Wireless location system

• US 5,890,068 A
• Filed: 10/03/1996
• Issued: 03/30/1999
• Est. Priority Date: 10/03/1996
• Status: Expired due to Term

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Cell-Loc Inc. (Pembina Pipeline Corp.)
• Inventors: Klukas, Richard W., Borsodi, Andrew L., Lachapelle, Gerard J., Astridge, Mark, Zaghloul, Hatim, Fattouche, Michel T.
• Primary Examiner(s): LE, THANH C
• Application Number:
  US08/725,560
• Time in Patent Office: 908 Days
• Field of Search: 455/456, 455/457, 455/507, 455/577, 455/524, 455/526, 455/440, 455/67.1, 455/67.6, 342/357, 342/451, 342/457, 701/207, 340/905
• US Class Current: 455/456.2
• CPC Class Codes: G01S 5/06 Position of source determined by co-ordinating a plurality of position lines defined by path-difference measurements G01S5/12 takes precedence
• CPC Class Codes: H04W 64/00 Locating users or terminals or network equipment for network management purposes, e.g. mobility management
• CPC Class Codes: Y10T 70/5757 Handle, handwheel or knob

Abstract

The position of a Wireless Transceiver (WT) is estimated using the Time Of Arrival (TOA) at several Monitoring Sites (MS) of the signal transmitted by the WT. Several sites are used to solve for the coordinates of the WT using hyperbolic trilateration. A time reference is used to synchronize the Monitoring Sites (MS). Super-Resolution (SR) techniques are applied to increase the resolution of the TOA estimate obtained via correlation of the received signal at each MS. SR techniques can improve the TOA resolution substantially.

Claims

1. A wireless location system comprising:

a plurality of monitoring stations (MS) and a central site (CS);

each monitoring station including (a) a transceiver for monitoring a signal s(t) transmitted from a wireless transmitter and generating an estimate of the time of arrival (TOA) of the received signal relative to a time reference using an inverse transform whose resolution is greater than Rayleigh resolution and (b) a communication link to transmit the estimated TOA from the monitoring station to the central site; and

the central site including (a) a difference unit that computes the time difference of arrival (TDOA) between each estimated TOA and a preselected TOA and (b) a position estimator which processes all TDOAs and generates an estimate of the position of the wireless transmitter.

2. The wireless location system of claim 1 in which each transceiver includes:

an RF receiver front end for filtering, amplifying and demodulating the received signal thereby generating a filtered, amplified and demodulated signal r_i (t); and

a first processor for estimating the time of arrival (TOA) of the filtered, amplified and demodulated signal r_i (t) relative to a time reference.

3. The wireless location system of claim 2 in which the RF receiver front end includes:

a receiver multicoupler for filtering and amplifying the received signal; and

an RF-to-baseband (RF-to-BB) receiver for demodulating the filtered and amplified signal thereby generating a filtered, amplified and demodulated signal r_i (t).

4. The wireless location system of claim 3 in which the receiver multicoupler includes:

a preselect filter for filtering the received signal;

a first low noise amplifier (LNA) for amplifying the filtered signal; and

a 1;

6 splitter for splitting the amplified and filtered signal into six replicas.

5. The wireless location system of claim 3 in which the RF-to-BB receiver includes:

plural intermediate frequencies (IF) and LNA stages for down-converting and amplifying the signal generated by the receiver multicoupler; and

a demodulator for demodulating the down-converted and amplified signal into a baseband signal r_i (t).

6. The wireless location system of claim 5 in which the demodulator is selected from the group consisting of an Inphase and Quadrature demodulator and an an FM demodulator.

7. The wireless location system of claim 6 in which the FM demodulator includes a phase lock loop (PLL).

8. The wireless location system of claim 2 in which the first processor includes:

a source which generates a generic received signal r(t);

a second processor which processes the generic received signal r(t) thereby generating the power spectral density X_r (f) of r(t);

an acquisition device which processes the generic received signal r(t) together with the demodulated signal r_i (t) to generate a time stump; and

a third processor which processes the power spectral density X_r (f), the generic received signal r(t), the time stamp and the demodulated signal r_i (t) to estimate the TOA of the demodulated signal r_i (t) relative to a time reference.

9. The wireless location system of claim 8 in which the second processor includes:

a first sliding correlator for correlating the generic received signal r(t) with itself thereby generating an autocorrelation function χ

_r (r) of r(t);

a first window for windowing parts of the autocorrelation function χ

_r (c) around its peak thereby generating a windowed autocorrelation function X_r,w (r); and

a Fourier transform for transforming the windowed autocorrelation function X_r,w (r) to the frequency domain thereby generating a power spectral density X_r (f) of r(t).

10. The wireless location system of claim 9 in which the first sliding correlator includes:

a first counter for incrementing the delay τ

from the value 0 to the duration of r(t);

a first delay element for delaying the generic received signal r(t) by the delay τ

thereby generating the delayed signal r(t+τ

); and

a first correlator for computing the correlation between the delayed signal r(t+τ

) and the generic received signal r(t) thereby generating an autocorrelation function χ

_r (τ

) of r(t).

11. The wireless location system of claim 9 in which the duration of the first window is equal to the duration of the second window when the transmitted signal s(t) is a precursor on a reverse control channel transmitted by an AMPs-based wireless transmitter.

12. The wireless location system in claim 9 in which the third processor includes:

a third sliding correlator for correlating the generic received signal r(t) with the demodulated signal r_i (t) thereby generating a crosscorrelation X_r,i (τ

);

a second window for windowing parts of the crosscorrelation function X_r,i (τ

) around its peak thereby generating a windowed autocorrelation function X_r,i,w (τ

);

a Fourier transform for transforming the windowed crosscorrelation function X_r,i,w (τ

) to the frequency domain thereby generating a signal X_r,i (f);

a first computing means for computing a function F(X_r,i (f),X_r (f)) of signal X_r,i (f) and of the autocorrelation X_r (f); and

a fourth processor for processing the function F(X_r,i (f),X_r (f)) and the time stamp thereby generating an estimate of the TOA of the demodulated signal r_i (t) relative to the time stamp.

13. The wireless location system of claim 12 in which the fourth processor includes:

a third window for windowing parts of the function F(X_r,i (f),X_r (f)) around its peak thereby generating a windowed function H_i (f);

an inverse superresolution (SR) transform for inverse transforming the windowed function H_i (f) from the frequency domain to the time domain thereby generating an estimate h_i (τ

) of the impulse response of the RF propagation channel;

a second peak detector to detect the first peak of the estimate h_i (τ

) thereby generating a time τ

_i of the occurrence of the peak; and

a second computing means for computing the estimate of the TOA of the demodulated signal r_i (t) relative to the time stamp using the time τ

_i of the occurrence of the peak.

14. The wireless location system of claim 13 in which the duration of the third window is 11 samples when observing a precursor on a reverse control channel transmitted by an AMPs-based wireless transmitter.

15. The transceiver of claim 13 in which the inverse SR transform consists of one of:

a method whereby group delay τ

_g of the windowed function H_i (f) is obtained thereby generating an estimate h_i (τ

) of the impulse response of the RF propagation channel as a single unit impulse with a delay τ

_g ;

or

a method whereby the estimate h_i (τ

) of the impulse response of the RF propagation channel is obtained by transforming H_i (f) from the frequency domain to the time domain using root MUSIC.

16. The wireless location system of claim 12 in which the duration of the first window is equal to the duration of the second window when the transmitted signal s(t) is a precursor on a reverse control channel transmitted by an AMPs-based wireless transmitter.

17. The wireless location system of claim 12 in which the first computing means computes the function X_r,i (f)/X_r (f).

18. The wireless location system of claim 8 in which the acquisition device includes:

a set of two truncating windows for truncating a segment of the generic received signal r(t) and a segment of the demodulated signal r_i (t) thereby generating truncated signals r_w (t) and r_w,i (t) respectively;

a second sliding correlator for correlating the truncated signal r_w (t) with the truncated signal r_w,i (t) thereby generating a cross-correlation function Y_r,i (τ

);

a first peak detector to detect the peak of the crosscorrelation function Y_r,i (τ

);

a decision device which compares the magnitude of the detected peak to a preselected threshold thereby generating a trigger signal if the magnitude of the peak of the crosscorrelation function Y_r,i (τ

) is larger than the preselected threshold; and

a GPS receiver which generates a time stamp whenever it receives a trigger from the decision device.

19. The wireless location system of claim 18 in which the GPS receiver operates in time transfer mode and is positioned to observe the same satellite at all monitoring stations.

20. The wireless location system of claim 1 in which the system is to be used with the AMPs cellular system, which includes a reverse control channel and a reverse voice channel and the transmitted signal s(t) is selected from the group consisting of:

a precursor signal on the reverse control channel (RECC);

precursor and registration signals on the reverse control channel (RECC); and

a blank-and-burst signal on the reverse voice channel (RVC).

21. The wireless location system of claim 1 in which the position estimator solves for one of:

two coordinates using three MSs using hyperbolic trilateration;

or

three coordinates using four MSs using hyperbolic trilateration;

or

two coordinates using more than three MSs using hyperbolic trilateration and one of parametric least squares, Kalman filtering or maximum likelihood;

or

three coordinates using more than four MSs using hyperbolic trilateration and one of parametric least squares, Kalman filtering or maximum likelihood.

22. A method of estimating the location of a wireless transmitter, the method comprising the steps of:

monitoring a signal s(t) transmitted from the wireless transmitter at each one of a plurality of MSs;

estimating the time of arrival (TOA) of the received signal relative to a time reference at each one of the plurality of MSs using an inverse transform whose resolution is greater than Rayleigh resolution;

transmitting the estimated TOA from each one of the plurality of MSs to a central site (CS);

computing the time difference of arrival (TDOA) between each estimated TOA and every other TOA at the CS; and

processing all TDOAs to generate an estimate of the position of the transmitter.

23. The method of claim 22 in which monitoring the signal s(t) at each MS includes:

filtering, amplifying and demodulating the received signal thereby generating a filtered, amplified and demodulated signal r_i (t); and

estimating the TOA of the filtered, amplified and demodulated signal r_i (t) relative to a time reference.

24. The method of claim 23 in which filtering and amplifying comprises:

preselect filtering the received signal;

low noise amplifying the filtered signal; and

1;

6 splitting the amplified and filtered signal into six replicas.

25. The method of claim 24 in which demodulating comprises:

down-converting to several intermediate frequencies and amplifying over several stages the filtered and amplified signal; and

demodulating the down-converted and amplified signal into a baseband signal r_i (t).

26. The method of claim 25 in which demodulating the down-converted and amplified signal is carried out by one of to Inphase and Quadrature demodulating the signal or FM demodulating the signal.

27. The method of claim 23 in which estimating the TOA of the demodulated signal r_i (t) relative to a time reference comprises:

generating a generic received signal r(t);

processing the generic received signal r(t) to generate a power spectral density X_r (f) of r(t);

processing the generic received signal r(t) together with the demodulated signal r_i (t) to generate a referenced time stamp; and

processing the power spectral density X_r (f), the generic received signal r(t), the referenced time stamp and the demodulated signal r_i (t) to estimate the TOA of the demodulated signal r_i (t) relative to a time reference.

28. The method of claim 27 in which processing the generic received signal r(t) to generate the power spectral density X_r (f) of r(t) includes:

slide-correlating the generic received signal r(t) with itself thereby generating an autocorrelation function χ

_r (r) of r(t);

windowing parts of the autocorrelation function χ

_r (τ

) around its peak thereby generating a windowed autocorrelation function X_r,w (τ

); and

Fourier transforming the windowed autocorrelation function X_r,w (τ

) to the frequency domain thereby generating a power spectral density X_r (f) of r(t).

29. The method of claim 28 in which slide-correlating the generic received signal r(t) with itself thereby generating the autocorrelation function τ

_r (r) of r(t) includes;

incrementing a counter by the delay τ

from the value 0 to the duration of r(t);

delaying the generic received signal r(t) by the delay τ

thereby generating the delayed signal r(t+τ

); and

computing the correlation between the delayed signal r(t+t) and the generic received signal r(t) thereby generating the autocorrelation function τ

_r (τ

) of r(t).

30. The method of claim 27 in which processing the generic received signal r(t) together with the demodulated signal r_i (t) to generate a time stamp includes:

truncating a segment of the generic received signal r(t) and truncating a segment of the demodulated signal r_i (t) thereby generating the truncated signals r_w (t) and r_w,i (t) respectively;

slide-correlating the truncated signal r_w (t) with the truncated signal r_w,i (t) thereby generating their crosscorrelation function Y_r,i (τ

);

detecting the peak of the crosscorrelation function Y_r,i (τ

);

comparing the magnitude of th e detected peak to a preselected threshold thereby generating a trigger signal if the magnitude of the peak of the crosscorrelation function Y_r,i (τ

) is larger than the preselected threshold; and

generating a referenced time stamp whenever a trigger is generated.

31. The method of claim 30 in which generating a referenced time stamp whenever a trigger is generated includes using either a GPS receiver or a high precision clock.

32. The method in claim 27 in which processing the power spectral density X_r (f), the generic received signal r(t), the referenced time stamp and the demodulated signal r_i (t) to estimate the TOA of the demodulated signal r_i (t) relative to a time reference includes:

slide-correlating the generic received signal r(t) with the demodulated signal r_i (t) thereby generating a crosscorrelation χ

_r,i (τ

);

windowing parts of the crosscorrelation function χ

_r,i (τ

) around its peak thereby generating a windowed autocorrelation function χ

_r,i,w (τ

);

Fourier transforming the windowed crosscorrelation function χ

_r,i,w (τ

) to the frequency domain thereby generating the signal X_r,i (f);

computing a function F(X_r,i (f),X_r (f)) of signal X_r,i (f) and of the autocorrelation X_r (f); and

processing the function F(X_r,i (f),X_r (f)) and the referenced time stamp thereby generating an estimate of the TOA of the demodulated signal r_i (t) relative to the referenced time stamp.

33. The method of claim 32 in which processing the function F(X_r,i (f),X_r (f)) and the time stamp thereby generating an estimate of the TOA of the demodulated signal r_i (t) relative to the referenced time stamp includes:

windowing parts of the function F(X_r,i (f),X_r (f)) around its peak thereby generating a windowed function H_i (f);

inverse superresolution (SR) transforming the windowed function H_i (f) from the frequency domain to the time domain thereby generating the estimate h_i (τ

) of the impulse response of the RF propagation channel;

detecting the first peak of the estimate h_i (τ

) thereby generating a time τ

_i of the occurrence of the first peak; and

estimating the TOA of the demodulated signal r_i (t) relative to the referenced time stamp using the time τ

_i of the occurrence of the first peak.

34. The transceiver of claim 33 in which inverse superresolution (SR) transforming the windowed function H_i (f) from the frequency domain to the time domain thereby generating the estimate h_i (τ

) of the impulse response of the RF propagation channel consists of one of;

a method whereby the group delay τ

_g of the windowed function H_i (f) is obtained thereby generating the estimate h_i (τ

) of the impulse response of the RF propagation channel as a single unit impulse with a delay τ

_g ;

or

a method whereby the estimate h_i (τ

) of the impulse response of the RF propagation channel is obtained by transforming H_i (f) from the frequency domain to the time domain using root MUSIC.

35. The transceiver of claim 32 in which computing a function F(X_r,i (f),X_r (f)) of signal X_r,i (f) and of the autocorrelation X_r (f) corresponds to computing the function X_r,i (f)/X_r (f).

36. The method of claim 33 in which windowing parts of the function F(X_r,i (f),X_r (f)) around its peak thereby generating a windowed function H_i (f) comprises discarding all the samples in F(X_r,i (f),X_r (f)) except for 11 samples around its peak when the transmitted signal s(t) is the RECC precursor transmitted by an AMPs-based wireless transmitter.

37. The method of claim 22 in which processing all TDOAs to generate an estimate of the position of the transmitter solves for one of:

two coordinates using three MSs using hyperbolic trilateration;

or

three coordinates using four MSs using hyperbolic trilateration;

or

two coordinates using more than three MSs using hyperbolic trilateration and one of parametric least squares, Kalman filtering or maximum likelihood;

or

three coordinates using more than four MSs using hyperbolic trilateration and one of parametric least squares, Kalman filtering or maximum likelihood.