Multisource Indoor Positioning System Based on Edge Computing

Abstract

Due to its significance in numerous applications, indoor positioning has garnered increasing interest. The WiFi fingerprinting method and pedestrian dead reckoning approach are the most popular methods for indoor positioning. However, both methods have certain limitations in terms of accuracy and response time, rendering them ineffective for some applications. To overcome these problems, we propose a system based on edge computing that combines these two methods. The results of our experiments show that our proposed method outperforms standalone methods by giving an average positioning error of 0.75 m in just 0.7 seconds.