Indoor Positioning And Navigation Based On WIFI

Abstract

Indoor spaces are critical for human living and working activities. With the rapid expansion of indoor environments, there is a growing demand for cost-effective indoor positioning solutions. WiFi, as the most widely used wireless communication technology, enables both human navigation via mobile devices and robot path planning through instruction issuance. This paper introduces the WCSA (Weighted Cosine Similarity Algorithm), a novel indoor positioning method based on cosine similarity and a location fingerprint database. The WCSA algorithm achieves high accuracy in indoor positioning systems. Additionally, by integrating Simultaneous Localization and Mapping(SLAM) with robotic environmental mapping, the system enables automated data collection during the offline training phase of indoor positioning algorithms. Extensive experiments demonstrate that the WCSA algorithm outperforms traditional methods such as K-Nearest Neighbors (KNN) and Cosine Similarity (COS) by over 0.8 meters, achieving a positioning accuracy of 1-3 meters. The proposed solution is versatile, supporting applications in both commercial scenarios(e.g., autonomous robot inspections in subways and shopping malls) and critical industries (e.g., emergency firefighting).