kunlin yang

Mr. kunlin yang


Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 November 2022 in Remote Sensing
Verified
Journal Article
Open access
PDF file
KYkunlin yang
kunlin yang
YLYang Liu
Yang Liu
The ionosphere serves as a critical medium for radio signal propagation in outer space. A good morphology of the global TEC distribution is very useful for both ionospheric studies and their relative applications. In this work, a deep learning framework was constructed for better spatial estimation in ionospheric TEC. Both the DCGAN and WGAN-GP were considered, and their performances were evaluated with spatial completion for a regional TEC. The performances were evaluated using the correlation coefficient, RMSE, and MAE. Moreover, the IAAC rapid products were used to make comparisons. The results show that both the DCGAN and WGAN-GP outperformed the IAAC CORG rapid products. The spatial TEC estimation clearly goes well with the solar activity trend. The RMSE differences had a maximum of 0.5035 TECu between the results of 2009 and 2014 for the DCGAN and a maximum of 0.9096 TECu between the results of 2009 and 2014 for the WGAN-GP. Similarly, the MAE differences had a maximum of 0.2606 TECu between the results of 2009 and 2014 for DCGAN and a maximum of 0.3683 TECu between the results of 2009 and 2014 for WGAN-GP. The performances of the CORG, DCGAN, and WGAN-GP were also verified for two selected strong geomagnetic storms in 2014 and 2017. The maximum RMSEs were 1.8354 TECu and 2.2437 TECu for the DCGAN and WGAN-GP in the geomagnetic storm on 18 February 2014, respectively, and the maximum RMSEs were 1.3282 TECu and 1.4814 TECu in the geomagnetic storm on 7 September 2017. The GAN-based framework can extract the detailed features of spatial TEC daily morphologies and the responses during geomagnetic storms.

16 June 2025 in Atmosphere
Verified
Journal Article
Open access
PDF file
YCYifei Chen
Yifei Chen
YLYang Liu
Yang Liu
KYkunlin yang
kunlin yang
LLLanhao Li
Lanhao Li
Chao XiongChao Xiong
Chao Xiong
JWJinling Wang
Jinling Wang
Predicting global ionospheric total electron content (TEC) is critical for high-precision GNSS applications, but some existing models fail to jointly capture spatial heterogeneity and multiscale temporal trends. To address the problem, this work proposes a spatio-temporal graph neural network (STGNN) that addresses these limitations through (1) a trainable positional attention mechanism to dynamically infer node dependencies without fixed geographical constraints and (2) a GRU–Transformer sequential module to hierarchically model local and global temporal patterns. The proposed network is validated by different solar and geomagnetic activities. With a training dataset with a time span between 2008 and 2018, the proposed model is tested in a high solar phase for the year 2015 and a low solar phase for the year 2018. For 2015, the experimental results show a 21.9% RMSE reduction at low latitudes compared to the results of the iTransformer model. For the geomagnetic storm event, the proposed STGNN achieves 16.0% higher stability. For the one-week (84 step) prediction test, the STGNN shows a 27.0% lower error compared to the MLPMultivariate model. The model’s self-adaptive spatial learning and multiscale temporal modeling uniquely enable TEC forecasting under diverse geophysical conditions.