Location detection system and method

• US 9,658,312 B2
• Filed: 08/26/2014
• Issued: 05/23/2017
• Est. Priority Date: 08/26/2013
• Status: Active Grant

• Current Assignee: Dublin City University
• Original Assignee: Dublin City University
• Inventors: Collins, Kevin, McGuinness, Kevin Anthony, O'Connor, Noel Edward
• Primary Examiner(s): Gonzalez, Amancio
• Application Number:
  US14/914,942
• Publication Number:
  US 20160202342A1
• Time in Patent Office: 1,001 Days
• Field of Search: 4554562
• CPC Class Codes: G01S 5/02521 using a radio-map
• CPC Class Codes: G01S 5/0278 involving statistical or probabilistic considerations G01S5/0252, G01S5/0294 take precedence
• CPC Class Codes: H04B 17/318 Received signal strength
• CPC Class Codes: H04L 43/045 for graphical visualisation of monitoring data
• CPC Class Codes: H04W 24/10 Scheduling measurement reports ; Arrangements for measurement reports
• CPC Class Codes: H04W 4/021 Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences
• CPC Class Codes: H04W 64/00 Locating users or terminals or network equipment for network management purposes, e.g. mobility management

Abstract

A system and method for detecting the location of a mobile device within an area, uses a plurality of wireless signals sent to or received by the device to derive a first estimate of the position of the device. This position estimate is then adjusted using a probabilistic model of the area, where the area is divided into zones or regions, and where each region is assigned a weighting according to the probability of the device being located in that region taking account of the prior position of the device. By using the weightings to adjust the position estimate, false or ambiguous position estimates are discounted and the true position of the device is more reliably detected. The method also includes techniques to take account of different wireless characteristics of different devices when each is supplied with the same model.

Claims

1. A method for detecting the location of a mobile device within an area, comprising the steps of:

storing a prior location estimate for the device;

determining, for each of a plurality of zones in said area, a signal-based value indicative of the likelihood of the device being located in the respective zone, based on measured wireless signal characteristics;

maintaining a set of transition values, each transition value representing a likelihood of a device moving between a pair of said zones;

calculating, for each of a plurality of zones in said area, a current location value by using a matrix-based approach to combine (i) the prior location estimate, (ii) said set of transition values, and (iii) said signal-based values;

wherein said set of transition values includes at least one transition value representing the likelihood of the device moving between a pair of zones which are physically accessible from one another and at least one transition value representing the likelihood of the device moving between a pair of zones which are physically inaccessible from one another; and

wherein said step of calculating a current location value includes;

using a discriminative model to calculate a current location value for at least one zone physically accessible from a zone corresponding to the prior location estimate, and a current location value for at least one zone physically inaccessible from said zone corresponding to the prior location estimate; and

determining whether the device is in a boundary region adjacent to the meeting of two or more zones, wherein for each candidate location Li of n location candidates L1, L2, L3, . . . , Ln, a probability value P(Li) is assigned according to the likelihood of the device being in that candidate location, and wherein the determination that the device is in a boundary region between two locations Li, Lj is made where the following conditions are met;

P(L_i|s)>

P(L_j|s)>

P(L_k|s), for i, j, k=1, 2, 3, . . . , n, k≠

j≠

i, and P(L_i|s)−

P(L_j|s)<

c, where c is a tuneable parameter and c<

1, and s is the observed signal strength vector on the mobile device.

2. The method of claim 1, wherein in a system of N zones, the prior location estimate comprises a set of N values (p1, p2, . . . pN);

wherein the set of transition values comprises a set of N² values expressible as a N×

N transition matrix T, and wherein the signal-based values comprise a set of N values (s1, s2, . . . sN); and

wherein said step of calculating, for each of a plurality of zones in said area, a current location value comprises obtaining the matrix product of a 1×

N matrix having elements [(p1×

s1) (p2×

s2) . . . (pN×

sN)] and the transition matrix T, resulting in a 1×

N location matrix whose elements e1, e2, . . . eN are indicative of the likelihood of the device being in each of the N zones respectively.

3. The method of claim 2, wherein said transition matrix T is doubly stochastic.

4. The method of claim 1, further comprising the step of storing one or more of said calculated current location values, and on a further iteration, using the stored current location values as the prior location estimate.

5. The method of claim 4, wherein on an initial iteration, the prior location estimate is initialised to a stored default value or set of values.

6. The method of claim 1, wherein the prior location estimate is stored as a set of values, each associated with a respective one of the plurality of zones, representing the likelihood of the device having been in the respective zone on a prior iteration of the method.

7. The method of claim 1, wherein said set of transition values is calculated from a representation of said zones as a graph of vertices and edges, each zone being represented as a vertex and being connected to at least one other vertex by an edge.

8. The method of claim 7, wherein said set of transition values is calculated such that each transition value is a function of the distance between a pair of vertices in the graph.

9. The method of claim 7, wherein the transition values between pairs of vertices are weighted.

10. The method of claim 9, wherein the weighting is assigned based on one or more of:

physical distance between zones, physical barriers between zones, traffic patterns between zones, and time to transition between zones.

11. The method of claim 7, wherein each transition value is also a function of a tuneable parameter.

12. The method of claim 11, further comprising the steps of modifying said tuneable parameter and adjusting the transition values in consequence.

13. The method of claim 12, wherein said tuneable parameter is modified in accordance with a calculated or estimated speed of movement of the device.

14. The method of claim 12, wherein said tuneable parameter is modified in accordance with a calculated or estimated congestion value associated with movement within said area.

15. The method of claim 12, wherein said tuneable parameter is modified in accordance with a value associated with the time between successive iterations of the method.

16. The method of claim 1, wherein the step of determining signal-based values comprises measuring, by the device, a plurality of wireless signal strengths, and comparing said measured signal strengths with a set of reference signal strengths associated with each of the plurality of zones.

17. The method of claim 16, wherein the measured signal strengths are adjusted before comparison using one or more of the following transformations:

a. in the event of no signal being detected from a previously detected wireless signal source, replacing a null value for said source with a previously measured value;

b. normalising the signal strength values to a common range;

c. shifting the signal strength values to reduce differences between wireless characteristics of the mobile device and those of a reference device from which the reference signal strengths were derived; and

d. calculating a standard z-score as follows;

18. The method of claim 17, comprising selecting device specific parameters for the transformations according to the specific type of mobile device used to measure the signal strengths.

19. The method of claim 18, wherein the device specific parameters are obtained by analysing measured signal strengths of the specific type of mobile device used.

20. The method of claim 1, wherein said step of calculating a current location value comprises adjusting the signal-based value for each zone according to a weighting derived from the prior location estimate, said weighting indicating the likelihood of the device having been in that zone on a previous iteration, resulting in an enhanced signal-based value for each of said zones;

and combining the enhanced signal-based value for each zone with the transition values for each zone pair involving that zone, to result in a current location value for each zone.