Minimum variance location estimation in wireless networks

• US 20080080429A1
• Filed: 10/03/2006
• Published: 04/03/2008
• Est. Priority Date: 10/03/2006
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Hart, Brian Donald
• Granted Patent: US 7,616,555 B2
• Time in Patent Office: Days
• US Class Current: 370/332
• CPC Class Codes: H04B 17/318 Received signal strength
• CPC Class Codes: H04W 64/00 Locating users or terminals or network equipment for network management purposes, e.g. mobility management
• CPC Class Codes: H04W 84/12 WLAN [Wireless Local Area Networks]

Abstract

In one embodiment, a method includes receiving received signal strength data, computing an aggregate square error surface based on the received signal strength data, computing a probability surface by applying a probability density function to the aggregate error surface, and computing a mean location of a wireless node.

Claims

1. Logic encoded in one or more tangible media for execution and when executed operable to:

collect received signal strength data;

compute an aggregate square error surface based on the received signal strength data;

compute a probability surface by computing a probability density function from the aggregate error surface; and

compute a mean location of a wireless node based on the probability surface.

2. The logic of claim 1 wherein the aggregate error surface characterizes a difference between the received signal strength values and expected received signal strength values.

3. The logic of claim 1 wherein the probability surface is computed based on the following expression:

$\mathrm{Pu}<br><br>\left(x,y\right)=\exp <br><br>\left(-\frac{1}{2<br><br>o^2}<br><br>S<br><br>\left(x,y\right)\right),$

wherein e is Euler'"'"'s number, σ

is a standard deviation of residual errors, and S represents a value of an error surface at a given location bin (x,y).

4. The logic of claim 1 wherein the probability surface is normalized such that a sum of the probabilities over an entire surface equals one.

5. The logic of claim 4 wherein the normalized probability surface is computed based on the following expression:

P(x,y)=Pu(x,y)/Σ

Pu(x,y).

15. The system of claim 4 wherein the normalized probability surface is computed based on the following expression:

P(x,y)=Pu(x,y)/Σ

Pu(x,y).

6. A method comprising:

collecting received signal strength data;

computing an aggregate square error surface based on the received signal strength data;

computing a probability surface by computing a probability density function from the aggregate error surface; and

computing a mean location of a wireless node based on the probability surface.

7. The method of claim 6 wherein the aggregate error surface characterizes a difference between the received signal strength values and expected received signal strength values.

8. The method of claim 6 wherein the probability surface is computed based on the following expression:

$\mathrm{Pu}<br><br>\left(x,y\right)=\exp <br><br>\left(-\frac{1}{2<br><br>o^2}<br><br>S<br><br>\left(x,y\right)\right),$

wherein e is Euler'"'"'s number, σ

is a standard deviation of residual errors, and S represents a value of an error surface at a given location bin (x,y).

9. The method of claim 6 wherein the probability surface is normalized such that a sum of the probabilities over an entire surface equals one.

10. The method of claim 9 wherein the normalized probability surface is computed based on the following expression:

P(x,y)=Pu(x,y)/Σ

Pu(x,y).

11. A system comprising:

one or more infrastructure radio transceivers; and

a wireless network infrastructure node operable to collect received signal strength data sensed by the one or more infrastructure radio transceivers;

compute an aggregate square error surface based on the received signal strength data;

compute a probability surface by computing a probability density function from the aggregate error surface; and

compute a mean location of a wireless node based on the probability surface.

12. The system of claim 11 wherein the aggregate error surface characterizes a difference between the received signal strength values and expected received signal strength values.

13. The system of claim 11 wherein the probability surface is computed based on the following expression:

$\mathrm{Pu}<br><br>\left(x,y\right)=\exp <br><br>\left(-\frac{1}{2<br><br>o^2}<br><br>S<br><br>\left(x,y\right)\right),$

wherein e is Euler'"'"'s number, σ

is a standard deviation of residual errors, and S represents a value of an error surface at a given location bin (x,y).

14. The system of claim 11 wherein the probability surface is normalized such that a sum of the probabilities over an entire surface equals one.

16. Logic encoded in one or more tangible media for execution and when executed operable to:

collect received signal strength data for a wireless node;

compute a plurality of aggregate square error surfaces, wherein one of the plurality of aggregate square error surfaces corresponds to a nominal transmission power, and remaining aggregate square error surfaces each correspond to at least one transmission power above or below the nominal transmission power;

compute an unnormalized probability surface for each of the aggregate square error surfaces;

compute normalized probability surfaces for each of the unnormalized probability surfaces relative to all the probability surfaces;

compute an aggregate probability surface across the normalized probability surfaces at each location bin; and

compute a mean x- and y-locations to yield an estimated location of the wireless node.

17. The logic of claim 16 wherein the plurality of aggregate square error surfaces are each composed of one or more error surfaces that characterize differences between the received signal strength values and expected received signal strength values at one or more locations.

18. The logic of claim 16 wherein the probability surface is computed based on the following expression:

$\mathrm{Pu}<br><br>\left(x,y\right)=\exp <br><br>\left(-\frac{1}{2<br><br>o^2}<br><br>S<br><br>\left(x,y\right)\right),$

wherein e is Euler'"'"'s number, σ

is a standard deviation of residual errors, and S represents a value of an error surface at a given location bin (x,y).

19. A method comprising:

collecting received signal strength data for a wireless node;

computing a plurality of aggregate square error surfaces, wherein one of the plurality of aggregate square error surfaces corresponds to a nominal transmission power, and remaining aggregate square error surfaces each correspond to at least one transmission power above or below the nominal transmission power;

computing an unnormalized probability surface for each of the aggregate square error surfaces;

computing normalized probability surfaces for each of the unnormalized probability surfaces relative to all the probability surfaces;

computing an aggregate probability surface across the normalized probability surfaces at each location bin; and

computing a mean x- and y-locations to yield an estimated location of the wireless node.

20. The method of claim 19 wherein the plurality of aggregate square error surfaces are each composed of one or more error surfaces that characterize differences between the received signal strength values and expected received signal strength values at one or more locations.