Serial search acquisition system with adaptive threshold and optimal decision for spread spectrum systems

• US 5,642,377 A
• Filed: 07/25/1995
• Issued: 06/24/1997
• Est. Priority Date: 07/25/1995
• Status: Expired due to Term

• Current Assignee: Nokia Technologies Oy (Nokia Corporation)
• Original Assignee: Nokia Mobile Phones UK Limited (Nokia Corporation)
• Inventors: Chung, Sanguoon, Noneman, John W.
• Primary Examiner(s): Chin, Stephen
• Assistant Examiner(s): PHAN, HAI
• Application Number:
  US08/506,336
• Time in Patent Office: 700 Days
• Field of Search: 375/200, 375/205, 375/206, 375/367, 380/46, 370/18
• US Class Current: 375/145
• CPC Class Codes: H04B 1/70755 Setting of lock conditions, e.g. threshold

Abstract

A PN code acquisition system for Code Division Multiple Access (CDMA) Direct Sequence Spread Spectrum (DSSS) systems, with automatic decision threshold, is based on the combination of the Maximum Likelihood (ML) and Serial Search (SS) acquisition approaches. In contrast to the conventional SS acquisition system, the disclosed system adaptively estimates optimal threshold by exploiting the statistics of the signal and noise, and makes an optimal decision based on the threshold. The system estimates the threshold by employing ML estimation and applies the threshold as in SS acquisition and makes a decision by comparing the updated threshold with the current signal strength. Together with post detection verification logic, this approach will increase detection probability and reduce false alarm probability significantly. An optimum system parameter design approach and the advantages of this approach as compared to conventional approaches (ML or SS acquisition) are demonstrated.

Claims

1. A method for PN code acquisition in a PN code signal transmitted in a Code Division Multiple Access (CDMA) Direct Sequence Spread Spectrum (DSSS) system, comprising the steps of:

receiving a transmitted PN code signal and multiplying said signal with a PN code replica signal to produce an input signal for use in determining the phase alignment of said PN code signals;

integrating the signal energy of said input signal over a time interval to obtain a first result;

establishing a noise estimate;

establishing a signal detection threshold having a value greater than that of said noise estimate;

establishing a signal classification threshold, having a value between that of said signal detection threshold and said noise estimate;

establishing a first number C_N of PN phases i to be searched and a total number q of PN phases for said PN code replica signal;

maintaining a count c_n of said PN phases i; and

searching for said alignment by detecting the signal energy levels and the noise levels of said input signal by the steps of;

evaluating whether said first result is equal to or exceeds the value of said signal detection threshold;

if NO, evaluating whether said first result exceeds the value of said signal classification threshold; and

if NO, updating the value of said noise estimate with the value of said first result and updating said count c_n by 1, and determining whether said updated count is equal to or exceeds C_N ;

if YES, ending the search;

if NO, determining whether the PN phase i of said PN code replica signal equals q;

if YES, ending the search;

if NO, updating the PN phase i of said PN code replica signal by 1, and continuing the search by repeating the foregoing steps using the phase updated PN code replica signal to multiply said transmitted PN code signal;

if YES, resetting said count c_n to 0, and determining whether the PN phase i of said PN code replica equals q;

if YES, ending the search;

if NO, updating the PN phase i of said PN code replica signal by 1, and continuing the search by repeating the foregoing steps using the phase updated PN code replica signal to multiply said transmitted PN code signal.

2. The method of claim 1 wherein said signal detection threshold comprises a first signal detection value T_SD1 multiplied by (1-y), where T_SD1 is the maximum integrated signal energy of said input signal obtained from past history till i-1, said first result is obtained at time t and phase i, and y is between 1/16 and 1/8;

and further comprising the steps of;

if said first result exceeds said signal detection threshold, then evaluating whether said first result exceeds said first signal detection value T_SD1 ;

if YES, changing said first signal detection value T_SD1 to an updated signal detection value equal to said first result, and updating the value T_SC1 of said signal classification threshold to an updated signal classification threshold value equal to the average of said updated signal detection value and said noise estimate value;

if NO, equating said updated signal detection value to said first signal detection value T_SD1 and equating said updated signal classification threshold value to said first signal classification threshold value T_SC1 ;

integrating the signal energy of said input signal over a second time interval to obtain a second result;

evaluating whether said second result is equal to or exceeds a second signal detection threshold value

if NO, evaluating whether said second result exceeds the value of a second signal classification threshold value T_SC2 ; and

if NO, updating the value of said noise estimate with the value of said second result and updating said count c_n by 1, and determining whether said updated count is equal to or greater than C_N ;

if YES, ending the search;

if NO, determining whether the PN phase i of said PN code replica signal equals q;

if YES, ending the search;

if NO, updating the PN phase i of said PN code replica signal by 1, and continuing the search by repeating the foregoing steps using the phase updated PN code replica signal to multiply said transmitted PN code signal;

if YES, resetting said count c_n to 0, and determining whether the PN phase i of said PN code replica equals q;

if YES, ending the search;

if NO, updating the PN phase i of said PN code replica signal by 1, and continuing the search by repeating the foregoing steps using the phase updated PN code replica signal to multiply said transmitted PN code signal.

3. The method of claim 2 wherein said number C_N of PN phases to be searched, is determined by the steps of:

1) calculating false alarm probability P_F using the relationship (P_F (T_SC)=exp (-T_SC /EN),

where T_SC is said updated signal classification threshold value, and E_N is said noise estimate; and

2) obtaining C_N using ##EQU11## where k=1 or 2.

4. The method of claim 2 further comprising, when said second result exceeds said second signal detection threshold value T_SD2, the steps of updating said number C_N and resetting said count c_n to 0.

5. The method of claim 1 wherein said signal detection threshold comprises a first signal detection threshold value times (1-y), where y is between 1/16 and 1/8, and said step of evaluating whether said first result exceeds said signal detection threshold comprises the steps of:

evaluating whether said first result exceeds said first signal detection threshold value times (1-y); and

,

if YES, evaluating whether said first result exceeds said first signal detection threshold value;

if YES, changing said first signal detection threshold value to an updated signal detection threshold value equal to said first result, and updating said signal classification threshold to an updated signal classification threshold value equal to the average of said updated signal detection threshold value and said noise estimate;

if NO, retaining said first signal detection threshold value and said signal classification threshold as an updated signal detection threshold value and an updated signal classification threshold value, respectively; and

further comprising the steps of;

integrating the signal energy of said input signal over a second predetermined time period to obtain a second result;

comparing the value of said second result to a second signal detection threshold value; and

,

if said second result value is less than said second signal detection threshold value, then, comparing said second result value to a second signal classification threshold value; and

if said second result value is less than said second signal classification threshold value, then,

updating said noise estimate and said phase count c_n ;

comparing said updated phase count value with said first number of phases to be searched C_N ; and

if said updated phase count value is equal to or greater than said first number of phases C_N, ending the search;

else,

if the total number q of phases in said PN space has not been reached, updating the phase i of said local PN code signal, and then repeating said PN code signals combining step, said integrating step, and said comparing steps with the updated values;

if said total number q is reached, ending the search;

if said second result value is equal to or greater than said second signal detection threshold value, then,

changing said second signal detection threshold value to the value of said second result, and updating said second signal classification threshold value to a value equal to the average of the sum of the value of said second result and said noise estimate value, and updating said number of phases to be searched C_N, and

then, setting said count c_n of said PN phases to zero; and

,

if the total number q of phases in said PN space has not been reached, updating the phase i of said local PN code signal, and then repeating said PN code signals combining step, said integrating step, and said comparing steps with the updated values;

else, ending the search.

6. In an apparatus for PN code acquisition for CDMA-DSSS systems comprising:

means for receiving a transmitted PN code signal and multiplying said signal with a PN code replica signal to produce an input signal for use in determining the phase alignment of said PN code signals;

means for integrating the signal energy of said input signal over a time interval and producing an energy level output signal;

signal energy detector means for using said energy level output signal to determine a signal level threshold and producing a signal indicative of said threshold;

the improvement comprising;

a noise energy level estimator for producing a noise estimate signal;

signal classification means for using said signal level threshold signal and said noise estimate signal to produce a signal classification threshold;

comparator means for comparing said energy level output signal to said signal classification threshold and for producing a first signal when said energy level output signal is smaller and a second signal when it is not;

means, responsive to said first signal, for making said noise estimate signal equal to said energy level output signal; and

means, responsive to said second signal, for changing the phase of said PN code replica signal.

7. The apparatus as in claim 6 wherein said means for changing the phase of said PN code replica signal comprises:

means for storing a first number C_N of PN phases i to be searched and a total number q of PN phases;

counter means for maintaining a count c_n of said PN phases searched and producing a count signal in accordance therewith;

means for updating said count c_n by 1 in response to said second signal from said comparator means; and

means for comparing the value of said count signal with said number C_N and ending said search when said count signal value equals or exceeds C_N and for changing the phase of said PN code replica signal when it does not.

8. The apparatus as in claim 6 further comprising means for determining the number of PN phases to be searched, C_N, including:

means for calculating false alarm probability P_F using the relationship (P_F (T_SC)=exp (-T_SC /E_N), where T_SC is said signal classification threshold, and E_N is the value of said noise estimate signal; and

means for obtaining C_N by ##EQU12## where k=1 or 2.

9. A method for searching a pseudorandom noise (PN) code space containing a total number q of PN chips or phases i, comprising the steps of:

predetermining values for;

a first number of PN phases to be searched, C_N ;

a first signal detection threshold, T_SD1 ;

a first noise estimate, E_N ; and

a first signal classification threshold, T_SC1, having a value between said first signal detection threshold and said first noise estimate;

maintaining a count c_n of said PN phases; and

(a) combining a received PN code signal with a first local PN code signal to obtain an input signal;

(b) integrating the signal energy of said input signal over a first predetermined time period τ

₁ to obtain a first result Z₁ ;

(c) comparing the value of said first result Z₁ to said first signal detection threshold T_SD1 value times (1-y), where y is between 1/16 and 1/8; and

,

(i) if said first result Z₁ value is less than said first signal detection threshold T_SD1 value times (1-y), comparing said first result Z₁ value to said first signal classification threshold T_SC1 value;

(1) if said first result Z₁ value is less than said first signal classification threshold T_SC1 value;

then,

updating the values of said first noise estimate E_N and said phase count c_n ;

comparing the updated phase count value with said first number of phases to be searched C_N ; and

if said updated phase count value is equal to or greater than said first number of phases C_N, ending the search;

else,

if the total number of phases q in said PN space has not been reached, updating the phase i of said local PN code signal, and then repeating said (a,b,c) combining, integrating, and comparing steps with said updated values;

if said total number of phases q is passed, ending the search;

(2) if the value of said first result Z₁ is equal to or greater than said first signal classification threshold T_SC1, value, then, setting said count c_n of said PN phases to zero; and

,

if said total number of phases q in said PN space has not been reached, updating the phase i of said local PN code signal; and

then repeating said (a,b,c) combining, integrating, and comparing steps with said updated values;

if said total number of phases q is passed, ending the search;

(ii) if the value of said first result Z₁ is equal to or greater than said first signal detection threshold T_SD1 value times (1-y), comparing said first result value to said first signal detection threshold T_SD1 value; and

,

if said first result value is greater than said first signal detection threshold value, then, changing said first signal detection threshold value to the value of said first result Z₁, and updating said first signal classification threshold T_SC1 value to a value equal to the average of said updated signal detection threshold value and said first noise estimate E_N value;

else,

retaining said first threshold T_SD1 and T_SC1 values as said updated threshold values; and

(d) integrating the signal energy of said input signal over a second predetermined time period τ

₂ to obtain a second result Z₂ ;

(e) comparing the value of said second result Z₂ to a second signal detection threshold T_SD2 value; and

,

(i) if said second result Z₂ value is less than said second signal detection threshold T_SD2 value, then,

comparing said second result Z₂ value to a second signal classification threshold T_SC2 value; and

if said second result Z₂ value is less than said second signal classification threshold T_SC2 value, then,

updating the value of said first noise estimate E_N and said phase count c_n ;

comparing said updated phase count c_n value with said first number of phases to be searched C_N ; and

if said updated phase count c_n value is equal to or greater than said first number of phases C_N, ending the search;

else,

if the total number q of phases in said PN space has not been reached, updating the phase i of said local PN code signal, and then repeating said (a,d,e) combining, integrating, and comparing steps with the updated values;

if said total number q is passed, ending the search;

(ii) if the value of said second result Z₂ is equal to or greater than said second signal detection threshold T_SD2 value, then,

changing said second signal detection threshold T_SD2 value to the value of said second result Z₂, and updating said second signal classification threshold T_SC2 value to a value equal to the average of the values of said second result Z₂ and said first noise estimate E_N, and updating said first number of phases to be searched C_N ;

and

then, setting said count c_n of said PN phases to zero; and

,

if said total number q of phases in said PN space has not been reached, updating the phase i of said local PN code signal, and then repeating said (a,d,e) combining, integrating, and comparing steps with updated values;

else,

ending the search.

10. The method of claim 9 wherein when said first result Z₁ value is less than said first signal classification threshold T_SC1 value, the updated value of said first noise estimate E_N comprises said first result Z₁ value.

11. The method of claim 9 wherein when said second result Z₂ value is less than said second signal classification threshold T_SC2 value, the updated value of said first noise estimate E_N comprises said second result Z₂ value.

12. The method of claim 9 wherein said number of phases to be searched C_N, is determined by the steps of:

1) calculating the false alarm probability P_F using the relationships (P_F (T_SC2)=exp (-T_SC2 /E_N); and

2) obtaining C_N by ##EQU13## where k=1 or 2.

13. An apparatus for PN code acquisition in CDMA-DSSS systems, with automatic decision threshold, comprising:

means for receiving a transmitted PN code signal and multiplying said signal with a PN code replica signal to produce an input signal for use in determining the phase alignment of said PN code signals;

means for integrating the signal energy of said input signal over a time interval to obtain a first result;

means for storing a noise estimate value;

means for storing a signal detection threshold value greater than said noise estimate value;

means for storing a signal classification threshold value between said signal detection threshold and said noise estimate values;

means for storing a first number C_N of PN phases i to be searched and a total number q of PN phases;

means for maintaining a count c_n of said PN phases; and

means for searching for said alignment by detecting the signal energy levels and the noise levels of said input signal, comprising;

means for evaluating whether said first result exceeds said signal detection threshold and producing a first signal if YES and a second signal if NO;

means, responsive to said first signal, for setting the value of said signal detection threshold equal to said first result and updating said signal classification threshold to the average value of said first result and said noise estimate;

means, responsive to said second signal, for evaluating whether said first result exceeds said signal classification threshold and producing a third signal if NO and a fourth signal if YES;

means, responsive to said third signal, for updating the value of said noise estimate by the value of said first result and updating said count c_n by 1, and determining whether said updated count is equal to or greater than C_N and producing a fifth signal if YES and a sixth signal if NO;

means, responsive to said fifth signal, for ending the search;

means, responsive to said sixth signal, for updating the PN phase i of said PN code replica signal by 1, and determining whether said updated phase equals q, and producing a seventh signal if YES and an eighth signal if NO;

means, responsive to said seventh signal, for ending the search;

means, responsive to said eighth signal, for continuing the search by actuating said foregoing means to repeat the foregoing functional steps using the phase updated first result;

means, responsive to said fourth signal, for setting said count c_n to zero; and

means, responsive to the setting of said count c_n to zero, for checking the PN phase i with respect to the total number q, and producing a ninth signal if the PN phase i is not equal to q, and a tenth signal if the PN phase i is equal to q;

means, responsive to said ninth signal, for updating the PN phase i of said PN code replica signal by 1, and continuing the search by actuating said searching means to repeat the foregoing functional operations using the phase updated first result; and

means, responsive to said tenth signal, for ending the search.

14. The apparatus of claim 13 wherein said means for storing a noise estimate value comprises a noise energy level estimator.

15. The apparatus of claim 13 further comprising means for determining the number of PN phases to be searched, C_N, including:

means for calculating false alarm probability P_F using the relationship (P_F (T_SC)=exp (-T_SC /E_N), where T_SC is said updated signal classification threshold, and E_N is the value of said noise estimate signal; and

means for obtaining C_N using ##EQU14## where k=1 or 2.

16. A CDMA-DSSS system for acquiring a transmitted PN code signal, comprising:

means for receiving a transmitted PN code signal;

means for producing a cross correlation between said received PN code signal and a locally generated PN code, to achieve phase alignment thereof, and outputting a cross correlation signal indicative thereof;

first comparator means for comparing said cross correlation signal with a signal detection threshold and producing a first signal if said cross correlation signal exceeds the signal detection threshold, and a second signal if it does not;

means, responsive to said first signal from said first comparator means, for updating said signal detection threshold to the value of said cross correlation signal;

second comparator means, responsive to said second signal from said first comparator means, for comparing said cross correlation signal with a signal classification threshold, having a value between said signal detection threshold and a noise estimate, and producing a first signal if said cross correlation signal exceeds the signal classification threshold, and a second signal if it does not;

means, responsive to said first signal from said second comparator means, for updating the phase of said locally generated PN code unless a first predetermined phase limit has been reached;

counter means for maintaining a count of the updated phases of said locally generated PN code;

means, responsive to said second signal from said second comparator means, for updating said noise estimate and comparing the count of the updated phases of said locally generated PN code in said counter means and a second predetermined phase limit, and updating the phase of said locally generated PN code unless said second predetermined phase limit has been reached; and

means for ending the updating of the phase of said locally generated PN code when either said first or second predetermined phase limit is reached.

17. A system as in claim 16 further comprising means, responsive to said first signal from said second comparator means, for setting the count in said counter means to zero.

18. A system as in claim 17 further comprising means for setting the number of phases to be counted after said count in said counter means has been set to zero, and wherein said number is said second predetermined phase limit.

19. A method for searching a pseudorandom noise (PN) code space containing a total number of PN chips or phases, q, comprising the steps of:

predetermining a first number of PN phases to be searched C_N ;

predetermining a first signal detection threshold T_SD having a value comprising T_SD1 (1-y), y being between 1/16 and 1/8 and T_SD1 being a threshold value based on signal energy;

predetermining a first noise estimate E_N having a value less than that of said threshold value T_SD1 ;

predetermining a first signal classification threshold T_SC, having a value T_SC1 between that of said threshold value T_SD1 and that of said first noise estimate E_N ;

maintaining a count c_n of said PN phases including a first phase count;

(a) combining a received PN code signal with a first local PN code signal, having a phase i, to obtain an input signal;

(b) integrating the signal energy of said input signal over a first predetermined time period to obtain a first result Z₁ ;

(c) comparing the value of said first result Z₁ to said first signal detection threshold T_SD value, and,

(i) if said first result Z₁ value is less than said first signal detection threshold T_SD value, comparing said first result Z₁ value to said first signal classification threshold value T_SC1 ; and

if said first result Z₁ value is less than said first signal classification threshold T_SC value, then, updating said first noise estimate E_N and said first phase count, and

comparing said updated phase count with said first number of phases to be searched C_N, and if said updated phase count is equal to or greater than said first number C_N, ending the search;

else,

if the total number of phases q in said PN space has not been reached, updating the phase i of said local PN code signal, and then repeating said combining, integrating, and comparing steps with updated values;

when said total number of phases q is reached, ending the search;

(ii) if said first result Z₁ value is equal to or greater than said first signal detection threshold T_SD value, comparing said first result Z₁ value to said threshold value T_SD1 and if said first result Z₁ value exceeds said threshold value T_SD1, setting said threshold value T_SD1 equal to said first result Z₁ and updating said first signal detection threshold value T_SD1 value to the average value of said first result Z₁ and said first noise estimate E_N ; and

(d) integrating the signal energy of said input signal over a second predetermined time period to obtain a second result Z₂ ;

(e) comparing the value of said second result Z₂ to a second signal detection threshold value T_SD2, and,

if said second result Z₂ is equal to or greater than said second signal detection threshold value T_SD2, setting said second signal detection threshold value T_SD2 equal to said second result Z₂, resetting said PN phase count c_n to zero, and if said total number of phases q has not been reached, updating the phase i of said local PN code signal and combining it with said received PN code signal, and repeating said integrating, and comparing steps for said second predetermined time period; and

when said first number C_N, or said total number of phases q, has been reached, ending the search.

20. A method for searching a pseudorandom noise (PN) code space containing a total number q of PN chips or phases i, comprising the steps of:

predetermining values for;

a first number of PN phases to be searched C_N ;

a first signal detection threshold T_SD ;

a first noise estimate E_N ; and

a first signal classification threshold T_SC, having a value between that of said first signal detection threshold and that of said first noise estimate; and

maintaining a count c_n of said PN phases including a first phase count c_n1 ;

(a) combining a received PN code signal with a first local PN code signal, having a phase i, to obtain an input signal;

(b) integrating the signal energy of said input signal over a first predetermined time period to obtain a first result Z₁ ;

(c) comparing the value of said first result Z₁ to said first signal detection threshold value T_SD1 times (1-y), where y is between 1/16 and 1/8, and,

(i) if said first result Z₁ is less than said first signal detection threshold value T_SD1 times (1-y), comparing said first result to said first signal classification threshold value T_SC1 ;

(1) if said first result Z₁ is less than said first signal classification threshold value T_SC1, then,

updating said first noise estimate value E_N1 to E_N2 and said first phase count c_n1 to c_n2, and

comparing said updated phase count c_n2 with said first number of phases C_N to be searched, and

if said count c_n2 is equal to or greater than said number C_N, ending the search, else,

if the total number of phases q in said PN space has not been reached, updating the phase i of said local PN code signal to i+1, and then repeating said (a,b,c) combining, integrating, and comparing steps with updated values;

if said total number q is reached, ending the search;

(2) if said first result Z₁ is equal to or greater than said first signal classification threshold value T_SC1, then, setting said count c_n of said PN phases to zero and,

if said total number of phases q in said PN space has not been reached, updating the phase i of said local PN code signal to i+1, and then repeating said (a,b,c) combining, integrating, and comparing steps with updated values;

if said total number q is reached, ending the search;

(ii) if said first result Z₁ is equal to or greater than said first signal detection threshold value T_SD1 times (1-y), comparing the value of said first result Z₁ to the value of said first signal detection threshold value T_SD1, and,

if said first result Z₁ is greater than said first signal detection threshold value T_SD1, changing said first signal detection threshold value T_SD1 to said first result Z₁, and updating said first signal classification threshold value T_SC1 to a value equal to the average of the value of said first result Z₁ and said first noise estimate value E_N1, else,

retaining said first threshold values T_SD1 and T_SC1 ; and

(d) integrating the signal energy of said input signal over a second predetermined time period to obtain a second result Z₂ ;

(e) comparing said second result Z₂ to a second signal detection threshold value T_SD2, and,

(i) if said second result Z₂ is less than said second signal detection threshold value T_SD2, comparing said second result Z₂ to a second signal classification threshold value T_SC2, and

if said second result Z₂ is less than said second signal classification threshold value T_SC2, then, updating said first noise estimate E_N1 to E_N2 and said first phase count c_n1 to c_n2, and comparing said updated phase count c_n2 with said first number of phases C_N to be searched, and

if said updated phase count c_n2 is equal to or greater than said number C_N, ending the search, else,

if the total number of phases q in said PN space has not been reached, updating the phase i of said local PN code signal to i+1, and then repeating said (a,b,c,d,e) combining, integrating, and comparing steps with updated values;

when said total number q or said number C_N is reached, ending the search;

(ii) if said second result Z₂ is equal to or greater than said second signal detection threshold value T_SD2, changing said second signal detection threshold value T_SD2 to the value of said second result Z₂, and updating said second signal classification threshold value T_SC2 to a value equal to the average of said second result Z₁ and said first noise estimate value E_N1, and updating said number of phases C_N to be searched, and

then, setting said count c_n of said PN phases to zero and,

if said total number of phases q in said PN space has not been reached, updating the phase i of said local PN code signal to i+1, and then repeating said (a,b,c,d,e) combining, integrating, and comparing steps with updated values;

when said total number q or said number C_N is reached, ending the search.