Abstract: A method for creating a correction function for improving the accuracy of a GPS device collects multiple time samples at multiple known locations wherein each time sample consists of GPS coordinates and associated satellite data from multiple satellites. The satellite data includes or permits determination of (i) satellite azimuth and elevation of an associated satellite, (ii) Signal-to-Noise Ratio of a received signal from the associated satellite, and optionally (iii) pseudo-range. For each time sample a respective error between the known location and the corresponding GPS coordinates is computed and an error correction function is created as a function of the respective GPS coordinates and the satellite data by applying deep learning/machine learning techniques to the multiple time samples.

Abstract: A method for improving accuracy of a raw GPS positioning of an untargeted pedestrian device wherein the pedestrian device receives from a nearby vehicle device a message containing a calculated offset between a raw GPS location of the vehicle and a corrected location of the vehicle, the message being received as a direct consequence of the pedestrian device and the vehicle device coming into mutual communication range without a need for pairing between the two devices. The calculated offset is applied to the raw GPS positioning of the pedestrian device to obtain a more accurate location of the pedestrian device.