Scholarlink

<p>Indoor Positioning Systems (IPS) have garnered substantial interest due to their critical role in navigation, asset tracking, and security applications. However, conventional IPS technologies often face significant limitations in accuracy, security, and scalability, especially in dynamic and interference-prone environments. Traditional IPS frameworks struggle to deliver high-precision localization and robust data protection, particularly in indoor environments susceptible to signal interference and cyber threats. There is an urgent need for an IPS that simultaneously ensures accurate real-time localization and maintains the integrity and confidentiality of positional data. To address these challenges, the proposed system integrates a Deep Temporal Fusion Transformer (DTFT) with Galois Field cryptography. The DTFT model is utilized to analyze both temporal dependencies and spatial patterns in sensor data, thereby improving positioning accuracy. For security, Galois Field-based encryption is implemented to safeguard positional data against unauthorized access and system vulnerabilities. The approach is validated using the WiFi RSS Fingerprint Localization Dataset in a real-world setting. Experimental evaluations reveal that the proposed system achieves a 25% improvement in localization accuracy over traditional IPS methods, reducing the average positioning error to 1.2&#xa0;m compared to the 1.6&#xa0;m typical of conventional systems. Additionally, the incorporation of Galois Field encryption introduces only a 10% increase in computational overhead, thereby offering a secure solution with minimal performance trade-offs. The proposed system effectively addresses key limitations of existing IPS frameworks by providing enhanced localization precision and robust data security. Its integration of deep learning with lightweight encryption mechanisms makes it a promising candidate for deployment in sensitive and interference-prone indoor environments.</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

A secure and high-accuracy indoor positioning system using deep temporal fusion transformer and Galois field encryption

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月3日

查看全文
  • Mohammad Mazyad Hazzazi,
  • Piyush Kumar Shukla,
  • Prashant Kumar Shukla,
  • Zaid Bassfar,
  • Amer Aljaedi,
  • Mohd Asif Shah

摘要

Indoor Positioning Systems (IPS) have garnered substantial interest due to their critical role in navigation, asset tracking, and security applications. However, conventional IPS technologies often face significant limitations in accuracy, security, and scalability, especially in dynamic and interference-prone environments. Traditional IPS frameworks struggle to deliver high-precision localization and robust data protection, particularly in indoor environments susceptible to signal interference and cyber threats. There is an urgent need for an IPS that simultaneously ensures accurate real-time localization and maintains the integrity and confidentiality of positional data. To address these challenges, the proposed system integrates a Deep Temporal Fusion Transformer (DTFT) with Galois Field cryptography. The DTFT model is utilized to analyze both temporal dependencies and spatial patterns in sensor data, thereby improving positioning accuracy. For security, Galois Field-based encryption is implemented to safeguard positional data against unauthorized access and system vulnerabilities. The approach is validated using the WiFi RSS Fingerprint Localization Dataset in a real-world setting. Experimental evaluations reveal that the proposed system achieves a 25% improvement in localization accuracy over traditional IPS methods, reducing the average positioning error to 1.2 m compared to the 1.6 m typical of conventional systems. Additionally, the incorporation of Galois Field encryption introduces only a 10% increase in computational overhead, thereby offering a secure solution with minimal performance trade-offs. The proposed system effectively addresses key limitations of existing IPS frameworks by providing enhanced localization precision and robust data security. Its integration of deep learning with lightweight encryption mechanisms makes it a promising candidate for deployment in sensitive and interference-prone indoor environments.