Location Corrections through Differential Networks (LOCD-IN)
Evan Dill, Russell Gilabert, NASA Langley Research Center; Maarten Uijt de Haag, Ohio University
In recent years, the development and use of internet enabled mobile devices with built in Global Navigation Satellite Systems (GNSS) receivers (e.g. cell phones, tablet computers) have seen substantial increases. Correspondingly, the quantity and variety of applications using the GNSS capabilities of these devices have grown to and include domains such as ground vehicle navigation, gaming and pedestrian navigation. However, these devices typically have limited precise positioning capabilities due to minimized Size, Weight, Power and Cost (SWaP-C) for their GNSS antennas and chip sets. Research suggests that the expected position accuracy for a typical mobile device (cell phone or tablet computer) is around 8-20 meters. This error hinders the implementation of applications that require a more precise position, such as establishing drop zones for unmanned aircraft system (UAS) package delivery or high accuracy E911 emergency rescue. Furthermore, any hardware modifications to mobile devices that would increase the SWaP-C of GNSS related components is highly undesirable for the manufacturers of these devices. As such, this work proposes the LOcation Corrections through DIfferential Networks (LOCD-IN) system as a means of increasing the position accuracy of cheap, network enabled mobile devices without making any hardware modifications to the device’s equipment.
LOCD-IN uses Global Positioning System (GPS) pseudorange, code phase and carrier phase (when available) measurements from a mobile device, combined with data provided via a network connection from stationary base stations (similar to systems such as CORS or NTRIPS). The raw GPS measurements from the mobile device are refined using established methods to calculate a geo-referenced position with improved accuracy. These established methods may include, but are not limited to, those used for differential GPS, Real-Time Kinematic (RTK) GPS and/or the Ground Based Augmentation System (GBAS) (formerly the Local Area Augmentation System (LAAS)). Such methods remove errors common to the measurements of both devices (the mobile and base station) thereby increasing the accuracy of position estimates for the mobile device.
This work focuses on the preliminary development, implementation and testing of the LOCD-IN system as a means of increasing the position accuracy of existing mobile devices. A prototype implementation of LOCD-IN is described in which multiple small, relatively cheap (<$300) base stations were constructed and installed around the campus of the NASA Langley Research Center. A wireless network and server were also created to aggregate each base stations measurement data and make it accessible to other devices on the network. Using this data, a procedure is established to estimate and select the base station in the network that will produce the highest quality position corrections for the mobile device. Additionally, a methodology is laid out that allows the use of various well defined GPS processing techniques to yield corrections for mobile devices with network access. Next, experimental results are presented using an unmodified commercial-of-the-shelf (COTS) tablet. These results are discussed and a comparison is made between the positioning capabilities of an un-aided mobile device and the same device using corrections derived from the LOCD-IN system. Lastly, conclusions and planned future work are presented.