Abstract The Indian Regional Navigation Satellite System (IRNSS), or NavIC, provides regional satellite positioning services across India and parts of Southeast Asia. In this study, we develop and evaluate a software-defined receiver (SDR) enhanced with deep learning techniques to acquire the IRNSS Standard Positioning Service (SPS) L5-band signal. The SDR architecture incorporates data-driven improvements in acquisition decision-making while retaining compatibility with the IRNSS signal structure as specified in the official ICD. Field experiments were conducted in Hanoi, Vietnam, a location situated at the fringe of NavIC’s primary service area. Signal data were collected using a low-cost RF front-end connected to a rooftop-mounted antenna. Experimental results demonstrate that the proposed SDR is capable of reliably acquiring and tracking up to four IRNSS satellites under nominal conditions. The average C/N0 ranged from 30 to 42 dB-Hz, and successful position solutions were obtained with a horizontal accuracy of approximately 25 meters. Additionally, the deep learning-based acquisition module improved robustness in low-SNR scenarios. This work represents the first implementation of a learning-aided IRNSS receiver validated in Vietnam and offers insights into extending NavIC-based positioning services to broader Southeast Asian regions.

Abstract This paper proposes a novel framework for detecting and mitigating multipath interference in Global Navigation Satellite Systems (GNSS), a common issue caused by reflections of signal off surfaces such as buildings and the ground. To address this challenge, the study integrates Differential GNSS (DGNSS) techniques with advanced machine learning models, such as Support Vector Machine (SVM), Convolutional Neural Network (CNN), and Random Forest—to automatically detect and exclude satellites affected by multipath. The methodology involves synchronized GNSS data collection from a stationary base station and a mobile rover using high-precision u-blox ZED-F9P receivers with polarized antennas, DGNSS corrections via single and double differencing, and feature vector construction from both corrected and raw observation data. Signal quality labels (“Clean” or “Noise”) are derived through skyplot analysis and environmental modeling. Three classification approaches are explored: direct classification using DGNSS-derived vectors (Approach 1), image-based classification (Approach 2), and classification using combined feature vectors (Approach 3). Experimental evaluation using 2,312 labeled samples shows that ensemble learning significantly outperforms single-model classifiers for multipath detection. Random Forest achieves the highest performance across all approaches, reaching up to 99.75% accuracy, while CNN outperforms traditional methods, reaching up to 86.77% accuracy in image-based classification in Approach 2. The findings demonstrate the effectiveness of the framework in identifying and excluding compromised satellite signals, which has the potential to enhance the accuracy of GNSS positioning, with potential for real-world application in complex urban environments.

Abstract Variations in the Earth’s rotation are intimately linked to geodynamical processes. Investigating secular change in the Earth’s rotation can provide a profound understanding of the Earth system. Currently, continuous sequence of the Earth’s variable rotation spans less than 400 years. Some researchers have employed historical records of ancient astronomical phenomena, particularly observations of solar and lunar eclipses, to study the ΔT value, which characterizes the secular change in the Earth’s rotation. This study examines the corresponding ΔT value by combining total solar eclipse observations from Chinese historical documents on June 25, A.D. 1275.

Abstract NRLMSISE is an empirical model that allows us to predict temperatures and densities of the main atmospheric components. The model is widely used to evaluate atmospheric impacts on satellite orbits and laser beam refraction which come through the atmosphere, such as those used for Earth-satellite distance measurements. Model of the atmosphere is a valuable part of the Satellite Laser Ranging processing software like Kyiv Geodynamics (Juliette). Juliette is written in C++ and exploits the C++ clone of NRLMSISE written by the second author. The C++ version produces the same outputs as an official Fortran code. Accurate modeling of atmospheric influences on satellite motion requires performing numerous calculations along satellite orbits or laser beam paths, which are computationally intensive. By decreasing calculation time of NRLMSISE, we would not only save the modeling time but also give a prospect for a wider application of the model due to lowering computational resource demands. Our work demonstrates how the traditional NRLMSISE model can be effectively translated into a neural network. This conversion achieves significant performance gains on both CPU and GPU while maintaining acceptable accuracy when compared to the C++ implementation of NRLMSISE. We demonstrate the process of moving NRLMSISE to a neural network, the resulting accuracy, ease of running the trained model on CUDA-enabled GPUs, and the obtained boost of performance on both CPU and GPU.

Abstract Disturbances in the ionospheric plasma due to geomagnetic storms and thunderstorms recorded by the DEMETER and Swarm satellites are presented. Lightning and, in particular, transient luminous events (sprites, jets, elves and halos) are associated with the electromagnetic links and interactions between the atmosphere, ionosphere and magnetosphere and with intense thunderstorm activity. DEMETER has clearly shown that thunderstorms and sprites can affect the ionosphere even at its orbital altitude (680 km). Additional information on ionospheric disturbances comes from data collected by the Swarm satellites. The Swarm constellation consists of three identical satellites. Two of them operate in circular polar orbits with an initial altitude of 460 km, while the third satellite has a circular orbit but an altitude of 530 km. The orbits of the first two satellites are almost in the same plane, but the third satellite is almost perpendicular to the first two. The payload, which includes a vector field magnetometer, an absolute scalar magnetometer and an electric field instrument, will make it possible to study the effects of thunderstorms on the ionosphere. Registrations of ionospheric disturbances collected by DEMETER and Swarm during selected strong geomagnetic storms and thunderstorms over Poland and Central Africa are presented, and the similarities and differences are discussed.

ABSTRACT The space plasma has relatively low energy but is dense in the low Earth orbit (LEO). In this study, we prepared various samples of anodic alloy surfaces with coating thicknesses of 20, 25, 35, and 45 μm to identify the most suitable characteristics for space applications. Ground-based tests were conducted at the Laboratory of Lean Satellite Enterprises and In-Orbit Experiments at the Kyushu Institute of Technology. A radio frequency (RF) plasma source was used to generate a simulated LEO plasma using argon gas in a vacuum chamber. The plasma properties were measured with a Langmuir probe under different test conditions. A negatively biased voltage of –450 V was applied to the samples to study charging/discharging phenomena. The samples were exposed to Ar-plasma for 1–2 h. The physical properties and structural morphology of various alloy samples were analyzed before and after exposure to plasma. This analysis involved ultraviolet (UV)/visible (Vis)/ Near-Infrared (NIR) absorption spectra, Energy Dispersive X-ray (EDX) analysis, and surface roughness testing. The results showed that space plasma notably impacts the physical properties and morphology of the alloys. A coated thickness of around 25 μm is considered more suitable for spacecraft surface structures due to its improved optical stability and resistance to plasma degradation, as indicated by the experimental results.

ABSTRACT The advances in Machine Learning (ML) and computer technologies enabled to process satellite images using programming. Environmental applications that handle Remote Sensing (RS) data for spatial analysis use such an approach, for example, Python’s library scikit-learn using algorithms on pattern identification, predictions or image classification. This paper presents an ML method of satellite image processing using Geographic Resources Analysis Support System (GRASS) Geographic Information System (GIS). The aim is to classify multispectral Landsat images using ML for identification of changes in salt pans of West Mauritania, Africa over the period 2014–2023. We define 10 classes of land cover categories and perform analysis of geological, lithological and landscape setting, and then introduce the principles, algorithms and processing of the ML methods of GRASS GIS. The following classification models were employed to implement image classification with training: Random Forest (RF), Decision Tree, Gradient Boosting and Support Vector Machine (SVM). The results were compared with clustering performed by k-means and maximum likelihood discriminant analysis. The cartographic visualisation and validation was implemented through accuracy analysis. Results for the best performing SVM model with seven-band input produced an overall accuracy of 76%, for the RF model – 73%, compared to 69% for Decision Tree Classifier – 69% and for Gradient Boosting Classifier – 67%. The SVM model embedded in GRASS GIS generates robust land cover maps with good accuracy from multispectral satellite images. The paper demonstrated an ML-based automated approach to satellite image processing, which links Artificial Intelligence (AI) with cartographic tasks.

ABSTRACT The orbits of Global Navigation Satellite System (GLONASS) satellites are computed from the broadcast ephemerides using the fourth order of the Runge–Kutta integration method. Usually, the initial conditions used in the integration of the differential equation of satellite motion are the three positions and the three velocities of satellites at the initial time, and the results are the position and velocity at a given time; the luni-solar perturbation is supposed to be constant during the integration interval. The algorithm used is known in the documentation as the simplified algorithm; this algorithm was modified and replaced by the one called in this investigation as the simplified–modified algorithm, where the luni-solar accelerations were taken as variable terms and three linear functions modeling these luni-solar accelerations were added to the simplified algorithm. The ode45 MATLAB solver, based on the fourth and fifth orders of the Runge–Kutta method, was used to solve the differential equations describing the motion of GLONASS satellites in orbit. The data used in this study is the broadcast orbit files of 24 GLONASS satellites between March 1 and 21, 2024. The results obtained showed an improvement of 1.76 m and 0.0027 m/s in the positions and velocities of GLONASS satellites, respectively, when the simplified–modified algorithm was applied, that is, the three luni-solar accelerations were assumed as variable terms.

ABSTRACT The copper belt of Anti-Atlas is recognized with several mineral occurrences of Cu, Zn, Mn, Ag, Au, and iron. We used ASTER and OLI in lithological and mineral detection and mapping. The lithological mapping was performed using principal components analysis (PCA), minimum noise fraction (MNF), and two classifiers: maximum likelihood (ML) and support vector machine (SVM). The hydrothermally altered zones were detected based on ASTER VNIR/SWIR bands by the integration of Ninomiya indices and constrained energy minimization (CEM) algorithm. In our study area, the enhanced band combinations of ASTER MNF1, PC4, and PC2 and OLI MNF1, PC5, and PC3 were applied for lithological discrimination. The OLI and ML classification shows the best lithological mapping accuracy with an overall accuracy of 91.74% and a 0.90 Kappa coefficient, followed by SVM with an overall accuracy of 88.82% and a 0.86 Kappa coefficient using the same sensor. The hydrothermal alteration mapping reveals alunite, chlorite, calcite, epidote, illite, kaolinite, montmorillonite, muscovite, and pyrophyllite minerals, principally in phyllic and argillic altered areas. The adopted methodology for lithological and mineralogical mapping can be used in other regions with similar criteria to the study area.

Abstract On the dwarf planet Ceres, there are bright spots known as faculae. Four types of faculae are distinguished: (a) floor faculae, (b) faculae on Ahuna Mons, (c) rim/wall faculae found on craters' rims or walls, and (d) ejecta faculae in the form of bright ejecta blankets. Our investigation on the interaction of the hypothesized subsurface originated jet of gas and the granular material indicated that floor faculae (a) could be a result of separation of fine bright component of regolith. Here, we consider the hypothesis that the ejecta faculae (d) may be the result of separation of grains due to explosive gas expansion during the formation of the impact crater. We consider the axisymmetric gas motion above the surface of Ceres. We transform our system of equations into a dimensionless form. Our numerical model indicates that the separation effect is strong enough to separate the grains (according to size, density, and other aerodynamics properties). In some cases, separation gives a monotonic, systematic effect: smaller particles are ejected farther than heavier particles. Generally, however, the distance over which the particles will be thrown depends in a rather complicated way on the parameters of the particles and the parameters of motion of the gas. This property fits the faculae of type (d). Because we used the dimensionless form of the equations, our results can be also applied to other celestial bodies where the regolith contains volatile substances. This paper is an extension of our investigations on the origin of faculae (a).