Scholarlink

<p>Spoofing attacks are one of the interferences that cause errors in the position estimated by Global Positioning System (GPS) receivers. To confront these attacks, there are various anti-spoofing methods. These methods are effective at different levels of signal processing to confront and defend against spoofing. This study addresses intermediate spoofing attacks with a strategy based on a single receiver and single antenna, with an emphasis on detection and mitigation. This article identifies spoofed channels using Receiver Autonomous Integrity Monitoring (RAIM). After RAIM detects the spoofing and a differential concept assists in identifying the fake PRN number, a Neural Network (NN) trained by Particle Swarm Optimization (PSO) begins the correction phase. In particular, a Multi-Layer Perceptron Neural Network (MLP NN) trained by PSO is proposed to follow proper PRN value trends. The proposed NN is trained by other Evolutionary Algorithms (EAs) such as Genetic Algorithm (GA), Independent Component Analysis (ICA), and Craziness PSO (CRPSO). Among all the examined EAs, NN training by PSO has the lowest Mean Square Error (MSE), but due to its computational complexity, the time spent to train such a trained NN is a bit more than other EAs.</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

GPS Spoofing Attack Detection and Correction in the Pseudo-Range of PRNs Using Neural Network

  • 期刊论文

  • 发表日期：

    2026年4月29日

查看全文
  • M. Ghandchi,
  • N. Orouji,
  • S. Tohidi,
  • M. R. Mosavi

摘要

Spoofing attacks are one of the interferences that cause errors in the position estimated by Global Positioning System (GPS) receivers. To confront these attacks, there are various anti-spoofing methods. These methods are effective at different levels of signal processing to confront and defend against spoofing. This study addresses intermediate spoofing attacks with a strategy based on a single receiver and single antenna, with an emphasis on detection and mitigation. This article identifies spoofed channels using Receiver Autonomous Integrity Monitoring (RAIM). After RAIM detects the spoofing and a differential concept assists in identifying the fake PRN number, a Neural Network (NN) trained by Particle Swarm Optimization (PSO) begins the correction phase. In particular, a Multi-Layer Perceptron Neural Network (MLP NN) trained by PSO is proposed to follow proper PRN value trends. The proposed NN is trained by other Evolutionary Algorithms (EAs) such as Genetic Algorithm (GA), Independent Component Analysis (ICA), and Craziness PSO (CRPSO). Among all the examined EAs, NN training by PSO has the lowest Mean Square Error (MSE), but due to its computational complexity, the time spent to train such a trained NN is a bit more than other EAs.