Scholarlink

<p>Ionospheric scintillations cause major impairments in global navigation satellite system (GNSS) received L band signals on either ground or space-based platforms from GNSS satellite constellations. The degradation of positional accuracy and availability of GNSS receivers frequently occurs during ionospheric scintillation, leading to loss of lock and reduced signal strength. In this paper, an attempt is made to apply bi-directional long short-term memory (Bi-LSTM) deep learning prediction model for predicting the amplitude scintillation index (S<sub>4</sub>) of GNSS receivers, observations collected from a low latitude GNSS station in Hyderabad, India. Three geomagnetic storm events of the year 2015 are considered for the analysis. Solar and geomagnetic indices, which reflect their impact on the ionosphere, are used as input variables alongside S<sub>4</sub> indices to train the bi-LSTM model for predicting S<sub>4</sub> index values. The comparison results indicate that the Bi-LSTM model can perform well as compared to the other deep learning models. The proposed S<sub>4</sub> prediction model can be considered as a prime contender for the S<sub>4</sub> prediction model in GNSS early warning ionospheric scintillation system for civil aviation and GNSS augmentations system applications.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Bi-directional LSTM deep learning model for low latitude GNSS ionospheric scintillations

  • 期刊论文

  • 发表日期：

    2025年6月3日

查看全文
  • Veera Kumar Bellapukonda,
  • Venkata Ratnam Devanaboyina,
  • Sridhar Miriyala

摘要

Ionospheric scintillations cause major impairments in global navigation satellite system (GNSS) received L band signals on either ground or space-based platforms from GNSS satellite constellations. The degradation of positional accuracy and availability of GNSS receivers frequently occurs during ionospheric scintillation, leading to loss of lock and reduced signal strength. In this paper, an attempt is made to apply bi-directional long short-term memory (Bi-LSTM) deep learning prediction model for predicting the amplitude scintillation index (S4) of GNSS receivers, observations collected from a low latitude GNSS station in Hyderabad, India. Three geomagnetic storm events of the year 2015 are considered for the analysis. Solar and geomagnetic indices, which reflect their impact on the ionosphere, are used as input variables alongside S4 indices to train the bi-LSTM model for predicting S4 index values. The comparison results indicate that the Bi-LSTM model can perform well as compared to the other deep learning models. The proposed S4 prediction model can be considered as a prime contender for the S4 prediction model in GNSS early warning ionospheric scintillation system for civil aviation and GNSS augmentations system applications.