Fingerprint data pre-process method for improving localization model
View Patent ↗Provided is a computer-implemented method including acquiring a first pieces of observation data that include a position of a mobile object and a received signal strength of a wireless signal observed by the mobile object; and correcting each position of the mobile object included in each piece of observation data of the first pieces of observation data using one position of the mobile object at a time before the received signal strength included in the piece of observation data is observed.
1. A computer-implemented method comprising:
acquiring a received signal strength distribution that includes a received signal strength of a wireless signal of a mobile computing device at each of a plurality of positions;
generating a received signal strength distribution in which a distribution of the plurality of positions included in the received signal strength distribution have been made uniform, by iteratively extracting, by importance sampling, each received signal strength corresponding to a sequentially changing number of observation positions among the plurality of positions from the received signal strength distribution, the generating further comprising selecting, using combinatorial optimization, a plurality of the plurality of positions such that an index value indicating non-uniformity is smallest; and
correcting, for each observation timing of the each received signal strength, each respective position of the mobile computing device from the plurality of positions included in the received signal strength distribution.
2. The computer-implemented method according to claim 1 , further comprising:
generating a model for estimating a position from a received signal strength, using the received signal strength distribution in which the distribution of the plurality of positions has been made uniform.
3. The computer-implemented method according to claim 1 , further comprising:
generating a probability density distribution indicating the distribution of the plurality of positions, wherein the extracting includes performing resampling of each of the plurality of positions according to a probability that has an inverse correlation relationship with a probability resulting from the probability density distribution.
4. The computer-implemented method according to claim 3 , wherein the extracting includes performing resampling of each of the plurality of positions with a probability that is inversely proportional to the probability resulting from the probability density distribution.
5. The computer-implemented method according to claim 1 , further comprising:
selecting the received signal strength distribution whose index value indicating non-uniformity is smallest, from among a plurality of the received signal strength distributions obtained using different ratios for the positions extracted from among the plurality of positions.
6. The computer-implemented method according to claim 5 , wherein the index value indicating the non-uniformity is a discrepancy.
7. The computer-implemented method according to claim 1 , wherein the some of the positions among the plurality of positions from the received signal strength distribution includes observation data for which a position of a mobile object has been corrected based on the extracting.
8. A computer program product for generating a received signal strength distribution, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a device to cause the device to perform operations comprising:
acquiring a received signal strength distribution that includes a received signal strength of a wireless signal of a mobile computing device at each of a plurality of positions;
generating a received signal strength distribution in which a distribution of the plurality of positions included in the received signal strength distribution have been made uniform, by iteratively extracting, by importance sampling, each received signal strength corresponding to a sequentially changing number of observation positions among the plurality of positions from the received signal strength distribution, the generating further comprising selecting, using combinatorial optimization, a plurality of the plurality of positions such that an index value indicating non-uniformity is smallest; and
correcting, for each observation timing of the each received signal strength, each respective position of the mobile computing device from the plurality of positions included in the received signal strength distribution.
9. The computer program product according to claim 8 , wherein the program instructions further cause the device to perform operations comprising:
generating a model for estimating a position from a received signal strength, using the received signal strength distribution in which the distribution of the plurality of positions has been made uniform.
10. The computer program product according to claim 8 , wherein the program instructions further cause the device to perform operations comprising:
generating a probability density distribution indicating the distribution of the plurality of positions, wherein the extracting includes performing resampling of each of the plurality of positions according to a probability that has an inverse correlation relationship with a probability resulting from the probability density distribution.
11. The computer program product according to claim 10 , wherein the extracting includes performing resampling of each of the plurality of positions with a probability that is inversely proportional to the probability resulting from the probability density distribution.
12. The computer program product according to claim 8 , further comprising:
selecting the received signal strength distribution whose index value indicating non-uniformity is smallest, from among a plurality of the received signal strength distributions obtained using different ratios for the positions extracted from among the plurality of positions.
13. The computer program product according to claim 12 , wherein the index value indicating the non-uniformity is a discrepancy.
14. The computer program product according to claim 8 , wherein the some of the positions among the plurality of positions from the received signal strength distribution includes observation data for which a position of a mobile object has been corrected based on the extracting.
15. A system for generating a received signal strength distribution, comprising:
a receiver configured for acquiring a received signal strength distribution that includes a received signal strength of a wireless signal of a mobile computing device at each of a plurality of positions; and
a processor device operatively coupled to a computer-readable storage medium, the processor device being configured for:
generating a received signal strength distribution in which a distribution of the plurality of positions included in the received signal strength distribution have been made uniform, by iteratively extracting, by importance sampling, each received signal strength corresponding to a sequentially changing number of observation positions among the plurality of positions from the received signal strength distribution, the generating further comprising selecting, using combinatorial optimization, a plurality of the plurality of positions such that an index value indicating non-uniformity is smallest; and
correcting, for each observation timing of the each received signal strength, each respective position of the mobile computing device from the plurality of positions included in the received signal strength distribution.
16. The system according to claim 15 , wherein the processor device is further configured for generating a model for estimating a position from a received signal strength, using the received signal strength distribution in which the distribution of the plurality of positions has been made uniform.
17. The system according to claim 15 , wherein the processor device is further configured for generating a probability density distribution indicating the distribution of the plurality of positions, wherein the extracting includes performing resampling of each of the plurality of positions according to a probability that has an inverse correlation relationship with a probability resulting from the probability density distribution.
18. The system according to claim 17 , wherein the extracting includes performing resampling of each of the plurality of positions with a probability that is inversely proportional to the probability resulting from the probability density distribution.
19. The system according to claim 15 , wherein the processor device is further configured for selecting the received signal strength distribution whose index value indicating non-uniformity is smallest, from among a plurality of the received signal strength distributions obtained using different ratios for the positions extracted from among the plurality of positions.
20. The system according to claim 15 , wherein the some of the positions among the plurality of positions from the received signal strength distribution includes observation data for which a position of a mobile object has been corrected based on the extracting.