A Low‐Latitude Three‐Dimensional Ionospheric Electron Density Model Based on Radio Occultation Data Using Artificial Neural Networks With Prior Knowledge

Yang, Ding; Fang, Hanxian

doi:10.1029/2022sw003299

Cited by 7 publications

(2 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ionosonde stations and incoherent scatter radars, for instance, are commonly used as a reference to the validation [12][13][14][15][16][17][18][19][20][21][22] since they provide accurate observations of the electron density. In-situ measurements provided by external satellite missions are also extensively used to assess the RO measurements [23][24][25][26][27][28][29][30]. Another typical evaluation metric is to compare the RO observations against climatological models of the ionosphere [31][32][33][34][35].…”

Section: Introductionmentioning

confidence: 99%

Study of Ionospheric Bending Angle and Scintillation Profiles Derived by GNSS Radio-Occultation with MetOp-A Satellite

et al. 2023

View full text Add to dashboard Cite

In this work, a dedicated campaign by MetOp-A satellite is conducted to monitor the ionosphere based on radio-occultation (RO) measurements provided by the onboard GNSS (Global Navigation Satellite System) Receiver for Atmospheric Sounding (GRAS). The main goal is to analyze the capabilities of the collected data to represent the bending angle and scintillation profiles of the ionosphere. We compare the MetOp-A products with those generated by other RO missions and explore the spatial/temporal distributions sensed by the MetOp-A campaign. Validation of dual frequency bending angles at the RO tangent points, S4 index, and Rate of the Total electron content Index (ROTI) is performed against independent products from Fengyun-3D and FORMOSAT-7/COSMIC-2 satellites. Our main findings constitute the following: (1) bending angle profiles from MetOp-A agree well with Fengyun-3D measurements; (2) bending angle distributions show a typical S-shape variation along the altitudes; (3) signatures of the sporadic E-layer and equatorial ionization anomaly crests are observed by the bending angles; (4) sharp transitions are observed in the bending angle profiles above ~200 km due to the transition of the daytime/nighttime in addition to the transition of the bottom-side/top-side; and (5) sporadic E-layer signatures are observed in the S4 index distributions by MetOp-A and FORMOSAT-7/COSMIC-2, with expected differences in magnitudes between the GPS (Global Positioning System) L1 and L2 frequencies.

show abstract

Section: Introductionmentioning

confidence: 99%

Study of Ionospheric Bending Angle and Scintillation Profiles Derived by GNSS Radio-Occultation with MetOp-A Satellite

et al. 2023

View full text Add to dashboard Cite

show abstract

“…A threehidden-layer DNN was used for a global 3D model ("ANN-TDD") based on COSMIC, Fengyun-3C and Digisonde data [Li et al, 2021]. The most recent work combined DNN with IRI ("ANN-IRI") to improve Ne prediction compared to pure data-driven ANNs, particular in the lower ionosphere [Yang and Fang, 2023]. These pioneer models reproduce the large-scale ionospheric phenomena and generally outperform the monthly-average model of IRI-2016 during the quiet time.…”

Section: Introdmentioning

confidence: 99%

Neural Network Models for Ionospheric Electron Density Prediction: A Neural Architecture Search Study

Pan,

Jin,

Zhang

et al. 2023

Preprint

View full text Add to dashboard Cite

Specification and forecast ionospheric parameters, such as ionospheric electron density (Ne), have been an important topic in space weather and ionosphere research. Neural networks (NNs) emerge as a powerful modeling tool for Ne prediction. However, heavy manual attention costs time to determine the optimal NN structures. In this work, we propose to use neural architecture search (NAS), an automatic machine learning method, to address this problem of NN models. NAS aims to find the optimal network structure through the alternated optimization of the hyperparameters and the corresponding network parameters. A total of 16-year data from Millstone Hill incoherent scatter radar (ISR) are used for NN models. One single-layer NN (SLNN) model and one deep NN (DNN) model are trained with NAS, namely SLNN-NAS and DNN-NAS, for Ne prediction and compared with their counterparts without NAS from previous studies, denoted as SLNN and DNN. Our results show that SLNN-NAS and DNN-NAS outperformed SLNN and DNN, respectively. NN models can reveal more finer details than the empirical ionospheric model developed using traditional data fitting approaches. DNN-NAS yields the best prediction accuracy measured by quantitative metrics and rankings of daily pattern prediction. The limited improvement of NAS is likely due to the network complexity and the limitation of fully connected NN without a memory mechanism.

show abstract

A storm-time global electron density reconstruction model in three-dimensions based on artificial neural networks

Habarulema,

Okoh,

Burešová

et al. 2024

Advances in Space Research

View full text Add to dashboard Cite

A Low‐Latitude Three‐Dimensional Ionospheric Electron Density Model Based on Radio Occultation Data Using Artificial Neural Networks With Prior Knowledge

Cited by 7 publications

References 46 publications

Study of Ionospheric Bending Angle and Scintillation Profiles Derived by GNSS Radio-Occultation with MetOp-A Satellite

Study of Ionospheric Bending Angle and Scintillation Profiles Derived by GNSS Radio-Occultation with MetOp-A Satellite

Neural Network Models for Ionospheric Electron Density Prediction: A Neural Architecture Search Study

A storm-time global electron density reconstruction model in three-dimensions based on artificial neural networks

Contact Info

Product

Resources

About