Frequency synchronizing method for a reference oscillator

• US 5,509,034 A
• Filed: 02/14/1994
• Issued: 04/16/1996
• Est. Priority Date: 02/14/1994
• Status: Expired due to Fees

• Current Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: Beukema, Troy J.
• Primary Examiner(s): Chin, Stephen
• Assistant Examiner(s): VO, DON NGUYEN
• Application Number:
  US08/196,624
• Time in Patent Office: 792 Days
• Field of Search: 375/343, 375/344, 375/346, 375/354, 375/362, 375/365, 375/368, 455/161, 455/164.1, 455/164.2, 455/182.2, 455/192.2
• US Class Current: 375/344
• CPC Class Codes: H03J 7/02 Automatic frequency control H03J7/18 takes precedence

Abstract

A method and apparatus of frequency synchronization is described for a reference oscillator. Using a reference oscillator set to a first frequency, the process begins by scanning for a radio frequency carrier modulated with a predetermined pattern. The radio frequency carrier is located such that a received pattern comprising a frequency shifted version of the predetermined pattern is identified and time synchronization is established. The received pattern, the predetermined pattern, and stepped alterations to the first frequency are combined to provide a set of results that represent pattern correlation versus reference oscillator frequency. The first frequency is offset using the set of results, such that frequency synchronization is improved.

Claims

1. A method comprising the steps of:

using a reference oscillator set to a first frequency, scanning for a radio frequency carrier modulated with a predetermined pattern;

locating the radio frequency carrier such that a received pattern comprising a frequency shifted version of the predetermined pattern is identified and time synchronization is established;

combining the received pattern, the predetermined pattern, and stepped alterations to the first frequency to provide a set of results that represent pattern correlation versus frequency, wherein the stepped alterations comprise at least two frequency translated versions of the received pattern;

interpolating the set of results using a centreid formula to offset the first frequency, such that frequency synchronization is improved.

2. The method of claim 1, wherein the step of locating includes performing the calculation:

##EQU4## where ρ

is a correlation coefficient;

n is an integer index;

m is an integer index;

N is a width of the predetermined pattern;

z is a complex data sequence;

y is a length N data sequence representing the predetermined pattern.

3. The method of claim 1, wherein the step of combining further comprises the step of performing the calculation:

space="preserve" listing-type="equation">z.sub.y (n)=z(n.sub.o +n)y(n), n={0 . . . N-1},

where

n is an integer index;

n_o is time index;

N is a width of the predetermined pattern;

z is a complex data sequence;

y is a length N data sequence representing the predetermined pattern.

4. The method of claim 1, wherein the step of combining further comprises the step of performing the calculation:

##EQU5## where ρ

is a correlation coefficient;

n_o is a time index;

k is an integer index;

m is an integer index;

N is a width of the predetermined pattern;

M is a positive integer;

z is a complex data sequence;

y is a length N data sequence representing the predetermined pattern.

5. The method of claim 1, wherein the step of interpolating further comprises finding the frequency offset, f_o, such that:

##EQU6## where ρ

is a correlation coefficient;

k_o is a frequency index;

fs is a sample rate of the received pattern;

M is a positive integer.

6. A method comprising the steps of:

using a reference oscillator set to a first frequency, scanning for a radio frequency carrier modulated with a predetermined pattern;

locating the radio frequency carrier such that a received pattern comprising a frequency shifted version of the predetermined pattern is identified and time synchronization is established;

combining the received pattern with the predetermined pattern to provide a first result;

altering the first result as a function of stepped alterations to the first frequency to provide a set of second results that represent pattern correlation versus frequency, wherein the stepped alterations comprise at least two frequency translated versions of the received pattern;

interpolating the set of second results using a centroid formula to offset the first frequency, such that frequency synchronization is improved.

7. The method of claim 6, wherein the step of locating includes performing the calculation:

##EQU7## where ρ

is a correlation coefficient;

n is an integer index;

m is an integer index;

N is a width of the predetermined pattern;

z is a complex data sequence;

y is a length N data sequence representing the predetermined pattern.

8. The method of claim 6, wherein the step of combining further comprises the step of performing the calculation:

space="preserve" listing-type="equation">z.sub.y (n)=z(n.sub.o +n) y(n), n={0 . . . N-1},

where

n is an integer index;

n_o is a time index;

N is a width of the predetermined pattern;

z is a complex data sequence;

y is a length N data sequence representing the predetermined pattern.

9. The method of claim 6, wherein the step of altering further comprises the step of performing the calculation:

##EQU8## where ρ

is a correlation coefficient;

n_o is a time index;

.

k is an integer index;

m is an integer index;

N is a width of the predetermined pattern;

M is a positive integer;

z is a complex data sequence;

y is a length N data sequence representing the predetermined pattern.

10. The method of claim 6, wherein the step of interpolating further comprises finding the frequency offset, f_o, such that:

##EQU9## where ρ

is a correlation coefficient;

k_o is a frequency index;

fs is a sample rate of the received pattern;

M is a positive integer.

11. A method comprising the steps of:

performing wideband time correlation on a received signal and an expected signal;

detecting time synchronization by comparing the received signal to the expected signal;

finding a peak time for correlating the received signal to the expected signal;

scaling the received signal to provide a scaled received signal;

stepped frequency correlating the scaled received signal and the expected signal to provide a set of values representing pattern correlation as a function of frequency;

verifying time synchronization;

determining a peak frequency for correlating the scaled received signal to the expected signal;

fine frequency interpolating the set of values using a centroid formula, thereby yielding a frequency offset;

adjusting a variable frequency oscillator by the frequency offset.

12. The method of claim 11, wherein the step of performing wideband time correlation includes performing the calculation:

##EQU10## where ρ

is a correlation coefficient;

n is an integer index;

m is an integer index;

N is a width of the expected signal;

z is a complex data sequence;

y is a length N data sequence representing the expected signal.

13. The method of claim 11, wherein the step of detecting further comprises correlating and comparing ρ

(n)>

thresh1, where thresh1 is the correlation threshold in the range 0 to 1, ρ

is a correlation coefficient, and n is an integer index.

14. The method of claim 11, wherein the step of finding further comprises the step of finding the index n_o, such that

15. The method of claim 11, wherein the step of scaling further comprises the step of performing the calculation:

space="preserve" listing-type="equation">z.sub.y (n)=z(n.sub.o +n)y(n), n={. . . N-1},

where

n is an integer index;

n_o is a time index;

N is a width of the expected signal;

z is a complex data sequence;

y is a length N data sequence representing the expected signal.

16. The method of claim 11, wherein the step of stepped frequency correlating further comprises the step of performing the calculation:

##EQU11## where ρ

is a correlation coefficient;

n_o is a time index;

k is an integer index;

m is an integer index;

N is a width of the expected signal;

M is a positive integer;

z is a complex data sequence;

y is a length N data sequence representing the expected signal.

17. The method of claim 11, wherein the step of verifying further comprises the comparing ρ

(k)>

thresh2, where thresh2 is a correlation threshold in the range 0 to 1, ρ

is a correlation coefficient, and k is an integer index.

18. The method of claim 11, wherein the step of determining a peak frequency further comprises the step of finding k_o such that:

space="preserve" listing-type="equation">ρ

(k.sub.o -1)<

ρ

(k.sub.o)>

ρ

(k.sub.o +1),

where ρ

is a correlation coefficient.

19. The method of claim 11, wherein the step of fine frequency interpolating further comprises finding the frequency offset, f_o, such that:

##EQU12## where ρ

is a correlation coefficient;

k_o is a frequency index;

fs is a sample rate of the received pattern;

M is a positive integer.

20. An apparatus for synchronizing with respect to a carrier frequency that is modulated with a predetermined pattern, the apparatus comprising:

an oscillator having an output and a control input;

a radio transceiver coupled to the output of the oscillator and having a received signal output;

a pattern detector, coupled to the received signal output, arranged and constructed to detection off a frequency shifted version of the predetermined pattern;

a combiner coupled to the pattern detector and having an output that provides a combined version of the predetermined pattern and the frequency shifted version of the predetermined pattern;

a correlator coupled to the output of the combiner and the received signal output and having an output that provides a set of results that represents pattern correlation versus frequency;

an oscillator controller that is coupled to the output of the correlator and the control input of the oscillator, such that the oscillator is controlled, at least in part, as an interpolation, using a centroid formula, of the set of results that represent pattern correlation versus frequency.

21. The apparatus of claim 20, wherein the pattern detector performs the calculation:

##EQU13## where ρ

is a correlation coefficient;

n is an integer index;

m is an integer index;

N is a width of the predetermined pattern;

z is a complex data sequence;

y is a length N data sequence representing the predetermined pattern.

22. The apparatus of claim 20, wherein the combiner performs the calculation:

space="preserve" listing-type="equation">z.sub.y (n)=z(n.sub.o +n)y(n), n={0 . . . N-1},

where

n is an integer index;

n_o is a time index;

N is a width of the predetermined pattern;

z is a complex data sequence;

y is a length N data sequence representing the predetermined pattern.

23. The apparatus of claim 20, wherein the correlator performs the calculation:

##EQU14## where ρ

is a correlation coefficient;

n_o is a time index;

k is an integer index;

m is an integer index;

N is a width of the predetermined pattern;

M is a positive integer;

z is a complex data sequence;

y is a length N data sequence representing the predetermined pattern.

24. The apparatus of claim 20, wherein the oscillator controller finds a frequency offset, f_o, such that:

##EQU15## where ρ

is a correlation coefficient;

k_o is a time index;

fs is a sample rate of the received signal output;

M is a positive integer.

25. The apparatus of claim 20, wherein the oscillator, the radio transceiver, the pattern detector, the combiner, the correlator, and the oscillator controller are disposed in a communication unit.