Scholarlink

Emerging artificial intelligence and sensor fusion technologies promise to advance indoor positioning and navigation systems. To achieve precise indoor positioning and heading estimation in complex environments, our study leverages smartphone capabilities. We employ a systematic methodology centered on indoor localization and heading estimation using Wi-Fi Received Signal Strength Indicator (RSSI) and orientation sensor data. Meticulous pre-processing refines acquired Wi-Fi RSSI data, integrated into well-structured machine learning algorithms. Accuracy is augmented by incorporating a heuristic trilateration algorithm alongside the Improved Weighted KNN technique. This approach addresses complexities inherent in indoor environments, paving the way for robust and practical indoor positioning and navigation systems. Our practical implementation achieves a positioning accuracy of 2.32 m, marking significant progress. Beyond accuracy, our method excels in practicality, affordability, convenience, and swiftness, distinguishing it from prior studies. It offers a more practical, cost-effective, and efficient indoor positioning solution with broader applicability.

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

Indoor Localization and Heading Estimation Leveraging Machine Learning Algorithms with Wi-Fi Fingerprints and Sensor Data

  • 章节

  • 发表日期：

    2026年

查看全文
  • Roshan Kumar Yadav,
  • Divyam Kumar,
  • Rohan Saraswat,
  • Shilpa Gite,
  • Biswajeet Pradhan

摘要

Emerging artificial intelligence and sensor fusion technologies promise to advance indoor positioning and navigation systems. To achieve precise indoor positioning and heading estimation in complex environments, our study leverages smartphone capabilities. We employ a systematic methodology centered on indoor localization and heading estimation using Wi-Fi Received Signal Strength Indicator (RSSI) and orientation sensor data. Meticulous pre-processing refines acquired Wi-Fi RSSI data, integrated into well-structured machine learning algorithms. Accuracy is augmented by incorporating a heuristic trilateration algorithm alongside the Improved Weighted KNN technique. This approach addresses complexities inherent in indoor environments, paving the way for robust and practical indoor positioning and navigation systems. Our practical implementation achieves a positioning accuracy of 2.32 m, marking significant progress. Beyond accuracy, our method excels in practicality, affordability, convenience, and swiftness, distinguishing it from prior studies. It offers a more practical, cost-effective, and efficient indoor positioning solution with broader applicability.