Application of a Multi‐Layer Artificial Neural Network in a 3‐D Global Electron Density Model Using the Long‐Term Observations of COSMIC, Fengyun‐3C, and Digisonde

Li, Wang; Zhao, Dongsheng; Hé, Changyong; Shen, Yi; Hu, Andong; Zhang, Kefei

doi:10.1029/2020sw002605

Cited by 19 publications

(15 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, a number of ML-based electron density models in the Earth’s ionosphere have been developed 15 – 20 . Several models provide the F2-peak parameters 15 , 18 , while others reproduce three dimensional electron density distributions 16 , 17 . However, most existing ML-based models binned the data into spatial cells in terms of geographic latitude and longitude and thus do not provide continuous output 16 , 19 , 20 .…”

Section: Introductionmentioning

confidence: 99%

“…One of the most efficient ways to make use of the vast amounts of data for empirical modeling is by applying machine learning (ML) techniques. In recent years, a number of ML-based electron density models in the Earth's ionosphere have been developed [15][16][17][18][19][20] . Several models provide the F2-peak parameters 15,18 , while others reproduce three dimensional electron density OPEN 1 Helmholtz Centre Potsdam -GFZ German Research Centre for Geosciences, Potsdam, Germany.…”

mentioning

confidence: 99%

See 1 more Smart Citation

A novel neural network model of Earth’s topside ionosphere

Smirnov

Shprits

Prol

et al. 2023

Sci Rep

View full text Add to dashboard Cite

The Earth’s ionosphere affects the propagation of signals from the Global Navigation Satellite Systems (GNSS). Due to the non-uniform coverage of available observations and complicated dynamics of the region, developing accurate models of the ionosphere has been a long-standing challenge. Here, we present a Neural network-based model of Electron density in the Topside ionosphere (NET), which is constructed using 19 years of GNSS radio occultation data. The NET model is tested against in situ measurements from several missions and shows excellent agreement with the observations, outperforming the state-of-the-art International Reference Ionosphere (IRI) model by up to an order of magnitude, especially at 100-200 km above the F2-layer peak. This study provides a paradigm shift in ionospheric research, by demonstrating that ionospheric densities can be reconstructed with very high fidelity. The NET model depicts the effects of numerous physical processes governing the topside dynamics and can have wide applications in ionospheric research.

show abstract

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

A novel neural network model of Earth’s topside ionosphere

Smirnov

Shprits

Prol

et al. 2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…In [7], the nonlinear least square estimation technique is used to allow the TEC modelling for a single-station. In the literature, several studies have focused on comprehension of storm-time behavior of the ionosphere to reduce the influence of ionospheric anomalies and irregularities on global positioning services and to advance the performance of the ionospheric models during the major geomagnetic storms [8,9,10,11]. Also, disturbances and irregularities on TEC due to seismic activity, Solar Flares and solar activity cause the deviations on precise of the satellite navigation and the positioning systems [12,13,14].…”

Section: Introductionmentioning

confidence: 99%

Comparison of the Performance of the Regression Models in GPS-Total Electron Content Prediction

2023

View full text Add to dashboard Cite

The ionosphere is an important layer that provides radio communication in the upper atmosphere. The ionosphere is located between 50 km and 1000 km above the atmosphere. Electron density, which is the most important parameter of the ionosphere, changes depending on location, time, seasons, altitude, solar, geomagnetic and seismic activity. A significant measurable amount of electron density is Total Electron Content (TEC), which is used to probe the structure of the ionosphere and upper atmosphere. The Global Positioning System (GPS), which has a low cost and widespread receiver network is prominent used in TEC estimation. The IONOLAB-TEC data estimated from GPS is used in this study. Prediction of TEC is important phenomenon to operate and to plan the Earth-space and satellite-to-satellite communication systems, to generate the earthquake precursor signals using TEC and to detect of anomalies in the ionosphere. In this study, IONOLAB-TEC data obtained from GPS is estimated using regression models. Among the tested algorithms, it is observed that the Exponential Gaussian Process Regression and Interactions Linear Regression algorithms are very successful and high-performance models for TEC estimation.

show abstract

“…For these two aspects, deep learning methods are promising as they are well known for progressively extracting higher‐level features from the raw input variables, and one of the benefits is that there is no need to identify or specify the functional relationship (e.g., linear or nonlinear) of the model. This is why deep learning methods are often used to model the ionospheric TEC (Li, Zhao, et al., 2021; Ruwali et al., 2021; Xiong et al., 2021; Zewdie et al., 2021; Zhukov et al., 2020).…”

Section: Introductionmentioning

confidence: 99%

An Investigation of Ionospheric TEC Prediction Maps Over China Using Bidirectional Long Short‐Term Memory Method

Shi

Zhang

et al. 2022

Space Weather

Self Cite

View full text Add to dashboard Cite

The ionospheric total electron content (TEC) is an important ionospheric parameter, and it is widely utilized in research such as space weather prediction and precise positioning. However, it is still challenging to develop an ionospheric TEC prediction model with high accuracy. In this study, a new ionospheric TEC model over China was developed using the bidirectional long short‐term memory (bi‐LSTM) method and observations from 257 ground‐based global navigation satellite system (GNSS) stations in the Crustal Movement Observation Network of China from January 2018 to December 2021. The root mean square errors of the bi‐LSTM‐based model’s 1 and 2 hr ahead predictions on the test data set (from June 2021 to December 2021) are 1.12 and 1.68 TECU, respectively, which are 75/50/32% and 72/48/22% smaller than those of the IRI‐2016, artificial neural network and LSTM‐based models, correspondingly. The bi‐LSTM‐based model shows the best performance, which is most likely due to the fact that the sequence information in both forward and backward directions is taken into consideration in the new model. In addition, the diurnal variation, seasonal variation of the ionospheric TEC, and variations under geomagnetic storm conditions are successfully captured by the bi‐LSTM‐based model. Moreover, the TEC maps resulting from the bi‐LSTM model agree well with those obtained from the final ionospheric product from the Chinese Academy of Sciences. Hence, the new model can be a good choice for the investigation of the spatiotemporal variation trend in the ionosphere and GNSS navigation.

show abstract

Application of a Multi‐Layer Artificial Neural Network in a 3‐D Global Electron Density Model Using the Long‐Term Observations of COSMIC, Fengyun‐3C, and Digisonde

Cited by 19 publications

References 43 publications

A novel neural network model of Earth’s topside ionosphere

A novel neural network model of Earth’s topside ionosphere

Comparison of the Performance of the Regression Models in GPS-Total Electron Content Prediction

An Investigation of Ionospheric TEC Prediction Maps Over China Using Bidirectional Long Short‐Term Memory Method

Contact Info

Product

Resources

About