOA information to inversely determine the aircraft’s position, ultimately providing the inertial navigation system with a satellite-independent correction reference. It is suitable for extreme suppression interference confrontation environment where satellite navigation cannot work for a long time. The main innovation is that a robust positioning method based on the position estimation of the interference source, which provides a extra positioning information source other than satellite navigation for inertial navigation in the application scenario where satellite navigation fails due to interference, and the corrected positioning information source does not need to add additional hardware sensors. The simulation results show that the algorithm can provide stable positioning for inertial navigation in the environment of four static strongly suppressed interference sources, and meet the navigation and positioning requirements in the environment of strong electromagnetic interference.

To evaluate the Doppler positioning performance of Low Earth Orbit (LEO) satellites, this paper analyzes the related errors and positioning performance of single-LEO navigation test satellite. Furthermore, the worldwide constellation Doppler positioning performance is analyzed in conjunction with low-orbit constellation simulation extrapolations. The results show that: ① The Doppler measurement error accuracy is at the decimetre level, which is more than one order of magnitude greater than the other error terms. Furthermore, the Doppler User Equivalent Range Rate Error (UERRE) accuracy of the comprehensive related error term is better than 0.27 m/s; ② The single-satellite Doppler positioning 3D error converges to 200 m in approximately eight minutes, with a post-convergence positioning accuracy of approximately 85 m. Furthermore, the single-star Doppler-equivalent PDOP eventually converges to around 200; ③ When the cut-off altitude angle is 10° or less and the cumulative observation time is 8 min or more, the global average of the constellation Doppler equivalent PDOP is superior to 28.8 m, the RMS is superior to 58.8 m, and better than 156.1 m on the 95%. Furthermore, the constellation Doppler positional accuracy (3D, 1σ) is superior to 7.8 m on average globally, and better than 15.9 m on the RMS, and better than 42.2 m on the 95%. The constellation Doppler-equivalent PDOP and positional accuracy are optimal at high latitudes, suboptimal at midlatitudes, and relatively poor at low latitudes.