Abstract: Device-based barometric pressure sensor calibration involves determining a specific location of the mobile device; determining a general location of the mobile device that encompasses and obfuscates the specific location; transmitting the general location to a server; receiving general calibration data for the general location; determining specific calibration data based on the general calibration data and the specific location; determining a calibration value based on the specific calibration data, the calibration value being for calibrating the barometric pressure sensor.

Abstract: A method involves determining weighted metric values for each metric of a plurality of metrics by applying a weight for each metric to a determined value of each metric. The method further involves using the weighted metric values to determine if a pressure sensor of a mobile device should be calibrated using information associated with the first location, and if a determination is made that the pressure sensor of the mobile device should be calibrated using information associated with the first location, then using the information associated with the first location to calibrate the pressure sensor of the mobile device.

Abstract: Field calibration of a pressure device involves collecting simultaneous pressure data or pressure and temperature data at two devices for multiple time points. Pressure differences between pairs of simultaneous data points of the collected pressure data are calculated. A model is fitted to the pressure differences and the temperatures and/or pressures, and model parameters are used to correct measurements from the second device. Alternatively, a pressure gradient is estimated for a region that encompasses the two devices for each time point. A distance is determined between the two devices. A pressure gradient difference is determined between the two devices for each time point. A pressure difference offset is obtained for one of the pairs of simultaneous data points for each time point. An average pressure difference offset is determined between the two devices and is used to correct measurements from one of the devices.

Abstract: Calibrating an unstable sensor of a mobile device. Systems and methods for calibrating a sensor of a mobile device determine a first estimated position of the mobile device without using any measurement from the sensor of the mobile device, generate a second estimated position of the mobile device using a measurement from the sensor, estimate a sensor error of the sensor using the first estimated position and the second estimated position, and use the sensor error to determine a calibration value for adjusting one or more measurements from the sensor.

Abstract: Determining one or more heights of one or more mobile devices above surfaces. Particular embodiments described herein include machines that retrieve first data (e.g., measurement value(s) determined by sensor(s) of a mobile device or estimated position(s) of the mobile device), determine a location context based on the first data, identify second data (e.g., measurement value(s) determined by sensor(s) of the mobile device or status indicator value(s) of feature(s) of the mobile device) to retrieve for use in determining an estimated height above a surface at which the mobile device is located based on the determined location context, retrieve the second data, and determine an estimated height above a surface at which the mobile device is located based on the retrieved second data.

Abstract: Multiple calibration results for calibrating a barometric pressure sensor based on data received from a device containing the sensor are determined and stored in a table. The table is updated based on rules regarding a relationship between each calibration result and a current calibration value. The calibration results are weighted and combined to determine a combined calibration result. The calibration value for calibrating the sensor is selected from the calibration results, the combined calibration results, or the current calibration value based on a selection criteria.

Abstract: Determining contexts of mobile devices. Particular embodiments described herein include machines that determine two estimated positions of a mobile device that respectively correspond to first and second locations at first and second times, acquire sets of terrain or structural information for first and second areas that respectively include the first and second estimated positions, use the acquired sets of information and the estimated positions to determine if the mobile device was near or within a structure at the first and second times, determine one or more values that are indicative of vertical movement by the mobile device during a period of time between the first time and the second time, compare the one or more values to one or more threshold conditions, and determine a context of the mobile device based on the comparison.

Abstract: Determining calibration values for atmospheric sensors that provide measured pressures used for estimating altitudes of mobile devices. Particular systems and methods determine if any uncalibrated reference-level pressure estimates associated with an unstable pressure sensor should not be used when calibrating the unstable pressure sensor, and calibrate the unstable pressure sensor using all of the uncalibrated reference-level pressure estimates except any uncalibrated reference-level pressure estimate that should not be used when calibrating the unstable pressure sensor.

Abstract: A method involves determining, at a mobile device or a service, an uncertainty in height above a reference altitude, an estimated 2D position of the mobile device, and an uncertainty in terrain height above the reference altitude using the estimated 2D position. An uncertainty in height above terrain, of the mobile device, is determined at the mobile device or a server using the uncertainty in height above the reference altitude and the uncertainty in terrain height above the reference altitude.

Abstract: A wholistic activity context is used to determine whether to calibrate a barometric pressure sensor of a mobile device. A pair of activity transitions are determined from three activities of the mobile device. A time relationship and a position relationship between the activity transitions is determined. An opportunity to calibrate the barometric pressure sensor occurs between the activity transitions. A calibration of the barometric pressure sensor is performed in response to determining that the time relationship and the position relationship indicate that the wholistic activity context surrounding the opportunity to calibrate the barometric pressure sensor is conducive to calibration.

Abstract: Calibrating a pressure sensor of a mobile device includes determining a first plurality of calibration values for a first plurality of visits to a first revisit zone to which a mobile device repeatedly returns; determining a first relative calibration adjustment value based on the first plurality of calibration values; determining an adjusted absolute calibration value based on i) an absolute calibration value used to calibrate pressure measurements made by a pressure sensor of the mobile device, and ii) the first relative calibration adjustment value; and calibrating pressure measurements made by the pressure sensor of the mobile device using the adjusted absolute calibration value.

Abstract: Estimating a difference in height between floors in a building for use in estimating a height or an altitude of one of the floors. A height difference is estimated between a first floor and a second floor based on outdoor temperatures of first and second time periods, an indoor temperature of the first or second time period, and first and second estimated differences in height between the first and second floors that is based on measurements of pressure from mobile devices when the mobile devices were on the first and second floors during the first and second time periods.

Abstract: Using a pressure sensor of a mobile device to improve the accuracy of determined contexts. Particular embodiments described herein include machines that determine a context of a mobile device, and determine if the determined context is accurate using multiple measurements of pressure from a pressure sensor of the mobile device.

Abstract: A method to identify a problematic 2D position of a mobile device can include: determining a reported 2D position of the mobile device; determining a piece of information about the mobile device; and comparing the reported 2D position and the piece of information about the mobile device. Upon determining that the reported 2D position and the piece of information about the mobile device are consistent with each other, the reported 2D position of the mobile device is used as an estimate of the actual 2D position of the mobile device, or upon determining that the reported 2D position and the piece of information about the mobile device are not consistent with each other, the reported 2D position is determined to be problematic, and the reported 2D position of the mobile device is removed from a list of reported 2D positions of the mobile device.

Abstract: Extending the spatial coverage of a reference pressure network. Particular embodiments described herein include machines that initially determine that an estimated position of a mobile device resides outside a coverage area of a network of reference pressure sensors. Reference-level pressures corresponding to reference pressure sensors of the network are determined using measurements of pressure from the network of reference pressure sensors. A pressure pattern for a region that includes the estimated position and an area not included in the coverage area of the network is determined using reference-level pressures that were not determined using measurements of pressure from the network of reference pressure sensors. The reference-level pressures corresponding to reference pressure sensors of the network and the pressure pattern are used to determine a reference-level pressure value for use in calibrating a pressure sensor of the mobile device or for use in estimating an altitude of the mobile device.

Abstract: Calibrating an unstable sensor of a mobile device. Systems and methods for calibrating a sensor of a mobile device determine a first estimated position of the mobile device without using any measurement from the sensor of the mobile device, generate a second estimated position of the mobile device using a measurement from the sensor, estimate a sensor error of the sensor using the first estimated position and the second estimated position, and use the sensor error to determine a calibration value for adjusting one or more measurements from the sensor.

Abstract: Calibrating weather stations using maximum allowed altitude errors. Particular embodiments described herein include machines that determine a maximum allowed pressure calibration error for a weather station, determine a temperature variation associated with an environment in which the weather station resides, determine a maximum allowed altitude error for the weather station using the maximum allowed pressure calibration error and the temperature variation, and use the maximum allowed altitude error to determine if a first maximum possible altitude error associated with a first approach for estimating an altitude of the weather station exceeds the maximum allowed altitude error. If the first maximum possible altitude error associated with the first approach does not exceed the maximum allowed altitude error, the first approach is used to estimate an altitude of the weather station for use in calibrating a pressure sensor of the weather station.

Abstract: Multiple calibration results for calibrating a barometric pressure sensor based on data received from a device containing the sensor are determined and stored in a table. The table is updated based on rules regarding a relationship between each calibration result and a current calibration value. The calibration results are weighted and combined to determine a combined calibration result. The calibration value for calibrating the sensor is selected from the calibration results, the combined calibration results, or the current calibration value based on a selection criteria.

Abstract: Extending the spatial coverage of a reference pressure network. Particular embodiments described herein include machines that initially determine that an estimated position of a mobile device resides outside a coverage area of a network of reference pressure sensors. Reference-level pressures corresponding to reference pressure sensors of the network are determined using measurements of pressure from the network of reference pressure sensors. A pressure pattern for a region that includes the estimated position and an area not included in the coverage area of the network is determined using reference-level pressures that were not determined using measurements of pressure from the network of reference pressure sensors. The reference-level pressures corresponding to reference pressure sensors of the network and the pressure pattern are used to determine a reference-level pressure value for use in calibrating a pressure sensor of the mobile device or for use in estimating an altitude of the mobile device.

Abstract: Calibrating a pressure sensor of a mobile device incudes determining an absolute calibration value used to calibrate pressure measurements by a pressure sensor of a mobile device; determining a first revisit zone as a first location to which the mobile device repeatedly returns; determining first and second calibrations for first and second visits to the first revisit zone; determining a first relative calibration adjustment value based on a difference between the first and second calibrations; determining an adjusted absolute calibration value based on a sum of the absolute calibration value and the first relative calibration adjustment value; and estimating an altitude of the mobile device based on a pressure measurement by the pressure sensor and the adjusted absolute calibration value.