Assimilation of ground-based GNSS data using a local ensemble Kalman filter.

Tropical cyclones become increasingly nonlinear and dynamically unstable in high-resolution models. The initial conditions are typically sub-optimal, leaving scope to improve the accuracy of forecasts with improved data assimilation. Simultaneously, the lack of real ground-based GNSS observations over the ocean poses significant challenges when evaluating the assimilation results in oceanic regions. In this study, an Observation System Simulation Experiment is carried out based on a tropical cyclone case. Assimilation experiments using the WRF-PDAF framework are conducted. Conventional and GNSS observation operators are implemented. A diverse array of synthetic observations, encompassing temperature (T), wind components (U and V), precipitable water (PW), and zenith total delay (ZTD), are assimilated utilizing the Local Error-Subspace Transform Kalman filter (LESTKF). The findings highlight the improvement in forecast accuracy achieved through the assimilation process over the ocean. Multiple observation types further improve the forecast accuracy. The study underscores the crucial role of GNSS data assimilation techniques. The assimilation of GNSS data presents potential for advancing weather forecasting capabilities. Thus, the construction of ground-based GNSS observation stations over the ocean is promising.

Study on the variation patterns of sea surface oil spill characteristics based on GNSS-R under different wind speeds.

Marine oil spills severely pollute marine environments, making rapid and accurate detection crucial. However, current Global Navigation Satellite System-Reflectometry (GNSS-R) oil spill detection studies overlook the impact of sea wind, leading to false or missed detections. This study simulates the mean square slope (MSS), scattering coefficient, coherent power, and delay-Doppler map (DDM), and explores the impact of sea wind on GNSS-R oil spill detection parameters. Results indicate that at wind speeds ≤3 m/s, the differences in MSS and scattering coefficient between oil and water are minimal, but become significant at speeds >3 m/s. Oil-covered sea surfaces exhibit unusually high coherent power and DDM in spill zones compared to clean surfaces, with both metrics decreasing as wind speed rises. Finally, using Cyclone Global Navigation Satellite System (CYGNSS) data, this study validated the reliability of GNSS-R oil spill parameter variation patterns derived from simulation data, providing references for oil spill detection research.

Extended Kalman filter-based robust roll angle estimation method for spinning vehicles.

Attitude measurement is a basic technique for monitoring vehicle motion states and safety. The spin motion of a vehicle couples the attitude angles with each other, which has an impact on the navigation and control of the vehicle. Global navigation satellite system (GNSS) signals-based roll angle measurement methods are important for vehicle attitude measurement. Most of existing studies use continuous signal power, but the case of loop lock loss leading to discontinuous power reception has not been considered. A robust estimation method for the roll angle based on the Tukey weight function is proposed to improve the measurement accuracy in cases of discontinuous reception. The characteristics of the GNSS signals, the geometric relationship between the signal power and roll angle of the vehicle are discussed. By installing a GNSS receiver with a single patched antenna on a rotating platform with a controllable rolling speed, the proposed method was verified by experiments. The robust estimation errors of different weight functions are analyzed. According to the characteristics of the gross measurement errors, a robust estimation method of multisatellite power observations is proposed to obtain a high-precision and stable estimation of the vehicle roll angle. The results show that the proposed algorithm can improve the accuracy of roll angle estimation even with gross measurement errors. As a result of the experiments, the estimation errors of the algorithm are 6.57° at a confidence level of 68 % and 15.49°at the confidence level of 95 %. In contrast, they are 11.38° and 37.31° for the traditional LS method. Moreover, the estimation accuracy of the algorithm is not significantly correlated with the vehicle rotational speed. Therefore, the vehicle roll angle can be estimated with high accuracy under a variety of rotational speeds.

Development of an IoT-Based Device for Data Collection on Sheep and Goat Herding in Silvopastoral Systems.

To evaluate the ecosystem services of silvopastoral systems through grazing activities, an advanced Internet of Things (IoT) framework is introduced for capturing extensive data on the spatial dynamics of sheep and goat grazing. The methodology employed an innovative IoT system, integrating a Global Navigation Satellite System (GNSS) tracker and environmental sensors mounted on the animals to accurately monitor the extent, intensity, and frequency of grazing. The experimental results demonstrated the high performance and robustness of the IoT system, with minimal data loss and significant battery efficiency, validating its suitability for long-term field evaluations. Long Range (LoRa) technology ensured consistent communication over long distances, covering the entire grazing zone and a range of 6 km in open areas. The superior battery performance, enhanced by a solar panel, allowed uninterrupted operation for up to 37 days with 5-min interval acquisitions. The GNSS module provided high-resolution data on movement patterns, with an accuracy of up to 10 m after firmware adjustments. The two-part division of the device ensured it did not rotate on the animals' necks. The system demonstrated adaptability and resilience in various terrains and animal conditions, confirming the viability of IoT-based systems for pasture monitoring and highlighting their potential to improve silvopastoral management, promoting sustainable practices and conservation strategies. This work uniquely focuses on documenting the shepherd's role in the ecosystem, providing a low-cost solution that distinguishes itself from commercial alternatives aimed primarily at real-time flock tracking.

A Hybrid Algorithm of LSTM and Factor Graph for Improving Combined GNSS/INS Positioning Accuracy during GNSS Interruptions.

In urban road environments, global navigation satellite system (GNSS) signals may be interrupted due to occlusion by buildings and obstacles, resulting in reduced accuracy and discontinuity of combined GNSS/inertial navigation system (INS) positioning. Improving the accuracy and robustness of combined GNSS/INS positioning systems for land vehicles in the presence of GNSS interruptions is a challenging task. The main objective of this paper is to develop a method for predicting GNSS information during GNSS outages based on a long short-term memory (LSTM) neural network to assist in factor graph-based combined GNSS/INS localization, which can provide a reliable combined localization solution during GNSS signal outages. In an environment with good GNSS signals, a factor graph fusion algorithm is used for data fusion of the combined positioning system, and an LSTM neural network prediction model is trained, and model parameters are determined using the INS velocity, inertial measurement unit (IMU) output, and GNSS position incremental data. In an environment with interrupted GNSS signals, the LSTM model is used to predict the GNSS positional increments and generate the pseudo-GNSS information and the solved results of INS for combined localization. In order to verify the performance and effectiveness of the proposed method, we conducted real-world road test experiments on land vehicles installed with GNSS receivers and inertial sensors. The experimental results show that, compared with the traditional combined GNSS/INS factor graph localization method, the proposed method can provide more accurate and robust localization results even in environments with frequent GNSS signal loss.

Design and Development of an Automatic Layout Algorithm for Laser GNSS RTK.

At the current stage, the automation level of GNSS RTK equipment is low, and manual operation leads to decreased accuracy and efficiency in setting out. To address these issues, this paper has designed an algorithm for automatic setting out that resolves the common problem of reduced accuracy in conventional RTK. First, the calculation of the laser rotation center is conducted using relevant parameters to calibrate the instrument's posture and angle. Then, by analyzing the posture information, the relative position and direction of the instrument to the point to be set out are determined, and the rotation angles in the horizontal and vertical directions are calculated. Following this, the data results are analyzed, and the obtained rotation angles are output to achieve automatic control of the instrument. Finally, a rotating laser composed of servo motors and laser modules is used to control the GNSS RTK equipment to locate the set-out point, thereby determining its position on the ground and displaying it in real-time. Compared to traditional GNSS RTK equipment, the proposed automatic setting out algorithm and the developed GNSS laser RTK equipment reduce the setting out error from 15 mm to 10.3 mm. This reduces the barrier to using GNSS RTK equipment, minimizes human influence, enhances the work efficiency of setting out measurements, and ensures high efficiency and stability under complex conditions.

The Real-Time Detection of Vertical Displacements by Low-Cost GNSS Receivers Using Precise Point Positioning.

Vertical displacements are traditionally measured with precise levelling, which is inherently time consuming. Rapid or even real-time height determination can be achieved by the Global Navigation Satellite System (GNSS). Nevertheless, the accuracy of real-time GNSS positioning is limited, and the deployment of a network of continuously operating GNSS receivers is not cost effective unless low-cost GNSS receivers are considered. In this study, we examined the use of geodetic-grade and low-cost GNSS receivers for static and real-time GNSS levelling, respectively. The results of static GNSS levelling were processed in four different software programs or services. The largest differences for ellipsoidal/normal heights reached 0.054 m/0.055 m, 0.046 m/0.047 m, and 0.058 m/0.058 m for points WRO1, BM_ROOF, and BM_CP, respectively. In addition, the values depended on the software used and the location of the point. However, the multistage experiment was designed to analyze various strategies for GNSS data processing and to define a method for detecting vertical displacement in a time series of receiver coordinates. The developed method combined time differentiation of coordinates estimated for a single GNSS receiver using the Precise Point Positioning (PPP) technique and Butterworth filtering. It demonstrated the capability of real-time detection of six out of eight displacements in the range between 20 and 55 mm at the three-sigma level. The study showed the potential of low-cost GNSS receivers for real-time displacement detection, thereby suggesting their applicability to structural health monitoring, positioning, or early warning systems.

GNSS Receiver Fingerprinting Based on Time Skew of Embedded CSAC Clock.

GNSS spoofing has become a significant security vulnerability threatening remote sensing systems. Hardware fingerprint-based GNSS receiver identification is one of the solutions to address this security issue. However, existing research has not provided a solution for distinguishing GNSS receivers of the same specification. This paper first theoretically proves that the CSACs (Chip-Scale Atomic Clocks) used in GNSS receivers have unique hardware noise and then proposes a fingerprinting scheme based on this hardware noise. Experiments based on the neural network method demonstrate that this fingerprint achieved an identification accuracy of 94.60% for commercial GNSS receivers of the same specification and performed excellently in anomaly detection, confirming the robustness of the fingerprinting method. This method shows a new real-time GNSS security monitoring method based on CSACs and can be easily used with any commercial GNSS receivers.

Monitoring Surface Deformations in a Fossil Landslide Zone and Identifying Potential Failure Mechanisms: A Case Study of Gümüshane State Hospital.

The escalating occurrence of landslides has drawn increasing attention from the scientific community, primarily driven by a combination of natural phenomena such as unpredictable seismic events, intensified precipitation, and rapid snowmelt attributable to climate fluctuations, compounded by inadequacies in engineering practices during site selection. Within the scope of this investigation, contemporary geodetic techniques using the GNSS were employed to monitor structural and surface deformations in and around a hospital edifice situated within an ancient fossil landslide region. Additionally, inclinometer measurements facilitated the determination of slip circle parameters. A subsequent analysis integrated these datasets to scrutinize both the hospital structure and its surrounding slopes. In addition to the finite element method, four different limit equilibrium methods (Bishop, GLE-Morgenstern-Price, Spencer, and Janbu) were used in the evaluation of stability. Since the safety number determined in all analyses was <1, it was determined that the slope containing the hospital building was unstable. The movement has occurred again due to the additional load created by the hospital building built on the currently stable slope, the effect of surface and groundwater, and the improperly designed road route. As a result of geodetic monitoring, it was determined that the sliding speed on the surface was in the N-E direction and was approximately 3 cm, and this situation almost coincided with inclinometer measurements.

An improved Gaussian process for filling the missing data in GNSS position time series considering the influence of adjacent stations.

Due to various unavoidable reasons or gross error elimination, missing data inevitably exist in global navigation satellite system (GNSS) position time series, which may result in many analysis methods not being applicable. Typically, interpolating the missing data is a crucial preprocessing step before analyzing the time series. The conventional methods for filling missing data do not consider the influence of adjacent stations. In this work, an improved Gaussian process (GP) approach is developed to fill the missing data of GNSS time series, in which the time series of adjacent stations are applied to construct impact factors, together with a comparison of the conventional GP and the commonly used cubic spline methods. For the simulation experiments, the root mean square error (RMSE), mean absolute error (MAE) and correlation coefficient (R) are adopted to evaluate the performance of the improved GP. The results show that the filled missing data of the improved GP are closer to the true values than those of the conventional GP and cubic spline methods, regardless of the missing percentages ranging from 5 to 30%, with an interval of 5%. Specifically, the mean relative RMSE and MAE improvements for the improved GP with respect to the conventional GP are 21.2%, 21.3% and 8.3% and 12.7%, 16.2% and 11.01% for the North (N), East (E) and Up (U) components, respectively. In the real experiment, eight GNSS stations are analyzed using improved GP, together with conventional GP and a cubic spline. The results indicate that the first three principal components (PCs) of the improved GP can perverse 98.3%, 99.8% and 77.0% of the total variance for the N, E and U components, respectively. This value is obviously higher than those of the conventional GP and cubic spline. Therefore, we can conclude that the improved GP can better fill in the missing data in GNSS position time series than the conventional GP and cubic spline because of the impacts of adjacent stations.

Wearable Multi-Sensor Positioning Prototype for Rowing Technique Evaluation.

The goal of this study is to determine the feasibility of a wearable multi-sensor positioning prototype to be used as a training tool to evaluate rowing technique and to determine the positioning accuracy using multiple mathematical models and estimation methods. The wearable device consists of an inertial measurement unit (IMU), an ultra-wideband (UWB) transceiver, and a global navigation satellite system (GNSS) receiver. An experiment on a rowing shell was conducted to evaluate the performance of the system on a rower's wrist, against a centimeter-level GNSS reference trajectory. This experiment analyzed the rowing motion in multiple navigation frames and with various positioning methods. The results show that the wearable device prototype is a viable option for rowing technique analysis; the system was able to provide the position, velocity, and attitude of a rower's wrist, with a positioning accuracy ranging between ±0.185 m and ±1.656 m depending on the estimation method.

Precise Positioning in Nitrogen Fertility Sensing in Maize (Zea mays L.).

This study documented the contribution of precise positioning involving a global navigation satellite system (GNSS) and a real-time kinematic (RTK) system in unmanned aerial vehicle (UAV) photogrammetry, particularly for establishing the coordinate data of ground control points (GCPs). Without augmentation, GNSS positioning solutions are inaccurate and pose a high degree of uncertainty if such data are used in UAV data processing for mapping. The evaluation included a comparative assessment of sample coordinates involving RTK and an ordinary GPS device and the application of precise GCP data for UAV photogrammetry in field crop research, monitoring nitrogen deficiency stress in maize. This study confirmed the superior performance of the RTK system in providing positional data, with 4 cm bias as compared to 311 cm with the non-augmented GNSS technique, making it suitable for use in agronomic research involving row crops. Precise GCP data in this study allow the UAV-based Normalized Difference Red-Edge Index (NDRE) data to effectively characterize maize crop responses to N nutrition during the growing season, with detailed analyses revealing the causal relationship in that a compromised optimum canopy chlorophyll content under limiting nitrogen environment was the reason for reduced canopy cover under an N-deficiency environment. Without RTK-based GCPs, different and, to some degree, misleading results were evident, and therefore, this study warrants the requirement of precise GCP data for scientific research investigations attempting to use UAV photogrammetry for agronomic field crop study.

Recent Advances on Jamming and Spoofing Detection in GNSS.

Increased interest in the development and integration of navigation and positioning services into a wide range of receivers makes them susceptible to a variety of security attacks such as Global Navigation Satellite Systems (GNSS) jamming and spoofing attacks. The availability of low-cost devices including software-defined radios (SDRs) provides a wide accessibility of affordable platforms that can be used to perform these attacks. Early detection of jamming and spoofing interferences is essential for mitigation and avoidance of service degradation. For these reasons, the development of efficient detection methods has become an important research topic and a number of effective methods has been reported in the literature. This survey offers the reader a comprehensive and systematic review of methods for detection of GNSS jamming and spoofing interferences. The categorization and classification of selected methods according to specific parameters and features is provided with a focus on recent advances in the field. Although many different detection methods have been reported, significant research efforts toward developing new and more efficient methods remain ongoing. These efforts are driven by the rapid development and increased number of attacks that pose high-security risks. The presented review of GNSS jamming and spoofing detection methods may be used for the selection of the most appropriate solution for specific purposes and constraints and also to provide a reference for future research.

A Study on Graph Optimization Method for GNSS/IMU Integrated Navigation System Based on Virtual Constraints.

In GNSS/IMU integrated navigation systems, factors like satellite occlusion and non-line-of-sight can degrade satellite positioning accuracy, thereby impacting overall navigation system results. To tackle this challenge and leverage historical pseudorange information effectively, this paper proposes a graph optimization-based GNSS/IMU model with virtual constraints. These virtual constraints in the graph model are derived from the satellite's position from the previous time step, the rate of change of pseudoranges, and ephemeris data. This virtual constraint serves as an alternative solution for individual satellites in cases of signal anomalies, thereby ensuring the integrity and continuity of the graph optimization model. Additionally, this paper conducts an analysis of the graph optimization model based on these virtual constraints, comparing it with traditional graph models of GNSS/IMU and SLAM. The marginalization of the graph model involving virtual constraints is analyzed next. The experiment was conducted on a set of real-world data, and the results of the proposed method were compared with tightly coupled Kalman filtering and the original graph optimization method. In instantaneous performance testing, the method maintains an RMSE error within 5% compared with real pseudorange measurement, while in a continuous performance testing scenario with no available GNSS signal, the method shows approximately a 30% improvement in horizontal RMSE accuracy over the traditional graph optimization method during a 10-second period. This demonstrates the method's potential for practical applications.

GNSS interference and spoofing dataset.

GNSS signals are vulnerable to spoofing and interference, which poses a threat to the security of critical national infrastructure. GNSS data sets with spoofing and jamming are lacking, which hinders the research of GNSS anti-spoofing and anti-interference techniques. This data article presents a dataset recorded by a low-cost sensor deployed on the balcony at the 5th floor of the Science Hall of Yunnan University (25°3'26'' N, 102°41'55'' E). The sensor suite includes a GNSS antenna, a u-blox GNSS receiver and an embedded computer. In the experiment, interferences including spoofing and jamming were irregularly emitted using a SDR HackRF One and a commercial jammer, respectively. The dataset collected by the receiver consists of two parts: (1) raw data; (2) processed data. The types of the raw data include hardware information, satellite information and receiver parameters of GPS, Campass, Galileo, GLONASS and QZSS systems. The processed data are extracted from the raw data, including the signals, Doppler shift, pseudorange observations, carrier phase, position (latitude, longitude, and altitude), satellite azimuth and elevation angles, etc. The provided datasets are interesting for the GNSS security, anti-jamming and anti-spoofing mechanisms based scientific communities.

UAV-UGV Collaborative Localisation with Minimum Sensing.

This paper presents a novel methodology to localise Unmanned Ground Vehicles (UGVs) using Unmanned Aerial Vehicles (UAVs). The UGVs are assumed to be operating in a Global Navigation Satellite System (GNSS)-denied environment. The localisation of the ground vehicles is achieved using UAVs that have full access to the GNSS. The UAVs use range sensors to localise the UGV. One of the major requirements is to use the minimum number of UAVs, which is two UAVs in this paper. Using only two UAVs leads to a significant complication that results an estimation unobservability under certain circumstances. As a solution to the unobservability problem, the main contribution of this paper is to present a methodology to treat the unobservability problem. A Constrained Extended Kalman Filter (CEKF)-based solution, which uses novel kinematics and heuristics-based constraints, is presented. The proposed methodology has been assessed based on the stochastic observability using the Posterior Cramér-Rao Bound (PCRB), and the results demonstrate the successful operation of the proposed localisation method.

Validity of Current Smartwatches for Triathlon Training: How Accurate Are Heart Rate, Distance, and Swimming Readings?

Smartwatches are one of the most relevant fitness trends of the past two decades, and they collect increasing amounts of health and movement data. The accuracy of these data may be questionable and requires further investigation. Therefore, the aim of the present study is to validate smartwatches for use in triathlon training. Ten different smartwatches were tested for accuracy in measuring heart rates, distances (via global navigation satellite systems, GNSSs), swim stroke rates and the number of swim laps in a 50 m Olympic-size pool. The optical heart rate measurement function of each smartwatch was compared to that of a chest strap. Thirty participants (15 females, 15 males) ran five 3 min intervals on a motorised treadmill to evaluate the accuracy of the heart rate measurements. Moreover, for each smartwatch, running and cycling distance tracking was tested over six runs of 4000 m on a 400 m tartan stadium track, six hilly outdoor runs over 3.4 km, and four repetitions of a 36.8 km road bike course, respectively. Three swimming protocols ranging from 200 m to 400 m were performed in triplicate in a 50 m Olympic-size pool, evaluating the tracked distance and the detected number of strokes. The mean absolute percentage errors (MAPEs) for the average heart rate measurements varied between 3.1% and 8.3%, with the coefficient of determination ranging from 0.22 to 0.79. MAPE results ranged from 0.8% to 12.1% for the 4000 m run on the 400 m track, from 0.2% to 7.5% for the 3.4 km outdoor run, and from 0.0% to 4.2% for the 36.8 km bike ride. For the swimming tests, in contrast, the deviations from the true distance varied greatly, starting at a 0.0% MAPE for the 400 m freestyle and reaching 91.7% for the 200 m medley with style changes every 25 m. In summary, for some of the smartwatches, the measurement results deviated substantially from the true values. Measurements taken while road cycling over longer distances with only a few curves were in relative terms more accurate than those taken during outdoor runs and even more accurate than those taken on the 400 m track. In the swimming exercises, the accuracy of the measured distances was severely deteriorated by the medley changes among the majority of the smartwatches. Altogether, the results of this study should help in assessing the accuracy and thus the suitability of smartwatches for general triathlon training.

Distributed Resources Allocation Method for Space-Ground Integrated Mobile Communication System.

This paper presents an innovative approach towards space-ground integrated communication systems by combining terrestrial cellular networks, UAV networks, and satellite networks, leveraging advanced slicing technology. The proposed architecture addresses the challenges posed by future user surges and aims to reduce network overhead effectively. Central to our approach is the introduction of a marginal mobile station (MS)-assisted network resource allocation decision architecture. Building upon this foundation, we introduce the DP-DQN model, an enhanced decision-making algorithm tailored for MSs in dynamic network environments. Furthermore, this study introduces a feedback mechanism to ensure the accuracy and adaptability of the marginalization model over time. Through extensive simulations and experimental validations, our DP-DQN-based edge decision method demonstrates substantial potential in alleviating core network overhead while improving success access ratios compared to conventional methods.

An Enhanced FGI-GSRx Software-Defined Receiver for the Execution of Long Datasets.

The Global Navigation Satellite System (GNSS) software-defined receivers offer greater flexibility, cost-effectiveness, customization, and integration capabilities compared to traditional hardware-based receivers, making them essential for a wide range of applications. The continuous evolution of GNSS research and the availability of new features require these software-defined receivers to upgrade continuously to facilitate the latest requirements. The Finnish Geospatial Research Institute (FGI) has been supporting the GNSS research community with its open-source implementations, such as a MATLAB-based GNSS software-defined receiver `FGI-GSRx' and a Python-based implementation `FGI-OSNMA' for utilizing Galileo's Open Service Navigation Message Authentication (OSNMA). In this context, longer datasets are crucial for GNSS software-defined receivers to support adaptation, optimization, and facilitate testing to investigate and develop future-proof receiver capabilities. In this paper, we present an updated version of FGI-GSRx, namely, FGI-GSRx-v2.0.0, which is also available as an open-source resource for the research community. FGI-GSRx-v2.0.0 offers improved performance as compared to its previous version, especially for the execution of long datasets. This is carried out by optimizing the receiver's functionality and offering a newly added parallel processing feature to ensure faster capabilities to process the raw GNSS data. This paper also presents an analysis of some key design aspects of previous and current versions of FGI-GSRx for a better insight into the receiver's functionalities. The results show that FGI-GSRx-v2.0.0 offers about a 40% run time execution improvement over FGI-GSRx-v1.0.0 in the case of the sequential processing mode and about a 59% improvement in the case of the parallel processing mode, with 17 GNSS satellites from GPS and Galileo. In addition, an attempt is made to execute v2.0.0 with MATLAB's own parallel computing toolbox. A detailed performance comparison reveals an improvement of about 43% in execution time over the v2.0.0 parallel processing mode for the same GNSS scenario.

A new Egyptian Grid Weighted Mean Temperature (EGWMT) model using hourly ERA5 reanalysis data in GNSS PWV retrieval.

Precise modeling of weighted mean temperature (Tm) is essential for Global Navigation Satellite System (GNSS) meteorology. In retrieving precipitable water vapor (PWV) from GNSS, Tm is a crucial parameter for the conversion of zenith wet delay (ZWD) into PWV. In this study, an improved Tm model, named EGWMT, was developed to accurately estimate Tm at any site in Egypt. This new model was established using hourly ERA5 reanalysis data from European Centre for Medium-Range Weather Forecasts (ECMWF) covering the period from 2008 to 2019 with a spatial resolution of 0.25° × 0.25°. The performance of the proposed model was evaluated using two types of data sources, including hourly ERA5 reanalysis data from 2019 to 2022 and radiosonde profiles over a six-year period from 2017 to 2022. The accuracy of the EGWMT model was compared to that of four other models: Bevis, Elhaty, ANN and GGTm-Ts using two statistical quantities, including mean absolute bias (MAB) and root mean square error (RMSE). The results demonstrated that the EGWMT model outperformed the Bevis, Elhaty, ANN and GGTm-Ts models with RMSE improvements of 32.5%, 30.8%, 39% and 48.2%, respectively in the ERA5 data comparison. In comparison with radiosonde data, the EGWMT model achieved RMSE improvements of 22.5%, 34%, 38% and 19.5% against Bevis, Elhaty, ANN and GGTm-Ts models, respectively. In order to determine the significance of differences in means and variances, statistical tests, including t-test and F-test, were conducted. The results confirmed that there were significant differences between the EGWMT model and the four other models.