Scholarlink

<p>Spoofing detection is an essential process in global navigation satellite system anti-spoofing. Signal quality monitoring (SQM) methods have been widely studied as simple and effective means to detect spoofing. However, the disadvantages of the existing SQM methods, such as long alarm times and low detection rates, necessitate the study of new methods. Therefore, to address these challenges, this paper proposes a novel SQM method based on a long short-term memory-detect (LSTM-Detect) model with a strong capacity for sequential signal processing. In particular, this method evaluates the distortion of the autocorrelation function (ACF) by the trained LSTM-Detect model for spoofing detection. The simulation results demonstrate that the LSTM-Detect model can detect a wide range of spoofing signals, varying in signal power advantages, code phase differences, and carrier phase differences. In the Texas Spoofing Test Battery datasets 2–6, the detection rate exceeds 98.5%, with an alarm time of less than 5 ms. Compared with five existing SQM methods, the LSTM-Detect model exhibits a more comprehensive spoofing detection performance.</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

Fast GNSS spoofing detection based on LSTM-detect model

  • 期刊论文

  • 发表日期：

    2025年1月23日

查看全文
  • Wenxuan Mao,
  • Jieyige Ren,
  • Shuyan Ni

摘要

Spoofing detection is an essential process in global navigation satellite system anti-spoofing. Signal quality monitoring (SQM) methods have been widely studied as simple and effective means to detect spoofing. However, the disadvantages of the existing SQM methods, such as long alarm times and low detection rates, necessitate the study of new methods. Therefore, to address these challenges, this paper proposes a novel SQM method based on a long short-term memory-detect (LSTM-Detect) model with a strong capacity for sequential signal processing. In particular, this method evaluates the distortion of the autocorrelation function (ACF) by the trained LSTM-Detect model for spoofing detection. The simulation results demonstrate that the LSTM-Detect model can detect a wide range of spoofing signals, varying in signal power advantages, code phase differences, and carrier phase differences. In the Texas Spoofing Test Battery datasets 2–6, the detection rate exceeds 98.5%, with an alarm time of less than 5 ms. Compared with five existing SQM methods, the LSTM-Detect model exhibits a more comprehensive spoofing detection performance.