Using TensorFlow-based Neural Network to estimate GNSS single frequency ionospheric delay (IONONet)

Pérez, Raül Orús

doi:10.1016/j.asr.2018.11.011

Cited by 55 publications

(20 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For any tested GIMs, the highest daily RMS values were observed around the beginning of March and October (about 60 and 300 DOY, respectively), which may be related to the equinox periods. Similar behavior was observed by Orús-Pérez (2019), who studied the accuracy of the Galileo NeQuick model over the whole year of 2017. It is worth noting that the UQRG model, even at these two RMS peaks, did not exceed 3 TECU with few exceptions.…”

Section: Comparison With Gps Stec Data: Self-consistency Analysissupporting

confidence: 81%

Validation of GNSS-derived global ionosphere maps for different solar activity levels: case studies for years 2014 and 2018

et al. 2021

View full text Add to dashboard Cite

Ionosphere Associate Analysis Centers (IAACs) of the International GNSS Service (IGS) independently produce global ionosphere maps (GIMs) of the total electron content (TEC). The GIMs are based on different modeling techniques, resulting in different TEC levels and accuracies. In this study, we evaluated the accuracy and consistency of the IAAC GIMs during high (2014) and low (2018) solar activity periods of the 24th solar cycle. In our study, we applied two different evaluation methods. First, we carried out a comparison of the GIM-derived slant TEC (STEC) with carrier phase geometry-free combination of GNSS signals obtained from 25 globally distributed stations. Second, vertical TEC (VTEC) from GIMs was compared to altimetry-derived VTEC obtained from the Jason-2 and Jason-3 satellites and complemented for plasmaspheric TEC. The analyzed GIMs obtained STEC RMS values reaching from 1.98 to 3.00 TECU and from 0.96 to 1.29 TECU during 2014 and 2018, respectively. The comparison to altimetry data resulted in VTEC STD values that varied from 3.61 to 5.97 TECU and from 1.92 to 2.78 TECU during 2014 and 2018, respectively. The results show that among the IAACs, the Center for Orbit Determination in Europe global maps performed best in low and high solar activity periods. However, the highest accuracy was obtained by a non-IGS product—UQRG GIMs provided by Universitat Politècnica de Catalunya. It was also shown that the best results were obtained using a modified single layer model mapping function and that the map time interval has a relatively small influence on the resulting map accuracy.

show abstract

Section: Comparison With Gps Stec Data: Self-consistency Analysissupporting

confidence: 81%

Validation of GNSS-derived global ionosphere maps for different solar activity levels: case studies for years 2014 and 2018

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Deep learning is considered to be a function of second‐generation NNs (Hinton & Salakhutdinov, 2006) and may be employed to better model spatiotemporal variations in ionospheric TEC (Orus Perez, 2019). Hochreiter and Schmidhuber (1997) developed a special deep‐learning architecture known as a long short‐term memory NN (LSTM NN).…”

Section: Introductionmentioning

confidence: 99%

Long Short‐Term Memory Neural Network for Ionospheric Total Electron Content Forecasting Over China

et al. 2021

View full text Add to dashboard Cite

• A novel extended encoder-decoder long short-term memory neural network (ED-LSTME) for ionospheric TEC forecasting over China is developed • ED-LSTME shows a strong capability in improving TEC forecasting at different geographical locations, seasons, and geomagnetic conditions • ED-LSTME is robust and outperforms all the six selected baselines when comparing with the models in terms of their performance Accepted Article This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as

show abstract

“…The rapid solution is released within 24 h, but the final solution has a latency of approximately 11 days. The GIMs are more accurate than the broadcast models [12] and have been available since the official start of 1998. Besides empirical ionosphere models, there are first-principle physics models that have the potential to provide ionospheric forecasts.…”

Section: Introductionmentioning

confidence: 99%

A Neural Network-Based TEC Model Capable of Reproducing Nighttime Winter Anomaly

Adolfs

Hoque

2021

Remote Sensing

View full text Add to dashboard Cite

With the availability of fast computing machines, as well as the advancement of machine learning techniques and Big Data algorithms, the development of a more sophisticated total electron content (TEC) model featuring the Nighttime Winter Anomaly (NWA) and other effects is possible and is presented here. The NWA is visible in the Northern Hemisphere for the American sector and in the Southern Hemisphere for the Asian longitude sector under solar minimum conditions. During the NWA, the mean ionization level is found to be higher in the winter nights compared to the summer nights. The approach proposed here is a fully connected neural network (NN) model trained with Global Ionosphere Maps (GIMs) data from the last two solar cycles. The day of year, universal time, geographic longitude, geomagnetic latitude, solar zenith angle, and solar activity proxy, F10.7, were used as the input parameters for the model. The model was tested with independent TEC datasets from the years 2015 and 2020, representing high solar activity (HSA) and low solar activity (LSA) conditions. Our investigation shows that the root mean squared (RMS) deviations are in the order of 6 and 2.5 TEC units during HSA and LSA period, respectively. Additionally, NN model results were compared with another model, the Neustrelitz TEC Model (NTCM). We found that the neural network model outperformed the NTCM by approximately 1 TEC unit. More importantly, the NN model can reproduce the evolution of the NWA effect during low solar activity, whereas the NTCM model cannot reproduce such effect in the TEC variation.

show abstract

Using TensorFlow-based Neural Network to estimate GNSS single frequency ionospheric delay (IONONet)

Cited by 55 publications

References 15 publications

Validation of GNSS-derived global ionosphere maps for different solar activity levels: case studies for years 2014 and 2018

Validation of GNSS-derived global ionosphere maps for different solar activity levels: case studies for years 2014 and 2018

Long Short‐Term Memory Neural Network for Ionospheric Total Electron Content Forecasting Over China

A Neural Network-Based TEC Model Capable of Reproducing Nighttime Winter Anomaly

Contact Info

Product

Resources

About