Likelihood-based geolocation prediction algorithms for CDMA systems using pilot strength measurements

• US 6,889,053 B1
• Filed: 07/26/1999
• Issued: 05/03/2005
• Est. Priority Date: 07/26/1999
• Status: Expired due to Term

• Current Assignee: Alcatel-Lucent USA, Inc. (Nokia Corporation)
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Chang, Kirk K., Jeske, Daniel R., Rege, Kiran M., Wang, Yung-Terng
• Primary Examiner(s): Cumming, William D.
• Application Number:
  US09/359,648
• Time in Patent Office: 2,108 Days
• Field of Search: 370/342, 375/262, 342/357.01, 342/357.02, 342/450
• US Class Current: 455/456.3
• CPC Class Codes: G01S 5/0278 involving statistical or probabilistic considerations G01S5/0252, G01S5/0294 take precedence
• CPC Class Codes: H04W 64/00 Locating users or terminals or network equipment for network management purposes, e.g. mobility management

Abstract

The location of a mobile wireless communication unit in the service area of a CDMA communications system is predicted utilizing two likelihood functions that define maximum likelihood estimators of the mobile unit'"'"'s location, based on attribute measurements, such as but not limited to pilot signal strength, being made at the location of the mobile unit and reported back to a base station. One of the likelihood functions comprises a frequentist likelihood function and the other comprises a Bayesian-modified likelihood function. The likelihood functions are based on the assumption that there is an RF model which provides the probability a mobile unit is able to detect one or more attributes associated with an arbitrary base station, given it is located at an arbitrary location within the service area. Each of the likelihoods are also incorporated into a sequential Bayesian procedure which outputs a posterior distribution indicative of the location of the mobile unit.

Claims

1. A method of predicting the location of a CDMA mobile communications unit in a wireless communications service area, comprising the steps of:

(a) receiving measure attribute information from a mobile unit location in the service area, said attribute information being specific to the location of the mobile unit in the service area;

(b) computing the probability of the mobile unit being at a specific location in the service area in response to said received attribute information using a likelihood probability function, said likelihood probability function having an iterative procedure for generating a maximum likelihood estimator of the mobile unit'"'"'s location in this service area, wherein said likelihood probability function includes a frequentist likelihood function;

(c) generating an output indicative of the likelihood of the mobile unit being at said location in the service area.

2. A method according to claim 1 wherein the maximum likelihood estimator comprises the coordinates in the service area which maximizes the likelihood probability function.

4. A method according to claim 1 wherein said procedure comprises sequential Bayesian procedure characteristics.

5. A method according to claim 4 wherein said sequential Bayesian procedure characteristics are implemented as a frequentist likelihood function.

6. A method according to claim 1 wherein said attribute information comprises a pilot signal strength measurement of at least one visible pilot signal at said location of the mobile unit.

7. A method according to claim 6 and additionally including a step (d) prior to step (b) of reducing the effective size of the service area so as to reduce the number of computations required in step (b).

8. A method according to claim 6 and following step (a) and before step (b), additionally including a step (d) of identifying a region A of support for the mobile unit in the service area and (e) identifying a set of possible pilot signals which can be detected by the mobile unit in said region of support.

9. A method according to claim 8 and additionally including a step (f) of computing an approximation {tilde over (θ

)}_ij(x,y) of the probability the mobile unit detects a pilot signal at said specific location (x,y) in said region A for all pilot signals in said set of possible pilot signals and where j is a sector of a multi-sector based station I in said region of support.

10. A method according to claim 9 wherein {tilde over (θ

)}_ij(x,y) is derived from an RF model for the RF power R_ij(x,y) received by the mobile unit and where the model is of the form of

R_ij(x,y)=T_ijG_ij(x,y)L_ij(x,y)F_ij(x,y)M_ij(x,y)

where T_ij is the transmit power associated with sector j of base station I, G_ij(x,y) is the antenna gain for the sector j of base station I along the direction pointing towards the location (x,y) within A, L_ij(x,y) is the distance loss between the base station I associated with the sector j and the location (x,y) within A, F_ij(x,y) is the shadow fading factor and M_ij(x,y) is the measured noise factor.

11. A method according to claim 10 wherein said attribute information comprises the measured visibility of a pilot signal transmitted from a base station and the likelihood probability function comprises an iterative frequentist maximum likelihood (ML) function in the form of $L_{\mathrm{ML}}^s$

(x,y)∝

LMLs-1⁡

(x,y)⁢

∏

ij∈

K⁢

 

⁢

[θ

~ij⁡

(x,y)]μ

ijs⁡

[1-θ

~ij⁡

(x,y)]1-μ

ijs,(x,y)∈

A

where x and y comprise rectilinear coordinates in the service area A, s is the number of measurement epochs, and μ

_ij^s is equal to one or zero depending on whether the mobile unit can detect a pilot signal ij at measurement epoch s.

14. A method according to claim 11 wherein the frequentist likelihood function L_ML^s(x,y) is combined with a discrete uniform prior distribution for the location of the mobile unit of the form $P_{\mathrm{ML}}^{\theta }$

⁡

(x,y)=1

A

,(x,y)∈

A

to generate a sequential Bayesian procedure which provides a posterior distribution for the location of the mobile unit of the form $P_{\mathrm{ML}}^s<br><br>\left(x,y\right)\propto <br><br>P_{\mathrm{ML}}^{s-1}<br><br>\left(x,y\right)<br><br>\underset{\mathrm{ij}\in <br><br>K}{\prod <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{{\left[{\widetilde{\mathrm{\theta <br><br>}}}_{\mathrm{ij}}<br><br>\left(x,y\right)\right]}^{{\mathrm{\mu <br><br>}}_{\mathrm{ij}}^s}<br><br>\left[1-{\widetilde{\mathrm{\theta <br><br>}}}_{\mathrm{ij}}<br><br>\left(x,y\right)\right]}^{1-{\mathrm{\mu <br><br>}}_{\mathrm{ij}}^s},\left(x,y\right)\in <br><br>A.$

where ∥

A∥

is the number of grid points contained within A, and where x and y comprise rectilinear coordinates in the service area A, s is the number of measurement epochs, and μ

_ij^s is equal to one or zero depending on whether the mobile unit can detect a pilot signal ij at measurement epoch s.

3. A method of predicting the location of a CDMA mobile communications unit in a wireless communications service area, comprising the steps of:

(a) receiving measured attribute information from a mobile unit location in the service area, said attribute information being specific to the location of the mobile unit in the service area;

(b) computing the probability of the mobile unit being at a specific location in the service area in response to said received attribute information using a likelihood probability function;

(c) generating an output indicative of the likelihood of the mobile unit being at said location in the service area;

wherein said likelihood probability function includes an iterative procedure for producing a maximum likelihood estimator of the mobile unit'"'"'s location in this service area wherein said likelihood probability function comprises a Bayes-modified likelihood function.

12. A method of predicting the location of a CDMA mobile communications unit in a wireless communications service area, comprising the steps of:

(a) receiving measure attribute information from a mobile unit location in the service area, said attribute information being specific to the location of the mobile unit in the service area;

(b) computing the probability of the mobile unit being at a specific location in the service area in response to said received attribute information using a likelihood probability function;

(c) generating an output indicative of the likelihood of the mobile unit being at said location in the service area;

wherein said attribute information comprises a pilot signal strength measurement of at least one visible pilot signal at said location of the mobile unit;

and following step (a) and before step (b), additionally including a step (d) of identifying a region A of support for the mobile unit in the service area and (e) identifying a set of possible pilot signals which can be detected by the mobile unit in said region of support;

and additionally including a step (f) of computing an approximation {tilde over (θ

)}_ij(x,y) of the probability the mobile unit detects a pilot signal at said specific location (x,y) in said region A for all pilot signals in said set of possible pilot signals and where j is a sector of a multi-sector based station I in said region of support;

wherein {tilde over (θ

)}_ij(x,y) is derived from an RF model for the RF power R_ij(x,y) received by the mobile unit and where the model is of the form of

R_ij(x,y)=T_ijG_ij(x,y)L_ij(x,y)F_ij(x,y)M_ij(x,y)

where T_ij is the transmit power associated with sector j of base station I, G_ij(x,y) is the antenna gain for the sector j of base station I along the direction pointing towards the location (x,y) within A, L_ij(x,y) is the distance loss between the base station I associated with the sector j and the location (x,y) within A, F_ij(x,y) is the shadow fading factor and M_ij(x,y) is the measured noise factor;

wherein said attribute information comprises the measured visibility of a pilot signal transmitted from a base station and the likelihood probability function comprises an iterative Bayes-modified maximum likelihood (ML) estimator in the form of $L_{B<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{ML}}^s<br><br>\left(x,y\right)\propto <br><br>L_{B<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{ML}}^{s-1}<br><br>\left(x,y\right)<br><br>\text{\hspace{1em}<br><br>}<br><br>\underset{\mathrm{ij}\in <br><br>K}{\prod <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>\frac{{{\left[n_{\mathrm{ij}}^{s-1}+{\mathrm{\alpha <br><br>}}_{\mathrm{ij}}<br><br>\left(x,y\right)\right]}^{{\mathrm{\mu <br><br>}}_{\mathrm{ij}}^s}<br><br>\left[s-1-n_{\mathrm{ij}}^{s-1}+{\mathrm{\beta <br><br>}}_{\mathrm{ij}}<br><br>\left(x,y\right)\right]}^{1-{\mathrm{\mu <br><br>}}_{\mathrm{ij}}^s}}{{\mathrm{\alpha <br><br>}}_{\mathrm{ij}}<br><br>\left(x,y\right)+{\mathrm{\beta <br><br>}}_{\mathrm{ij}}<br><br>\left(x,y\right)+s-1},\left(x,y\right)\in <br><br>A$

where x and y comprise rectilinear coordinates in the service area A s is the number of measurement epochs, η

_ij<sup>s−

1</sup> is the number of times a pilot signal in section j of a multi-sector base station I is visible through the first s−

1 measurement epochs, α

_ij(x,y) and β

_ij(x,y) are parameters of a beta distribution, and μ

_ij^s is equal to one or zero depending on whether the mobile unit can detect a pilot signal ij at measurement epoch s.

13. A method according to claim 12 where ${\alpha }_{}$

ij⁡

(x,y)={1,if⁢

 

⁢

θ

~ij⁡

(x,y)≤

1/2 

⁢

θ

~ij⁡

(x,y)1- 

⁢

θ

~ij⁡

(x,y),if⁢

 

⁢

θ

~ij⁡

(x,y)>

1/2⁢

⁢

β

ij⁡

(x,y)={ 

⁢

1- 

⁢

θ

~ij⁡

(x,y)θ

~ij⁡

(x,y),if⁢

 

⁢

θ

~ij⁡

(x,y)≤

1/21,if⁢

 

⁢

θ

~ij⁡

(x,y)>

1/2.

15. A method of predicting the location of a CDMA mobile communications unit in a wireless communications service area, comprising the steps of:

(a) receiving measure attribute information from a mobile unit location in the service area, said attribute information being specific to the location of the mobile unit in the service area;

(b) computing the probability of the mobile unit being at a specific location in the service area in response to said received attribute information using a likelihood probability function;

(c) generating an output indicative of the likelihood of the mobile unit being at said location in the service area;

wherein said attribute information comprises a pilot signal strength measurement of at least one visible pilot signal at said location of the mobile unit;

and following step (a) and before step (b), additionally including a step (d) of identifying a region A of support for the mobile unit in the service area and (e) identifying a set of possible pilot signals which can be detected by the mobile unit in said region of support;

and additionally including a step (f) of computing an approximation {tilde over (θ

)}_ij(x,y) of the probability the mobile unit detects a pilot signal at said specific location (x,y) in said region A for all pilot signals in said set of possible pilot signals and where j is a sector of a multi-sector based station I in said region of support;

wherein {tilde over (θ

)}_ij(x,y) is derived from an RF model for the RF power R_ij(x,y) received by the mobile unit and where the model is of the form of

R_ij(x,y)=T_ijG_ij(x,y)L_ij(x,y)F_ij(x,y)M_ij(x,y)

where T_ij is the transmit power associated with sector j of base station I, G_ij(x,y) is the antenna gain for the sector j of base station I along the direction pointing towards the location (x,y) within A, L_ij(x,y) is the distance loss between the base station I associated with the sector j and the location (x,y) within A, F_ij(x,y) is the shadow fading factor and M_ij(x,y) is the measured noise factor;

wherein the frequentist likelihood function L_BML^s(x,y) is combined with a discrete uniform prior distribution for the location of the mobile unit of the form $P_{B<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{ML}}^{\mathrm{\theta <br><br>}}<br><br>\left(x,y\right)=\frac{1}{<br><br>A<br><br>},\left(x,y\right)\in <br><br>A$

to generate a sequential Bayesian procedure which provides a posterior distribution for the location of the mobile unit of the form $P_{B<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{ML}}^s<br><br>\left(x,y\right)\propto <br><br>P_{\mathrm{BML}}^{s-1}<br><br>\left(x,y\right)<br><br>\text{\hspace{1em}<br><br>}<br><br>\underset{\mathrm{ij}\in <br><br>K}{\prod <br><br>}<br><br>\frac{{{\left[n_{\mathrm{ij}}^{s-1}+{\mathrm{\alpha <br><br>}}_{\mathrm{ij}}<br><br>\left(x,y\right)\right]}^{{\mathrm{\mu <br><br>}}_{\mathrm{ij}}^s}<br><br>\left[s-1-n_{\mathrm{ij}}^{s-1}+{\mathrm{\beta <br><br>}}_{\mathrm{ij}}<br><br>\left(x,y\right)\right]}^{1-{\mathrm{\mu <br><br>}}_{\mathrm{ij}}^s}}{{\mathrm{\alpha <br><br>}}_{\mathrm{ij}}<br><br>\left(x,y\right)+{\mathrm{\beta <br><br>}}_{\mathrm{ij}}<br><br>\left(x,y\right)+s-1},\left(x,y\right)\in <br><br>A.$

where ∥

A∥

is the number of grid points contained within A, and where x and y comprise rectilinear coordinates in the service area A, s is the number of measurement epochs, n is the number of times a pilot signal in section j of a multi-sector base station I is visible through the first s−

1 measurement epochs, α

_ij(x,y) and β

_ij(x,y) are parameters of a beta distribution, and μ

_ij^s is equal to one or zero depending on whether the mobile unit can detect a pilot signal ij at measurement epoch s.

16. A method according to claim 15 where ${\alpha }_{}$

ij⁡

(x,y)={1,if⁢

 

⁢

θ

~ij⁡

(x,y)≤

1/2 

⁢

θ

~ij⁡

(x,y)1- 

⁢

θ

~ij⁡

(x,y),if⁢

 

⁢

θ

~ij⁡

(x,y)>

1/2⁢

⁢

β

ij⁡

(x,y)={ 

⁢

1- 

⁢

θ

~ij⁡

(x,y)θ

~ij⁡

(x,y),if⁢

 

⁢

θ

~ij⁡

(x,y)≤

1/21,if⁢

 

⁢

θ

~ij⁡

(x,y)>

1/2.