The multi-constellation Global Navigation Satellite Systems (GNSS) offers promising potential for the retrieval of real-time (RT) atmospheric parameters to support time-critical meteorological applications, such as nowcasting or regional short-term forecasts. In this study, we processed GNSS data from the globally distributed Multi-GNSS Experiment (MGEX) network of about 30 ground stations by using the precise point positioning (PPP) technique for retrieving RT multi-GNSS tropospheric delays. RT satellite orbit and clock product streams from the International GNSS Service (IGS) were used. Meanwhile, we assessed the quality of clock and orbit products provided by different IGS RT services, called CLK01, CLK81, CLK92, GFZC2, and GFZD2, respectively. Using the RT orbit and clock products, the performances of the RT zenith total delays (ZTD) retrieved from single-system as well as from multi-GNSS combined observations were evaluated by comparing with the U.S. Naval Observatory (USNO) final troposphere products. With the addition of multi-GNSS observations, RT ZTD estimates with higher accuracy and enhanced reliability compared to the single-system solution can be obtained. Compared with the Global Positioning System (GPS)-only solution, the improvements in the initialization time of ZTD estimates are about 5.8% and 8.1% with the dual-system and the four-system combinations, respectively. The RT ZTD estimates retrieved with the GFZC2 products outperform those derived from the other IGS-RT products. In the GFZC2 solution, the accuracy of about 5.05 mm for the RT estimated ZTD can be achieved with fixing station coordinates. The results also confirm that the accuracy improvement (about 22.2%) can be achieved for the real-time estimated ZTDs by using multi-GNSS observables, compared to the GPS-only solution. In the multi-GNSS solution, the accuracy of real-time retrieved ZTDs can be improved by a factor of up to 2.7 in the fixing coordinate mode, compared with that in the kinematic mode.

The new generations of global navigation satellite system (GNSS) space vehicles can transmit three or more frequency signals. Multi-frequency observations bring a significant improvement to precise point positioning ambiguity resolution (PPP AR). However, the multi-frequency satellite code and phase biases need to be properly handled before conducting PPP AR. The traditional satellite bias correction methods, for example, the commonly used differential code biases (DCB), are limited to the dual-frequency ionosphere-free (IF) case and become more and more difficult to extend to multi-GNSS and multi-frequency cases. In this contribution, we propose the observable-specific signal bias (OSB) correction method for un-differenced and uncombined (UDUC) PPP AR. The OSB correction method, which includes observable-specific satellite code and phase bias correction, can directly apply kinds of OSBs to GNSS original observation data, thus, it is more appropriate for multi-GNSS and multi-frequency cases. In order to verify the performance of multi-frequency UDUC-PPP AR based on the OSB correction method, triple-frequency GPS observation data provided by 142 Multi-GNSS Experiment (MGEX) stations were used to estimate observable-specific satellite phase biases at the PPP service end and some of them were also used to conduct AR at the PPP user end. The experiment results showed: the averaged time-to-first-fix (TTFF) of triple-frequency GPS kinematic UDUC-PPP AR with observable-specific satellite code bias (SCB) corrections could reach about 22 min with about 29% improvement, compared with that without observable-specific SCB corrections; TTFF of triple-frequency static UDUC-PPP AR with observable-specific phase-specific time-variant inter-frequency clock bias (IFCB) corrections took about 15.6 min with about 64.3% improvement, compared with that without observable-specific IFCB corrections.

Compared with the traditional ionospheric-free linear combination precise point positioning (PPP) model, the un-differenced and uncombined (UDUC) PPP model using original observations can keep all the information of the observations and be easily extended to any number of frequencies. However, the current studies about the multi-frequency UDUC-PPP ambiguity resolution (AR) were mainly based on the triple-frequency BeiDou navigation satellite system (BDS) observations or simulated data. Limited by many factors, for example the accuracy of BDS precise orbit and clock products, the advantages of triple-frequency signals to UDUC-PPP AR were not fully exploited. As Galileo constellations have been upgraded by increasing the number of 19 useable satellites, it makes using Galileo satellites to further study the triple-frequency UDUC-PPP ambiguity resolution (AR) possible. In this contribution, we proposed the method of multi-frequency step-by-step ambiguity resolution based on the UDUC-PPP model and gave the reason why the performance of PPP AR can be improved using triple-frequency observations. We used triple-frequency Galileo observations on day of year (DOY) 201, 2018 provided by 166 Multi-GNSS Experiment (MGEX) stations to estimate original uncalibrated phase delays (UPD) on each frequency and to conduct both dual- and triple-frequency UDUC-PPP AR. The performance of UDUC-PPP AR based on post-processing mode was assessed in terms of the time-to-first-fix (TTFF) as well as positioning accuracy with 2-h observations. It was found that triple-frequency observations were helpful to reduce TTFF and improve the positioning accuracy. The current statistic results showed that triple-frequency PPP-AR reduced the averaged TTFF by 19.6% and also improved the positioning accuracy by 40.9%, 31.2% and 23.6% in the east, north and up directions respectively, compared with dual-frequency PPP-AR. With an increasing number of Galileo satellites, it is expected that the robustness and accuracy of the triple-frequency UCUD-PPP AR can be improved further.

In the construction of super high-rise buildings, vibration monitoring of climbing scaffolding is crucial for ensuring construction safety. This study proposes a vibration state prediction model based on Variational Mode Decomposition (VMD) and Temporal Convolutional Network (TCN), referred to as the VMD-TCN model. Using the construction of the Tianjin Zhonghai City Plaza super high-rise building as a case study, this model was applied to 48 h of climbing scaffolding vibration data for modeling and prediction. The results demonstrate that VMD significantly enhances the multi-band feature extraction capability of vibration signals. Compared to predictions using raw, undecomposed signals, the VMD-TCN model reduces the root mean square error (RMSE) by 43.9%, 43.2%, and 34.7% for 1 min, 3 min, and 5 min prediction tasks, respectively, while improving the coefficient of determination (R²) by 21.0%, 33.0%, and 37.6%. Furthermore, the computational efficiency of the VMD-TCN model surpasses that of the VMD-GRU model by approximately 88–91%, making it well-suited for engineering applications with high real-time requirements. Additionally, the VMD-TCN model maintains high predictive accuracy across different sensor placements and data collection periods, demonstrating strong generalization capabilities. The findings of this study provide scientific support for intelligent monitoring and safety early warning of climbing scaffolding, contributing to improved safety and management efficiency in super high-rise building construction.

Long-term and accurate prediction of air pollutant concentrations can serve as a foundation for air pollution warning and prevention, which is crucial for social development and human health. In this study, we provide a model for predicting the concentration of air pollutants based on big data spatiotemporal correlation analysis and deep learning methods. Based on an improved temporal convolutional network (TCN) and a bi-directional gated recurrent unit (BiGRU) as the fundamental architecture, the model adds two attention mechanisms to improve performance: Squeeze and Excitation Networks (SENet) and Convolutional Block Attention Module (CBAM). The improved TCN moves the residual connection layer to the network’s front end as a preprocessing procedure, improving the model’s performance and operating efficiency, particularly for big data jobs like air pollution concentration prediction. The use of SENet improves the model’s comprehension and extraction of long-term dependent features from pollutants and meteorological data. The incorporation of CBAM enhances the model’s perception ability towards key local regions through an attention mechanism in the spatial dimension of the feature map. The TCN-SENet-BiGRU-CBAM model successfully realizes the prediction of air pollutant concentrations by extracting the spatiotemporal features of the data. Compared with previous advanced deep learning models, the model has higher prediction accuracy and generalization ability. The model is suitable for prediction tasks from 1 to 12 h in the future, with root mean square error (RMSE) and mean absolute error (MAE) ranging from 5.309~14.043 and 3.507~9.200, respectively.