KLKaihui Lv

Kaihui Lv

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Kaihui Lv has a total of 9 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A Zenith Tropospheric Delay Prediction Model Based on VMD-LSTM Neural Network

    by Kaihui Lv, Chenglin Cai, Yihao Cai, Yun Fang, Zexian Li, Lingfeng Cheng, Wenhui Guan

    Zenith Tropospheric Delay (ZTD) is one of the primary error sources impacting the accuracy of Global Navigation Satellite System (GNSS) Precise Point Positioning (PPP) and Very Long Baseline Interferometry (VLBI). Enhancing the accuracy of ZTD prediction is critical for improving navigation, positioning, and VLBI precision. This study introduces an innovative tropospheric prediction model that integrates Variational Mode Decomposition (VMD) with a Long Short-Term Memory (LSTM) neural network to improve ZTD prediction accuracy and long-term stability. The experimental data used in this study were obtained from the Vienna Mapping Function 3 Optimized (VMF3_OP) ZTD dataset. VMD was employed to decompose the ZTD into periodic and residual components, allowing for deeper exploration of its informational characteristics. Each component is then individually used to train an enhanced LSTM neural network model integrated with the Self-Attention mechanism, enabling accurate prediction of the global ZTD. By combining all VMD-LSTM forecasted components, the final ZTD prediction was obtained. The results demonstrate that the VMD-LSTM model's average Root Mean Square Error (RMSE) in 2023 showed improvements of 62.3%, 61.9%, 76.1%, and 74.2% compared to the GPT2w, LSTM, Hopfield, and Saastamoinen models, respectively. These findings highlight the superior performance of the VMD-LSTM model in ZTD forecasting.


  • Open Access

    Regional Tropospheric Delay Prediction Model Based on LSTM-Enhanced Encoder Network

    by Yuanfang Peng, Chenglin Cai, Zexian Li, Kaihui Lv, Xue Zhang, Yihao Cai

    Precise modeling of Zenith Tropospheric Delay (ZTD) is essential for real-time high-precision positioning in global navigation satellite systems (GNSS). Due to the stochastic variability of atmospheric water vapor across different regions, tropospheric delay exhibits strong regional characteristics. Empirical tropospheric delay models built on reanalysis of meteorological data often show significant accuracy discrepancies across regions, failing to meet the needs for precise regional ZTD forecasting. Deep learning methods excel in learning complex patterns and dependencies from time series data. Our study utilized ZTD data from 178 NGL stations in Australia during 2023 as ground truth values and modeled them using an LSTM-Enhanced encoder network. This model incorporated both spatial and temporal information as well as correlations with GPT3 ZTD. Predictions were compared with those from GPT3 ZTD, ERA5 ZTD, ANN ZTD, GRNN ZTD and LSTM ZTD. The results showed that the LSTM-Enhanced encoder ZTD achieved an RMSE of 14.43 mm, a mean bias close to zero, with mean absolute error and mean correlation coefficient of 12.42 mm and 0.95, respectively. The proposed model outperforms the GPT3, ERA5, ANN, GRNN, and LSTM models, with respective RMSE improvements of approximately 62.3%, 12.3%, 61%, 59.9%, and 60%. In addition, we compared the spatial and temporal properties of the proposed model with those of the GPT3 model and the ERA5 model. The discussion section further analyzed the prediction performance of different neural network approaches under different prediction periods.