Suppression on multipath signal effects

• US 5,903,597 A
• Filed: 05/20/1996
• Issued: 05/11/1999
• Est. Priority Date: 05/20/1996
• Status: Expired due to Fees

• Current Assignee: Trimble Navigation Limited (Trimble Inc.)
• Original Assignee: Trimble Navigation Limited (Trimble Inc.)
• Inventors: Pon, Rayman
• Primary Examiner(s): Bocure, Tesfaldet
• Application Number:
  US08/650,631
• Time in Patent Office: 1,086 Days
• Field of Search: 375/209, 375/346, 375/285, 375/267, 375/347, 375/343, 342/378, 364/728.03, 364/728.07
• US Class Current: 375/150
• CPC Class Codes: G01S 19/22 Multipath-related issues
• CPC Class Codes: H04B 7/005 Control of transmission; Equalising

Abstract

Method and apparatus for formation of an autocorrelation difference function of an incoming digital signal, with bit transition interval λτ_chip, that reduces the effects of presence of a multipath signal in an incoming composite signal. First and second autocorrelation functions AC(τ;E;qE) and AC(τ;L;qL) are formed with respective first and second selected time shift values τ=t_E and τ=t_L, which replicate an estimate of an autocorrelation function AC(τ;P;qP) at an intermediate time shift value τ=t_P that satisfies t_E <t_P <t_L with t_L -t_E <2 Δτ_chip, and which depend upon one or more parameters qE and qL, respectively. Independently chosen, non-uniform weighting functions w1(t;qE) and w2(t;qL) are used for formation of the respective autocorrelation functions AC(τ;E;qE) and AC(τ;L;qL) that depend upon one or more parameters qE and qL. Reduction of non-multipath and/or noise effects is achieved by particular choices of weighting functions. The correlator spacing t_L -t_E may be narrow (e.g., Δt_L-E ≦0.3Δτ_chip) or may be any larger value, up to 2Δτ_chip, with appropriate choice of the weighting functions w1 and w2.

Claims

1. A method of formation of an autocorrelation difference function of an incoming signal that reduces the effects of presence of a multipath signal in the incoming signal, the method comprising the steps of:

receiving an incoming digital signal s(t) that can vary with time t and that has a digital signal bit period with a selected length Δ

τ

_chip ;

forming a signal product s(t)S_d (t+τ

)w(t+τ

;

q), where S_d (t) is a selected reference signal, w(t;

q) is a selected, non-constant weighting signal that depends upon one or more selected parameters q, and τ

is a selected time shift value;

forming a first autocorrelation function AC(τ

;

q) of the incoming signal s(t) by integration or summation of the signal product over a selected time interval that includes a time interval defined by t1≦

t≦

t2, where t1 and t2 are selected time values that satisfy a relation t2-t1=Δ

τ

_chip ;

forming a second autocorrelation function AC(τ

;

E;

q) that is a replica of the autocorrelation function AC(τ

;

q) but that is translated to the left along the τ

-axis by a first selected time shift value Δ

t_P-E ;

forming a third autocorrelation function AC(τ

;

L;

q) that is a replica of the autocorrelation function AC(τ

;

q) but that is translated to the right along the τ

-axis by a second selected time shift value Δ

t_L-P, where Δ

t_P-E +Δ

t_L-P <

2Δ

τ

_chip ;

forming an autocorrelation difference function Δ

AC(τ

;

q)=AC(τ

;

E;

q)-AC(τ

;

L;

q);

determining at least one value to of the time shift variable τ

at which the autocorrelation difference function Δ

AC(τ

;

q) changes sign; and

interpreting the time value t=t₀ as an estimate of the time at which a signal, which is substantially free of the presence of a multipath signal, was received.

2. The method of claim 1, further comprising the step of selecting said weighting function w(t;

q) to be a notch function.

3. The method of claim 1, further comprising the step of selecting said weighting function w(t;

q) to be an anti-notch function.

4. The method of claim 1, further comprising the step of selecting said weighting function w(t;

q) to be a step function with Q distinct, non-zero step function amplitudes, where Q is a selected integer>

1.

5. The method of claim 1, further comprising the step of selecting said weighting function w(t;

q) to have at least one positive value and at least one negative value.

6. The method of claim 1, further comprising the step of selecting said first selected time shift value Δ

t_P-E and said second selected time shift value Δ

t_L-P to be substantially equal to each other.

7. The method of claim 1, further comprising the step of selecting the sum of said first selected time shift value and said second selected time shift value to satisfy Δ

t_P-E +Δ

t_L-P ≦

0.3 Δ

τ

_chip.

8. Apparatus for formation of an autocorrelation difference function of an incoming signal that reduces the effects of presence of a multipath signal in the incoming signal, the apparatus comprising:

a signal antenna that receives an incoming digital signal s(t) that can vary with time t;

a signal receiver/processor, including a computer, that receives the incoming signal s(t) from the signal antenna, that forms a signal product s(t)S_d (t+τ

)w(t+τ

;

q), where S_d (t) is a selected reference signal, w(t;

q) is a selected, non-constant weighting signal that depends upon one or more parameters q, and τ

is a selected time shift, and that is programmed to;

(i) form a first autocorrelation function AC(τ

;

q) of the incoming signal s(t) by integration or summation of the signal product over a selected time interval that includes a time interval defined by t1≦

t≦

t2, where t1 and t2 are selected time values that satisfy a relation t2-t1=Δ

τ

_chip ;

(ii) form a second autocorrelation function AC(τ

;

E;

q) that is a replica of the autocorrelation function AC(τ

;

q) but that is translated to the left along the τ

-axis by a first selected time shift value Δ

t_P-E ;

(iii) form a third autocorrelation function AC(t;

L;

q) that is a replica of the autocorrelation function AC(τ

;

q) but that is translated to the right along the τ

-axis by a second selected time shift value Δ

t_L-P, where Δ

t_P-E +Δ

t_L-P <

2Δ

τ

_chip ;

(iv) form an autocorrelation difference function Δ

AC(τ

;

q)=AC(τ

;

E;

q)-AC(τ

;

L;

q);

(v) determine at least one value to of the time shift variable τ

at which the autocorrelation difference function Δ

AC(τ

;

q) changes sign; and

(vi) use the time value t=t₀ to estimate the time at which a signal, which is substantially free of the presence of a multipath signal, was received at the signal antenna.

9. The apparatus of claim 8, wherein said weighting function w(t;

q) used by said signal receiver/processor to form said signal product is selected to be a notch function.

10. The apparatus of claim 8, wherein said weighting function w(t;

q) used by said signal receiver/processor to form said signal product is selected to be an anti-notch function.

11. The apparatus of claim 8, wherein said weighting function w(t;

q) used by said signal receiver/processor has at least one positive value and has at least one negative value.

12. The apparatus of claim 8, wherein said weighting function w(t;

q) is selected to be a step function with Q distinct step function amplitudes, where Q is a selected integer≧

1.

13. The apparatus of claim 8, wherein said first selected time shift value Δ

t_P-E and said second selected time shift value Δ

t_L-P are substantially equal to each other.

14. The apparatus of claim 8, wherein the sum of said first time shift value and said second time shift value satisfy Δ

t_P-E +Δ

t_L-P ≦

0.3 Δ

τ

_chip.

15. A method of formation of an autocorrelation difference function of an incoming signal that reduces the effects of presence of a multipath signal in the incoming signal, the method comprising the steps of:

receiving an incoming digital signal s(t) that can vary with time t and that has a digital signal bit period with a selected length Δ

τ

_chip ;

forming a first signal product s(t)S_d (t+τ

)w(t+τ

;

E;

qE), where S_d (t) is a selected reference signal, w(t;

E;

qE) is a first selected, non-constant weighting signal that depends upon one or more selected parameters qE, and τ

is a selected time shift value;

forming a second signal product s(t)S_d (t+τ

)w(t+τ

;

L;

qL), where w(t;

L;

qL) is a second selected, non-constant weighting signal that depends upon one or more selected parameters qL;

forming a first autocorrelation function AC(τ

;

E;

qE) of the incoming signal s(t) by integration or summation of the first signal product over a selected time interval that includes a time interval defined by t1≦

t≦

t2, where t1 and t2 are selected time values that satisfy a relation t2-t1=Δ

τ

_chip ;

forming a second autocorrelation function AC(τ

;

L;

qL) of the incoming signal by integration or summation of the second signal product over the selected time interval, where the first and second autocorrelation functions AC(τ

;

E;

qE) and AC(τ

;

L;

qL) are formed with a selected time shift difference Δ

t_L-E satisfies Δ

t_L-E <

2Δ

τ

_chip ;

forming an autocorrelation difference function Δ

AC(τ

;

qE;

qL)=AC(τ

;

E;

qE)-AC(τ

;

L;

qL);

determining at least one value to of the time shift variable τ

at which the autocorrelation difference function Δ

AC(τ

;

qE;

qL) changes sign; and

interpreting the time value t=t₀ as an estimate of the time at which a signal, which is substantially free of the presence of a multipath signal, was received.

16. The method of claim 15, further comprising the step of selecting said weighting function w(t;

q) to be a notch function.

17. The method of claim 15, further comprising the step of selecting said weighting function w(t;

q) to be an anti-notch function.

18. The method of claim 15, further comprising the step of selecting said weighting function w(t;

q) to be a step function with Q distinct, non-zero step function amplitudes, where Q is a selected integer≧

1.

19. The method of claim 15, further comprising the step of selecting at least one of said weighting functions w(t;

E;

qE) and w(t;

L;

qL) to have at least one positive value and at least one negative value.

20. The method of claim 15, further comprising the step of selecting said time shift difference to satisfy Δ

t_L-E ≦

0.3 Δ

τ

_chip.

21. Apparatus for formation of an autocorrelation difference function of an incoming signal that reduces the effects of presence of a multipath signal in the incoming signal, the apparatus comprising:

a signal antenna that receives an incoming digital signal s(t) that can vary with time t and that has a digital signal bit period with a selected length Δ

τ

_chip ;

a signal receiver/processor, including a computer, that receives the incoming signal s(t) from the signal antenna and that is programmed to;

(i) form a first signal product s(t)S_d (t+τ

)w(t+τ

;

E;

qE), where S_d (t) is a selected reference signal, w(t;

E;

qE) is a first selected, non-constant weighting signal that depends upon one or more selected parameters qE, and τ

is a selected time shift value;

(ii) form a second signal product s(t)S_d (t+τ

)w(t+τ

;

L;

qL), where w(t;

L;

qL) is a second selected, non-constant weighting signal that depends upon one or more selected parameters qL;

(iii) form a first autocorrelation function AC(τ

;

E;

qE) of the incoming signal by integration or summation of the first signal product over a selected time interval that includes a time interval defined by t1≦

t≦

t2, where t1 and t2 are selected time values that satisfy a relation t2-t1=Δ

τ

_chip ;

(iv) form a second autocorrelation function AC(τ

;

L;

qL) of the incoming signal by integration or summation of the second signal product over the selected time interval, where the first and second autocorrelation functions AC(τ

;

E;

qE) and AC(τ

;

L;

qL) are formed with a selected time shift difference Δ

t_L-E that satisfies Δ

t_L-E <

2Δ

τ

_chip ;

(v) form an autocorrelation difference function Δ

AC(τ

;

qE;

qL)=AC(τ

;

E;

qE)-AC(τ

;

L;

qL);

(vi) determine at least one value t₀ of the time shift variable τ

at which the autocorrelation difference function Δ

AC(τ

;

qE;

qL) changes sign; and

(vii) use the time value t=t₀ to estimate the time at which a signal, which is substantially free of the presence of a multipath signal, was received at the signal antenna.

22. The apparatus of claim 21, wherein said weighting function w(t;

q) used by said signal receiver/processor to form said signal product is selected to be a notch function.

23. The apparatus of claim 21, wherein said weighting function w(t;

q) used by said signal receiver/processor to form said signal product is selected to be an anti-notch function.

24. The apparatus of claim 21, wherein said weighting function w(t;

q) is selected to be a step function with Q distinct step function amplitudes, where Q is a selected integer≧

1.

25. The apparatus of claim 21, wherein at least one of said weighting functions w(t;

E;

qE) and w(t;

L;

qL) has at least one positive value and has at least one negative value.

26. The apparatus of claim 21, wherein said time shift difference Δ

t_L-E satisfies Δ

t_L-E≦

0.3 Δ

τ

_chip.

27. A method of formation of an autocorrelation difference function of an incoming signal that reduces the effects of presence of a multipath signal in the incoming signal, the method comprising the steps of:

receiving an incoming digital signal s(t) that can vary with time t and that has a digital signal bit period with a selected length Δ

τ

_chip ;

sampling the incoming signal non-uniformly over a sequence of selected time intervals J_m ={t|t1+m(t2-t1)≦

t≦

t1+(m+1)(t2-t1)} (m=0, 1, 2, . . . ) at a sequence of times t=t_n (n 1, 2, . . . , N), where each sample time t_n lies in a selected subinterval J'"'"'_m of J_m defined by J'"'"'_m ={t|t1'"'"'+m(t2-t1)≦

t≦

t2'"'"'+m(t2-t1)}, where t1<

t1'"'"'<

t2'"'"'<

t2, where t1 and t2 are selected time values that satisfy a relation t2-t1=Δ

τ

_chip ;

forming a sequence of signal products s(t_n)S_d (t_n +τ

), where S_d (t) is a selected reference signal and τ

is a selected time shift value;

forming an autocorrelation function AC(τ

;

q) of the incoming signal by summation of the signal products s(t_n)S_d (t_n +τ

) (n=1, 2, . . . , N) over a selected time interval that includes at least one time interval J_m ;

forming a second autocorrelation function AC(τ

;

E;

q) that is a replica of the autocorrelation function AC(τ

;

q) but that is translated to the left along the t-axis by a positive time shift value Δ

t_P-E ;

forming a third autocorrelation function AC(τ

;

L;

q) that is a replica of the autocorrelation function AC(τ

;

q) but that is translated to the right along the t-axis by a positive time shift value Δ

t_L-P, where Δ

t_P-E +Δ

t_L-P <

2Δ

τ

_chip ;

forming an autocorrelation difference function Δ

AC(τ

;

q)=AC(τ

;

E;

qE)-AC(τ

;

L;

qL);

determining at least one value to of the time shift variable τ

at which the autocorrelation difference function Δ

AC(τ

;

q) changes sign; and

using the time value t=t₀ to estimate the time at which a signal, which is substantially free of the presence of a multipath signal, was received at the signal antenna.

28. Apparatus for formation of an autocorrelation difference function of an incoming signal that reduces the effects of presence of a multipath signal in the incoming signal, the apparatus comprising:

a signal antenna that receives an incoming digital signal s(t) that can vary with time t and that has a digital signal bit period with a selected length Δ

τ

_chip ;

a signal receiver/processor, including a computer, that receives the incoming signal s(t) from the signal antenna and that is programmed to;

(i) sample the incoming signal non-uniformly over a sequence of selected time intervals J_m ={t|t1+m(t2-t1)≦

t≦

t1+(m+1)(t2-t1)} (m=0, 1, 2, . . . ) at a sequence of times t=t_n (n 1, 2, . . . , N), where each sample time t_n lies in a selected subinterval J'"'"'_m of J_m defined by J'"'"'_m ={t|t1'"'"'+m(t2-t1)≦

t≦

t2'"'"'+m(t2-t1)}, where t1<

t1'"'"'<

t2'"'"'<

t2, where t1 and t2 are selected time values that satisfy a relation t2-t1=Δ

τ

_chip ;

(ii) form a sequence of signal products s(t_n)S_d (t_n +τ

), where S_d (t) is a selected reference signal and τ

is a selected time shift value;

(iii) form an autocorrelation function AC(τ

;

q) of the incoming signal by summation of the signal products s(t_n)S_d (t_n +τ

) (n=1, 2, . . . , N) over a selected time interval that includes at least one time interval J_m ;

(iv) form a second autocorrelation function AC(τ

;

E;

q) that is a replica of the autocorrelation function AC(τ

;

P;

q) but that is translated to the left along the τ

-axis by a positive time shift value Δ

t_P-E ;

(v) form a third autocorrelation function AC(τ

;

L;

q) that is a replica of the autocorrelation function AC(τ

;

P;

q) but that is translated to the right along the τ

-axis by a positive time shift value Δ

t_L-P, where Δ

t_P-E +Δ

t_L-P <

2Δ

τ

_chip ;

(vi) form an autocorrelation difference function Δ

AC(τ

;

q)=AC(τ

;

E;

qE)-AC(τ

;

L;

qL);

(vii) determine at least one value to of the time shift variable τ

at which the autocorrelation difference function Δ

AC(τ

;

q) changes sign; and

(viii) use the time value t=t₀ to estimate the time at which a signal, which is substantially free of the presence of a multipath signal, was received at the signal antenna.

29. A method for accumulating autocorrelation data representing a timing relationship between a first time varying digital signal containing a desired signal and a second time varying digital signal that is a replica of the desired signal, where the desired signal has a digital bit value time interval of selected length Δ

τ

_chip, the method comprising the steps of;

generating a first weighting signal, having at least two distinct values;

generating a second weighting signal, having at least two distinct values;

mixing the first weighting signal, the first signal and the second signal to generate a first weighted signal;

mixing the second weighting signal, the first signal and the second signal to generate a second weighted signal;

accumulating the first weighted signal over a first predetermined time interval to form a first accumulated weighted signal;

accumulating the second weighted signal over a second predetermined time interval to form a second accumulated weighted signal, where each of the first and the second weighting signals is chosen to de-emphasize the contribution, to the respective first and second accumulated weighted signals, of at least one time interval that does not include and is spaced apart from an estimated time at which the first signal was received; and

forming a third accumulated weighted signal that is the difference of the accumulated first weighted signal and the second accumulated weighted signal.

30. The method of claim 29, further comprising the step of shifting said first signal in time relative to said second signal.

31. The method of claim 30, further comprising the step of shifting said first signal to an earlier time relative to said second signal.

32. The method of claim 31, further comprising the step of selecting the magnitude of said shift in time of said first signal relative to said second signal to be greater than 0.5Δ

τ

_chip.

33. The method of claim 30, further comprising the step of shifting said first signal to a later time relative to said second signal.

34. The method of claim 33, further comprising the step of selecting the magnitude of said shift in time of said first signal relative to said second signal to be greater than 0.5Δ

τ

_chip.

35. The method of claim 29, further comprising the steps of:

identifying a transition time at which said second signal makes a transition from a first digital bit value to a second digital bit value; and

generating said first weighting signal so that said first weighting signal has non-zero values for a selected time interval that includes the transition time.

36. A method for accumulating autocorrelation data representing a timing relationship between a first time varying digital signal containing a desired signal and a second time varying digital signal that is a replica of the desired signal, where the desired signal has a digital bit value time interval of selected length Δ

τ

_chip, the method comprising the steps of;

identifying a transition time at which the second signal makes a transition from a first digital bit value to a second digital bit value;

generating a first weighting signal so that the first weighting signal has non-zero values for a selected time interval that includes the transition time.

mixing the first weighting signal, the first signal and the second signal to generate a first weighted signal; and

accumulating the first weighted signal.

37. The method of claim 36, further comprising the step of shifting said first signal in time relative to said second signal.

38. The method of claim 37, further comprising the step of shifting said first signal to an earlier time relative to said second signal.

39. The method of claim 38, further comprising the step of selecting the magnitude of said shift in time of said first signal relative to said second signal to be greater than 0.5Δ

τ

_chip.

40. The method of claim 37, further comprising the step of shifting said first signal to a later time relative to said second signal.

41. The method of claim 40, further comprising the step of selecting the magnitude of said shift in time of said first signal relative to said second signal to be greater than 0.5Δ

τ

_chip.

42. Apparatus for accumulating autocorrelation data representing a timing relationship between a first time varying digital signal containing a desired signal and a second time varying digital signal that is a replica of the desired signal, where the desired signal has at least one transition time at which the desired signal makes a transition from a first digital value to a second digital value, the apparatus comprising:

signal mixing means for receiving and mixing the first signal and the second signal to generate a mixed signal; and

a signal sampler to sample the mixed signal at each of a sequence of at least two selected sampling times, to accumulate the sampled values of the mixed signal, and to form an accumulated mixed signal, wherein at least one of the sampling times is selected to lie in a neighborhood surrounding a transition time at which the second signal makes a transition from a first digital bit value to a second digital bit value.

43. The apparatus of claim 42, further comprising time shift means for receiving at least one of said first signal and said second signal and for shifting said first signal in time relative to said second signal by a selected time shift Δ

t1,

wherein said signal mixing means receives and mixes said second signal and said first signal time shifted by Δ

t1 to form said mixed signal.

44. The apparatus of claim 43, wherein:

said time shift means also shifts said first signal in time relative to said second signal by a second selected time shift Δ

t2, where Δ

t2 is different from Δ

t1;

said signal mixing means also receives and mixes said second signal and said first signal time shifted by Δ

t2 to form a second mixed signal;

said signal sampler also samples the second mixed signal at each of a sequence of at least two selected sampling times, accumulates the sampled values of the second mixed signal, and forms an accumulated second mixed signal; and

said apparatus further comprises signal difference means for receiving and forming a difference of said accumulated mixed signal and the accumulated second mixed signal.