Hybrid location using pattern recognition of location readings and signal strengths of wireless access points
A query device scans radio frequencies for visible transmitting devices. The querying device receives at least a signal strength and identifier information associated with each of the transmitting devices. The list of visible devices is used to query a database containing location information for a plurality of visible devices. The list may be sent to a locationing system that may perform a location analysis on the resulting data to return a location to the query device. The weighted average of the locations returned in the database query may be computed to determine the location of the querying device, with the weight for each of the locations being the current signal strength detected by the querying device. Neural network analysis may also be used to determine the location of the querying device. Learning and seeding operations many also be used to populate the database with location information for transmitting devices.
1. A computer-implemented method for training a pattern recognition system for determining a location of a user device, comprising the steps of:
traversing one or more regions with a device configured to communicate with one or more transmitting devices;
scanning one or more radio frequencies for one or more visible transmitting devices within a region among the one or more regions;
receiving a signal strength pattern and identifiers for the one or more visible transmitting devices within the region;
querying a database containing location points for a plurality of visible transmitting devices by using the identifiers;
performing a weighted average analysis of those location points for each identifier to determine an average of those location points for each identifier of each visible transmitting device having location points within the database, resulting in a set of weighted averages, wherein a weight assigned to each location point is based on the signal strength for the visible transmitting device corresponding to that location point;
mapping the signal strength pattern and the set of weighted averages to the region to create a learning example for the pattern recognition system;
repeating the steps of scanning, receiving, querying, weighted averaging, and mapping to create a plurality of learning examples;
training the pattern recognition system using the plurality of learning examples; and
training additional pattern recognition systems using the plurality of learning examples.
2. The method as recited in claim 1 , wherein the one or more regions include a GPS coordinate, one or more rooms, one or more halls, one or more floors within a building, one or more buildings, one or more outdoor areas, or a combination of these regions.
3. The method as recited in claim 1 , wherein the one or more radio frequencies includes one or more wireless channels, and wherein the step of scanning includes the step of scanning wireless channels at one or more points within the region.
4. The method as recited in claim 1 , wherein the one or more radio frequencies includes one or more wireless channels, and wherein the step of scanning includes the step of scanning wireless channels over a period of time with a processor.
5. The method as recited in claim 1 , wherein the step of scanning includes the steps of:
enabling a user to initiate the step of scanning;
scanning the one or more radio frequencies at a cadence with a processor; and
enabling the user to terminate the step of scanning.
6. The method as recited in claim 1 , further comprising the step of storing the signal strength pattern and location information for each visible transmitting device in the database, and wherein a number of inputs of the pattern recognition system is based on a number of unique visible transmitting devices stored in the database.
7. The method as recited in claim 1 , wherein the pattern recognition system includes a neural network.
8. The method as recited in claim 1 , further comprising the steps of selecting the pattern recognition algorithm among the additional pattern recognition algorithms with a lowest error level after training.
9. The method as recited in claim 1 , further comprising the step of retraining the pattern recognition system after a predetermined period of time.
10. The method as recited in claim 1 , further comprising the step of retraining the pattern recognition system after a number of records greater than a threshold have been added to a database.
11. The method as recited in claim 1 , further comprising the step of enabling a user to enter a text label associated with the region.
12. The method as recited in claim 1 , further comprising the step of retraining the pattern recognition system after a threshold error level is passed.
13. The method as recited in claim 1 , further comprising the step of selecting a determined location by a pattern recognition algorithm among a plurality of pattern recognition algorithms with the highest confidence level as a final location of a querying device.
14. The method as recited in claim 1 , further comprising the step of selecting a location output the most often determined by two or more pattern recognition algorithms among a plurality of pattern recognition algorithms as a final location of a querying device.
15. The method as recited in claim 1 , further comprising the step of weighing a plurality of outputs from a plurality of pattern recognition algorithms by an amount associated with a confidence level or with a level of preference.
16. The method as recited in claim 15 , further comprising the steps of:
averaging the plurality of outputs from the plurality of pattern recognition algorithms using a weighted average; and
using the resulting average presented as a final location of a querying device.
17. The method as recited in claim 1 , further comprising the step of using only those pattern recognition algorithms among a plurality of pattern recognition algorithms that achieve an error level less than a threshold during the training phase.
18. A system for training a pattern recognition system to determine a location of a device, comprising:
a computer processor;
a memory in communication with the computer processor when the system is activated, the memory comprising computer readable instructions that upon execution by the computer processor cause the system to:
scan one or more radio frequencies for one or more visible transmitting devices within a region among one or more regions as the one or more regions are traversed with a device configured to communicate with one or more transmitting devices;
receive a signal strength pattern and identifiers for the one or more visible transmitting devices within the region;
query a database containing location points for a plurality of visible transmitting devices by using the identifiers;
perform a weighted average analysis of those location points for each identifier to determine an average of those location points for each identifier of each visible transmitting device having location points within the database, resulting in a set of weighted averages, wherein a weight assigned to each location point is based on the signal strength for the visible transmitting device corresponding to that location point;
map the signal strength pattern and the set of weighted averages to the region to create a learning example for the pattern recognition system;
create a plurality of learning examples by repeating the steps of scanning, receiving, querying, weighted averaging, and mapping;
train the pattern recognition system using the plurality of learning examples; and
train additional pattern recognition systems using the plurality of learning examples.