Location determination using fingerprint data
Methods, program products, and systems for using a location fingerprint database to determine a location of a mobile device are described. A mobile device can use location fingerprint data received from a server to determine a location of the mobile device at the venue. The mobile device can obtain, from a sensor of the mobile device, a vector of sensor readings, each sensor reading can measure an environment variable, e.g., a signal received by the sensor from a signal source. The mobile device can perform a statistical match between the vector and the location fingerprint data. The mobile device can then estimate a current location of the mobile device based on the statistical match.
1. A method comprising:
receiving, by a mobile device and from a server, location fingerprint data, the location fingerprint data comprising a plurality of locations at a venue and a fingerprint of each location, the venue comprising a space accessible by a pedestrian and one or more constraints of movement of the pedestrian, each fingerprint comprising one or more expected measurements of an environment variable at the corresponding location;
receiving a measurement from one or more sensors of the mobile device, the measurement comprising one or more readings of the one or more sensors on the environment variable; and
determining a location of the mobile device at the venue using the measurement and the location fingerprint data, including performing a statistical classification categorizing the measurement using a particle filter, wherein performing the statistical classification comprises:
determining, from the location fingerprint data, a plurality of vectors, each vector corresponding to a possible location and including expected measurements of the environment variable at the corresponding possible location, each possible location corresponding to a category, each possible location being designated as a particle in the particle filter;
determining a probability density that includes a respective probability that the mobile device is located at each of the possible locations by applying a function to a difference between the measurement and each of the vectors;
classifying the measurement into a category according to the probability density; and
designating a possible location corresponding to the category as the location of the mobile device.
2. The method of claim 1 , wherein:
each environment variable comprises a signal from a signal source detectable by the mobile device, the one or more signal sources include at least one of a wireless access point, a light source, a sound source, a magnetic source, or a heat source; and
the one or more expected measurements include at least one of a received signal strength indication (RSSI), a round-trip time, a magnetic field strength, a temperature, a noise level, or an air pressure level.
3. The method of claim 1 , wherein receiving the location fingerprint data comprises:
requesting, by the mobile device, the location fingerprint data from the server when the mobile device determines that the mobile device is located in proximity to the venue based on a threshold; and
receiving the location fingerprint data in response.
4. The method of claim 1 , wherein determining the probability density comprises:
determining a point-wise kernel estimation of a respective probability that the measurement is measured at each of the possible locations, the point-wise kernel estimation having a kernel function, the kernel function being applied to a difference between the measurement and a vector corresponding to the respective possible location, the kernel function being a symmetric function that integrates to one.
5. The method of claim 4 , wherein the kernel function includes a Gaussian function.
6. The method of claim 4 , wherein the kernel function is applied to an a priori domain provided by the particle filter.
7. The method of claim 1 , wherein determining the location of the mobile device comprises aggregating statistics of a group of possible locations.
8. The method of claim 7 , wherein determining the location of the mobile device comprises calculating the location using the group of locations based on a context, the context including one or more readings of at least one of an accelerometer, a magnetic sensor, a gyroscope, or a barometer.
9. A system comprising:
a mobile device;
a non-transitory computer-readable medium storing instructions operable to cause the mobile device to perform operations comprising:
receiving, by a mobile device and from a server, location fingerprint data, the location fingerprint data comprising a plurality of locations at a venue and a fingerprint of each location, the venue comprising a space accessible by a pedestrian and one or more constraints of movement of the pedestrian, each fingerprint comprising one or more expected measurements of an environment variable at the corresponding location;
receiving a measurement from one or more sensors of the mobile device, the measurement comprising one or more readings of the one or more sensors on the environment variable; and
determining a location of the mobile device at the venue using the measurement and the location fingerprint data, including performing a statistical classification categorizing the measurement using a particle filter, wherein performing the statistical classification comprises:
determining, from the location fingerprint data, a plurality of vectors, each vector corresponding to a possible location and including expected measurements of the environment variable at the corresponding possible location, each possible location corresponding to a category, each possible location being designated as a particle in the particle filter;
determining a probability density that includes a respective probability that the mobile device is located at each of the possible locations by applying a function to a difference between the measurement and each of the vectors;
classifying the measurement into a category according to the probability density; and
designating a possible location corresponding to the category as the location of the mobile device.
10. The system of claim 9 , wherein:
each environment variable comprises a signal from a signal source detectable by the mobile device, the one or more signal sources include at least one of a wireless access point, a light source, a sound source, a magnetic source, or a heat source; and
the one or more expected measurements include at least one of a received signal strength indication (RSSI), a round-trip time, a magnetic field strength, a temperature, a noise level, or an air pressure level.
11. The system of claim 9 , wherein receiving the location fingerprint data comprises:
requesting, by the mobile device, the location fingerprint data from the server when the mobile device determines that the mobile device is located in proximity to the venue based on a threshold; and
receiving the location fingerprint data in response.
12. The system of claim 9 , wherein determining the probability density comprises:
determining a point-wise kernel estimation of a respective probability that the measurement is measured at each of the possible locations, the point-wise kernel estimation having a kernel function, the kernel function being applied to a difference between the measurement and a vector corresponding to the respective possible location, the kernel function being a symmetric function that integrates to one.
13. The system of claim 12 , wherein the kernel function includes a Gaussian function.
14. The system of claim 12 , wherein the kernel function is applied to an a priori domain provided by the particle filter.
15. The system of claim 9 , wherein determining the location of the mobile device comprises aggregating statistics of a group of possible locations.
16. The system of claim 15 , wherein determining the location of the mobile device comprises calculating the location using the group of locations based on a context, the context including one or more readings of at least one of an accelerometer, a magnetic sensor, a gyroscope, or a barometer.
17. A non-transitory computer-readable medium storing instructions operable to cause one or more processors to perform operations comprising:
receiving, by a mobile device and from a server, location fingerprint data, the location fingerprint data comprising a plurality of locations at a venue and a fingerprint of each location, the venue comprising a space accessible by a pedestrian and one or more constraints of movement of the pedestrian, each fingerprint comprising one or more expected measurements of an environment variable at the corresponding location;
receiving a measurement from one or more sensors of the mobile device, the measurement comprising one or more readings of the one or more sensors on the environment variable; and
determining a location of the mobile device at the venue using the measurement and the location fingerprint data, including performing a statistical classification categorizing the measurement using a particle filter, wherein performing the statistical classification comprises:
determining, from the location fingerprint data, a plurality of vectors, each vector corresponding to a possible location and including expected measurements of the environment variable at the corresponding possible location, each possible location corresponding to a category, each possible location being designated as a particle in the particle filter;
determining a probability density that includes a respective probability that the mobile device is located at each of the possible locations by applying a function to a difference between the measurement and each of the vectors;
classifying the measurement into a category according to the probability density; and
designating a possible location corresponding to the category as the location of the mobile device.
18. The non-transitory computer-readable medium of claim 17 , wherein:
each environment variable comprises a signal from a signal source detectable by the mobile device, the one or more signal sources include at least one of a wireless access point, a light source, a sound source, a magnetic source, or a heat source; and
the one or more expected measurements include at least one of a received signal strength indication (RSSI), a round-trip time, a magnetic field strength, a temperature, a noise level, or an air pressure level.
19. The non-transitory computer-readable medium of claim 17 , wherein receiving the location fingerprint data comprises:
requesting, by the mobile device, the location fingerprint data from the server when the mobile device determines that the mobile device is located in proximity to the venue based on a threshold; and
receiving the location fingerprint data in response.
20. The non-transitory computer-readable medium of claim 17 , wherein determining the probability comprises:
determining a point-wise kernel estimation of a respective probability that the measurement is measured at each of the possible locations, the point-wise kernel estimation having a kernel function, the kernel function being applied to a difference between the measurement and a vector corresponding to the respective possible location, the kernel function being a symmetric function that integrates to one.
21. The non-transitory computer-readable medium of claim 20 , wherein the kernel function includes a Gaussian function.
22. The non-transitory computer-readable medium of claim 20 , wherein the kernel function is applied to an a priori domain provided by the particle filter.
23. The non-transitory computer-readable medium of claim 17 , wherein determining the location of the mobile device comprises aggregating statistics of a group of possible locations.
24. The non-transitory computer-readable medium of claim 23 , wherein determining the location of the mobile device comprises calculating the location using the group of locations based on a context, the context including one or more readings of at least one of an accelerometer, a magnetic sensor, a gyroscope, or a barometer.