GNSS Fault Monitoring using Android Devices

Since the introduction of raw GNSS measurements in Android devices, the applications of Android GNSS have been expanding rapidly. The users have become able to use GNSS measurements including code and carrier phase to carry out post-processing such as differential GNSS for high accuracy and precision navigation. However, one of the limiting factors that restrict Android GNSS applications is the lack of accurate GNSS measurement noise models and measurement fault detection mechanisms. As most Android devices have small isotropic GNSS antenna due size constraints, and unknown operating orientations, they have greater measurement noise and more frequent faults such as cycle slips than geodetic receivers. Therefore, if accurate noise models and fault detection mechanisms are not present, it is difficult to effectively apply most position, navigation, and time algorithms to Android measurements. There have been several efforts in the past to model the Android measurement noise, but in this study we outline the limitations of those approaches and propose a more effective method to accurately and empirically model the measurement noise. In addition, we propose an innovative Android fault detection methodology to detect any potential cycle slips in the GNSS carrier phase measurements. Afterwards, both the noise models and the fault detection mechanisms are compared to the pre-existing ones provided by the Android raw GNSS measurements using live data in order to demonstrate the effectiveness and potential benefits of this research in improving the applicability of Android GNSS measurements.