Research Article
A Study on the Error Characteristics of Smartphone Inertial Sensors
@INPROCEEDINGS{10.1007/978-3-030-62205-3_12, author={Huifeng Li}, title={A Study on the Error Characteristics of Smartphone Inertial Sensors}, proceedings={Mobile Wireless Middleware, Operating Systems and Applications. 9th EAI International Conference, MOBILWARE 2020, Hohhot, China, July 11, 2020, Proceedings}, proceedings_a={MOBILWARE}, year={2020}, month={11}, keywords={Sensor errors Smartphones Gaussian distribution}, doi={10.1007/978-3-030-62205-3_12} }- Huifeng Li
Year: 2020
A Study on the Error Characteristics of Smartphone Inertial Sensors
MOBILWARE
Springer
DOI: 10.1007/978-3-030-62205-3_12
Abstract
The inertial sensors embedded in current smartphones are being used in a variety of applications, including motion monitoring, safe driving, panoramic roaming, Pedestrian Dead Reckoning (PDR), etc. Since the performance of these sensors has significant influences on these applications, it is of great value to comprehensively understand how the measurements returned by these sensors are statistically distributed. Most existing studies assume white Gaussian noises in sensor measurements, which is not experimentally confirmed in realistic and dynamic scenarios with commercial off-the-shelf (COTS) smartphones. In this paper, we study the statistical error characteristics of sensor measurements through extensive experiments in practice. The experimental results reveal that, when the device is stationary, the sensor measurement errors fully obey the standard Gaussian distribution; when the speed of smartphones increases, the sensor measurement errors begin rising, and the discrepancy between its distribution and the Gaussian distribution is enlarged. This paper establishes foundation for studying the statistical characteristics of the measurement errors of smartphone inertial sensors.
