System and method for model-free position estimation and tracking

• US 20070139269A1
• Filed: 12/16/2005
• Published: 06/21/2007
• Est. Priority Date: 12/16/2005
• Status: Active Grant

• Current Assignee: Piece Future Pte., Ltd.
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Chen, Aiyou, Lambert, Diane, Harko, Cristina, Whiting, Philip
• Granted Patent: US 8,089,407 B2
• Time in Patent Office: Days
• US Class Current: 342/450
• CPC Class Codes: G01S 5/02528 Simulating radio frequency fingerprints

Abstract

A system for, and method of, position estimation of, or tracking, an object. In one embodiment, the system includes: (1) a map generator configured to create, without requiring a propagation model, a mean map for the region and (2) a measurement matching engine associated with the map generator and configured to match received measurements regarding the object to a number of nearest neighboring grid points in the mean map.

Claims

1. A system for position estimation of an object relative to a region, comprising:

a map generator configured to create, without requiring a propagation model, a mean map for said region; and

a measurement matching engine associated with said map generator and configured to configured to match received measurements regarding said object to a number of nearest neighboring grid points in said mean map.

2. The system as recited in claim 1 wherein said map generator is further configured to create, without requiring a propagation model, a standard deviation map for said region and said measurement matching engine is further configured to employ said standard deviation map to weight an average of said number of nearest neighboring grid points.

3. The system as recited in claim 1 wherein said mean map is a signal strength mean map.

4. The system as recited in claim 2 wherein said standard deviation map is a signal strength standard deviation map.

5. The system as recited in claim 1 wherein said nearest neighboring grid points are determined by generating vectors from moving averages of test measurements and determining distances between said vectors.

6. The system as recited in claim 1 further comprising a path smoother associated with said measurement matching engine and configured to make smooth static position estimates regarding said object without requiring a motion model.

7. The system as recited in claim 1 wherein said map generator is configured to create said mean map by interpolating sample means over a triangular partition of a given area, ignoring height.

8. The system as recited in claim 7 wherein said map generator is configured to create said mean map by linearly interpolating said sample means.

9. The system as recited in claim 2 wherein said map generator is configured to create said standard deviation map by interpolating an estimate of standard deviation over a partition of a given area, ignoring height.

10. The system as recited in claim 9 wherein said map generator is configured to create said standard deviation map by linearly interpolating said estimate of said standard deviation.

11. The system as recited in claim 6 wherein said path smoother is further configured to locally linearly smooth a moving window of said static position estimates.

12. A method of determining a position of an object in a region having a plurality of access points, comprising:

measuring a vector of components representing the position, each component being a characteristic of a corresponding one of the access points, the characteristic being either a transmission signal intensity of the corresponding access point at the position or a time delay for signals sent between the corresponding access point and the position; and

estimating the position of the object based on the measured vector, a set of reference vectors corresponding to grid points in the region, and error vectors corresponding to the reference vectors; and

wherein each component of each reference vector is an estimate of the value of the characteristic for one of the access points at the corresponding grid point; and

wherein each error vector provides estimates of errors in the corresponding reference vector.

13. The method as recited in claim 12 wherein the estimating includes evaluating a weighted average over a proper subset of the grid points, the proper subset being those grid points whose corresponding reference vectors are close to the measured vector.

14. The method as recited in claim 12 wherein the weighted average depends on components of the reference vectors and components of the error estimates.

15. The method as recited in claim 12 further comprising identifying those of the reference vectors that are at a smaller Euclidean distance from the measured vector in a space of said vectors than others of the reference vectors to determine the proper subset of the grid points.

16. The method as recited in claim 15 wherein the averaging including weighting each grid point by a corresponding weight that depends on the reference vector corresponding to the same grid point.

17. The method as recited in claim 16 wherein each weight depends on the error vector corresponding to the same grid point.

18. A method of tracking an object in a region having a plurality of access points, comprising:

measuring vectors corresponding to a sequence of positions of the object, each component of one of the vectors being a characteristic of a corresponding of one of the access points, the characteristic being either a transmission signal intensity of the corresponding access point at the position corresponding to the one of the vectors or a time delay for transmitted a signal between the corresponding access point and the position corresponding to the one of the vectors;

at each position of the sequence, estimating a position of the object from the measured vectors corresponding to a number of nearby ones of the positions; and

wherein the estimating steps are performed such that the number of nearby ones of the positions varies with the position being estimated.

19. The method as recited in claim 18 wherein the estimating steps are performed such that the number of nearby ones of the positions depends on the speed of the object.

20. The method as recited in claim 18 wherein the estimating steps are performed such that the number of nearby ones of the positions decreases as the speed of the object increases.

21. The method as recited in claim 18 wherein the estimating steps are performed such that the number of nearby ones of the positions depends on estimated errors of the measured vectors at the position being estimated.

22. The method as recited in claim 18 wherein the estimating steps are performed such that the number of nearby ones of the positions increases as the estimated errors of the measured vectors at the positions being estimated increase.

23. The method as recited in claim 18 further comprising determining a track of the object from the estimated positions by a method, the method including weighting the estimated positions to determine the track.

24. The method as recited in claim 23 wherein the method includes performing a loess smoothing of the sequence formed by the estimated positions.