Improved heading estimation for smartphone-based indoor positioning systems

Existing positioning systems such as global positioning system (GPS) or WiFi positioning system (WPS) are limited mainly to outdoor applications or are of poor accuracy without associated maps. Indoor work environment where one spend most of the time and wide adoption of smartphones with numerous embedded sensors provide new opportunities for researches on indoor user positioning. In this paper, we present an indoor positioning scheme using recently introduced smartphones equipped with sensors such as accelerometer, magnetometer, and gyroscope. We propose a practical indoor positioning method to handle complicated human movements and noisy inertial sensors in smartphones. We focus on providing indoor locations with accuracy by applying improved heading estimation solution. The key concept is to reduce errors caused by directional change since the critical component of positioning system is a heading orientation. We have simply used iPhone 4S for experiments and validated its practical usage in indoor environments. The extensive experimental results show that our system is needily applicable as a fundamental system for various mobile services on smartphones providing 2.42 times accuracy over than traditional methods on the average.