Ahmed Sherif, Mostafa Rabah, Ashraf El-Kutb Mousa, Ahmed Zaki, Mohamed Anwar, Ahmed Sedeek
{"title":"利用COSMIC-2 GNSS无线电掩星和人工神经网络对埃及上空电离层TEC进行建模","authors":"Ahmed Sherif, Mostafa Rabah, Ashraf El-Kutb Mousa, Ahmed Zaki, Mohamed Anwar, Ahmed Sedeek","doi":"10.1515/jag-2023-0079","DOIUrl":null,"url":null,"abstract":"Abstract The ionospheric delay significantly impacts GNSS positioning accuracy. To address this, an Artificial Neural Network (ANN) was developed using the high-quality COSMIC-2 ionospheric profile dataset to predict the Total Electron Content (TEC). ANNs are adept at addressing both linear and nonlinear challenges. For this research, eight distinct ANNs were cultivated. These ANNs were designed with the following inputs Year, Month, Day, Hour, Latitude, and Longitude. Along with solar and geomagnetic parameters such as the F10.7 solar radio flux index, the Sunspot Number (SSN), the Kp index, and the ap index. The goal was to discern the most influential parameters on ionosphere prediction. After pinpointing these key parameters, an enhanced model utilizing a pioneering technique of a secondary ANN was employed with the main ANN to predict TEC values for events in 2023. The study’s findings indicate that solar parameters markedly enhance the model’s accuracy. Notably, the augmented model featuring a prelude secondary network achieved a stellar correlation coefficient of 0.99. Distributionally, 41 % of predictions aligned within the (−1≤ ΔTEC ≤1) TECU spectrum, 28 % nestled within the (1< ΔTEC ≤2) and (−2≤ ΔTEC <−1) TECU ambit, while a substantial 30 % spanned the broader (2< ΔTEC ≤5) and (−5≤ ΔTEC <−2) TECU range. In essence, this research underscores the potential of incorporating solar parameters and advanced neural network techniques to refine ionospheric delay predictions, thus boosting GNSS positioning precision.","PeriodicalId":45494,"journal":{"name":"Journal of Applied Geodesy","volume":null,"pages":null},"PeriodicalIF":1.2000,"publicationDate":"2023-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Ionospheric TEC modeling using COSMIC-2 GNSS radio occultation and artificial neural networks over Egypt\",\"authors\":\"Ahmed Sherif, Mostafa Rabah, Ashraf El-Kutb Mousa, Ahmed Zaki, Mohamed Anwar, Ahmed Sedeek\",\"doi\":\"10.1515/jag-2023-0079\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract The ionospheric delay significantly impacts GNSS positioning accuracy. To address this, an Artificial Neural Network (ANN) was developed using the high-quality COSMIC-2 ionospheric profile dataset to predict the Total Electron Content (TEC). ANNs are adept at addressing both linear and nonlinear challenges. For this research, eight distinct ANNs were cultivated. These ANNs were designed with the following inputs Year, Month, Day, Hour, Latitude, and Longitude. Along with solar and geomagnetic parameters such as the F10.7 solar radio flux index, the Sunspot Number (SSN), the Kp index, and the ap index. The goal was to discern the most influential parameters on ionosphere prediction. After pinpointing these key parameters, an enhanced model utilizing a pioneering technique of a secondary ANN was employed with the main ANN to predict TEC values for events in 2023. The study’s findings indicate that solar parameters markedly enhance the model’s accuracy. Notably, the augmented model featuring a prelude secondary network achieved a stellar correlation coefficient of 0.99. Distributionally, 41 % of predictions aligned within the (−1≤ ΔTEC ≤1) TECU spectrum, 28 % nestled within the (1< ΔTEC ≤2) and (−2≤ ΔTEC <−1) TECU ambit, while a substantial 30 % spanned the broader (2< ΔTEC ≤5) and (−5≤ ΔTEC <−2) TECU range. In essence, this research underscores the potential of incorporating solar parameters and advanced neural network techniques to refine ionospheric delay predictions, thus boosting GNSS positioning precision.\",\"PeriodicalId\":45494,\"journal\":{\"name\":\"Journal of Applied Geodesy\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":1.2000,\"publicationDate\":\"2023-10-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Applied Geodesy\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1515/jag-2023-0079\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"REMOTE SENSING\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Applied Geodesy","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1515/jag-2023-0079","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"REMOTE SENSING","Score":null,"Total":0}
Ionospheric TEC modeling using COSMIC-2 GNSS radio occultation and artificial neural networks over Egypt
Abstract The ionospheric delay significantly impacts GNSS positioning accuracy. To address this, an Artificial Neural Network (ANN) was developed using the high-quality COSMIC-2 ionospheric profile dataset to predict the Total Electron Content (TEC). ANNs are adept at addressing both linear and nonlinear challenges. For this research, eight distinct ANNs were cultivated. These ANNs were designed with the following inputs Year, Month, Day, Hour, Latitude, and Longitude. Along with solar and geomagnetic parameters such as the F10.7 solar radio flux index, the Sunspot Number (SSN), the Kp index, and the ap index. The goal was to discern the most influential parameters on ionosphere prediction. After pinpointing these key parameters, an enhanced model utilizing a pioneering technique of a secondary ANN was employed with the main ANN to predict TEC values for events in 2023. The study’s findings indicate that solar parameters markedly enhance the model’s accuracy. Notably, the augmented model featuring a prelude secondary network achieved a stellar correlation coefficient of 0.99. Distributionally, 41 % of predictions aligned within the (−1≤ ΔTEC ≤1) TECU spectrum, 28 % nestled within the (1< ΔTEC ≤2) and (−2≤ ΔTEC <−1) TECU ambit, while a substantial 30 % spanned the broader (2< ΔTEC ≤5) and (−5≤ ΔTEC <−2) TECU range. In essence, this research underscores the potential of incorporating solar parameters and advanced neural network techniques to refine ionospheric delay predictions, thus boosting GNSS positioning precision.