Sequence-based positioning technique

• US 20050136944A1
• Filed: 11/30/2004
• Published: 06/23/2005
• Est. Priority Date: 05/31/2002
• Status: Active Grant

• Current Assignee: AiRISTA Flow, Inc. (AiRISTA, LLC), AiRISTA International Oy (AiRISTA, LLC)
• Original Assignee: Ekahau Oy
• Inventors: Misikangas, Pauli, Myllymaki, Petri
• Granted Patent: US 7,349,683 B2
• Time in Patent Office: Days
• US Class Current: 455/456.100
• CPC Class Codes: G01S 5/02521 using a radio-map
• CPC Class Codes: H04W 64/00 Locating users or terminals or network equipment for network management purposes, e.g. mobility management

Abstract

A target device'"'"'s location is estimated by a location estimation module (LEM) that comprises a probabilistic model (PM) for a plurality of sample points, each of which comprises a sample location and an expected distribution of signal values at that sample point. The location estimation module (LEM) makes a sequence (OS) of observations o_n, n=1, 2, 3 . . . , of signal values. Each observation corresponds to a respective location q_n along the target device'"'"'s path. The sequence of observations and the respective location constitute a hidden Markov model. The module estimates the target device'"'"'s location q_n based on the probabilistic model (PM) and the sequence of observations, wherein the sequence of observations comprises one or more future observations o_n+m for which m is a positive integer. In other words, the target device'"'"'s location is estimated, at least partially, based on knowledge of its future.

Claims

1. A method for estimating a target device'"'"'s location, the method comprising:

maintaining a probabilistic model for a plurality of sample points, each sample point comprising a sample location and an expected distribution of signal values at that sample point;

making a sequence of observations o_n, n=1 . . . N, of signal values wherein each observation corresponds to a respective location q_n along the target device'"'"'s path, wherein the sequence of observations and the respective location constitute a hidden Markov model;

estimating the target device'"'"'s location q_t based on the probabilistic model and the sequence of observations, wherein the sequence of observations comprises one or more future observations o_t+m for which m is a positive integer, such that t+m<

N.

2. A method according to claim 1, wherein the sequence of observations comprises one or more past observations o_n−

m.

3. A method according to claim 2, further comprising estimating the target device'"'"'s location q_n based on:

a forward recursion of the one or more past observations; and

a backward recursion of the one or more future observations.

4. A method according to claim 1, further comprising estimating the target device'"'"'s transition probabilities between the sample points and estimating the target device'"'"'s location partially based on the transition probabilities.

5. A method according to claim 2, further comprising estimating the transition probabilities and storing them with the probabilistic model before making the sequence of observations.

6. A method according to claim 1, further comprising performing the location-estimating step in the target device.

7. A method according to claim 1, further comprising performing the location-estimating step in a fixed equipment to which the target device reports the sequence of observations via a radio network.

8. A method according to claim 1, wherein the at least one measurable signal value comprises signal strength.

9. A method according to claim 1, wherein the at least one measurable signal value comprises bit error rate or ratio.

10. A location estimation module for estimating a target device'"'"'s location, the location estimation module comprising:

a probabilistic model for a plurality of sample points, each sample point comprising a sample location and an expected distribution of signal values at that sample point;

means for making a sequence of observations o_n, n=1 . . . N, of signal values, wherein each observation corresponds to a respective location q_n along the target device'"'"'s path, wherein the sequence of observations and the respective location constitute a hidden Markov model;

means for estimating the target device'"'"'s location q_t based on the probabilistic model and the sequence of observations, wherein the sequence of observations comprises one or more future observations o_t+m for which m is a positive integer, such that t+m≦

N.

11. A location estimation module according to claim 10, wherein the sequence of observations comprises one or more past observations o_n−

m.

12. A location estimation module according to claim 11, further comprising means for estimating the target device'"'"'s location q_n based on:

a forward recursion of the one or more past observations; and

a backward recursion of the one or more future observations.

13. A location estimation module according to claim 10, further comprising means for estimating the target device'"'"'s transition probabilities between the sample points and for estimating the target device'"'"'s location partially based on the transition probabilities.

14. A location estimation module according to claim 10, further comprising means for estimating the transition probabilities and storing them with the probabilistic model before making the sequence of observations.

15. A location estimation module according to claim 10, wherein the location estimation module is configured to be located in the target device.

16. A location estimation module according to claim 10, wherein the location estimation module is configured to be located in a fixed equipment to which the target device reports the sequence of observations via a radio network.

17. A location estimation module according to claim 10, wherein the at least one measurable signal value comprises signal strength.

18. A location estimation module according to claim 10, wherein the at least one measurable signal value comprises bit error rate or ratio.