Digital frequency-locked loop for use with staggered sampling systems

• US 4,723,216 A
• Filed: 08/01/1985
• Issued: 02/02/1988
• Est. Priority Date: 08/01/1985
• Status: Expired due to Fees

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Premerlani, William J.
• Primary Examiner(s): Krass, Errol A.
• Assistant Examiner(s): Trans, V. N.
• Application Number:
  US06/761,267
• Time in Patent Office: 915 Days
• Field of Search: 364/484, 324/78 D, 324/78 R, 324/78 F, 324/79 R, 324/79 D, 455/260, 455/183, 375/81, 375/120
• US Class Current: 702/126
• CPC Class Codes: H03D 2200/0025 Gain control circuits
• CPC Class Codes: H03D 2200/005 Analog to digital conversion
• CPC Class Codes: H03D 2200/0062 Computation of input samples, e.g. successive samples
• CPC Class Codes: H03D 3/006 by sampling the oscillations and further processing the samples, e.g. by computing techniques H03D3/007 takes precedence

Abstract

A digital frequency locked loop is disclosed for use with programmable signal processing apparatus wherein frequency locking is desired from staggered data samples. Sampling means are provided for sampling an input signal, at instants determined by the sampling signal, to provide N data signals from M cycles of the input signal wherein each data signal represents the magnitude of the input signal at the sampling instant. Signal processing means are provided for performing a unique approximation to the optimal mean square estimation analysis to provide a frequency deviation signal which is proportional to the difference in frequency between the sampling and input signals. The frequency deviation signal is used to modulate the period of the sampling signal such that the frequency of the sampling signal is maintained in constant proportion to the frequency of the input signal.

Claims

1. A method for providing a sampling signal having a first frequency corresponding to a first period wherein said first frequency is maintained in substantially constant proportion to a second frequency of an input signal, said method comprising the steps of:

(a) sampling said input signal at instants determined by said sampling signal to provide a plurality of data signals wherein each said data signal represents the magnitude of said input signal at the sampling instant;

(b) performing an approximation to the optimal mean square estimation analysis on said data values to minimize the sum of the squares of the errors between said data values and assumed values with respect to phase angle, amplitude and frequency to thereby provide a frequency deviation signal which is indicative of the frequency difference between said first and second frequencies; and

(c) modulating said first period in response to said frequency deviation signal such that said frequency deviation signal is minimized and, therefore, said first frequency is maintained in substantially constant proportion to said second frequency.

2. The method as recited in claim 1 wherein N of said data signals respectively designated x_n for n=0 through N-1 are provided over M cycles of said input signal and wherein said step of performing an approximation to the optical mean square estimation analysis comprises the substeps of:

providing four weighted sums from said data signals in accordance with the following expressions;

##EQU4## wherein;

a_n =cos [(2π

M/N) (n+1/2-N/2)];

b_n =sin [(2π

M/N) (n+1/2-N/2)];

c_n =(a_n /N).(n+1/2-N/2); and

d_n =(b_n /N).(n+1/2-N/2);

for n=0, 1, 2 . . . (N-1);

providing a difference value and an amplitude value from said four weighted sums in accordance with the following expressions;

space="preserve" listing-type="equation">D=-SUM.sub.1 SUM.sub.4 +SUM.sub.2 SUM.sub.3 +G (SUM.sub.2.sup.2 -SUM.sub.1.sup.2),

space="preserve" listing-type="equation">A=(SUM.sub.1.sup.2 +SUM.sub.2.sup.2),

wherein;

G=1/[2N sin (2π

/N]; and

providing said frequency deviation signal in accordance with the follow expression;

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

=(6·

D)/(λ

·

M·

A).

3. The method as recited in claim 2 further including the step of placing maximum and minimum limitations on said frequency deviation signal in accordance with the following expression:

space="preserve" listing-type="equation">-0.2≦

λ

≦

0.2,

such that said sampling period is modulated with a deviation signal within this range.

4. The method as recited in claim 1 wherein N of said data signals respectively designated x_n for n=0 through N-1 are provided over M cycles of said input signal and wherein said step of performing an approximation to the optimal mean square estimation analysis comprises the substeps of:

providing four weighted sums from said data signals in accordance with the following expressions;

##EQU5## wherein;

a_n =cos [(2π

M/N) (n+1/2-N/2)];

b_n =sin [(2π

M/N) (n+1/2-N/2)];

c_n =(a_n /N)·

(n+1/2-N/2); and

d_n =(b_n /N)·

(n+1/2-N/2);

for n=0, 1, 2 . . . (N-1);

providing a difference value from said four weighted sums in accordance with the following expression;

space="preserve" listing-type="equation">D=-SUM.sub.1 SUM.sub.4 +SUM.sub.2 SUM.sub.3 +G (SUM.sub.2.sup.2 -SUM.sub.1 .sup.2),

wherein;

G=1/[2N sin (2π

/N)];

providing said frequency deviation signal in accordance with the follow expression;

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

=(6·

D)/(π

·

M·

I),

wherein

I is equal to a constant value;

said method further comprising the steps of;

providing an amplitude value in accordance with the following formula;

space="preserve" listing-type="equation">A=SUM.sub.1.sup.2 +SUM.sub.2.sup.2 ; and

modulating the amplitude of said input signal in response to said amplitude value such that the amplitude of said input signal is maintained to equal said constant value I.

5. The method of claim 4 further comprising the step of placing maximum and minimum limitations on said frequency deviation signal in accordance with the following expression:

space="preserve" listing-type="equation">-0.2≦

λ

≦

0.2,

such that said sampling period is modulated with a deviation signal within this range.

6. Apparatus for providing a sampling signal having a first frequency corresponding to a first period wherein said first frequency is maintained in substantially constant proportion to a second frequency of an input signal, said apparatus comprising:

(a) means for sampling said input signal at instants determined by said sampling signal to provide a plurality of data signals wherein each said data signal represents the magnitude of said input signal at the sampling instant; and

(b) signal processing means for performing an approximation to the optimal mean square estimation analysis on said data values to minimize the sum of the squares of the errors between said data values and assumed values with respect to phase angle, amplitude and frequency to thereby provide a frequency deviation signal which is indicative of the frequency difference between said first and second frequencies, said signal processing means being further adapted to provide said sampling signal and to modulate said first period in response to said frequency deviation signal such that said frequency deviation signal is minimized and, therefore, said first frequency is maintained in substantially constant proportion to said second frequency.

7. Apparatus as recited in claim 6 wherein N of said data signals respectively designated x_n for n=0 through N-1 are provided over M cycles of said input signals and wherein said signal processing means is adapted to:

provide four weighted sums from said data signals in accordance with the following expressions;

##EQU6## wherein;

a_n =cos [(2π

M/N)(n+1/2-N/2)];

b_n =sin [(2π

M/N)(n+1/2-N/2)];

c_n =(a_n /N)·

(n+1/2-N/2); and

d_n =(b_n /N)·

(n+1/2-N/2);

for n=0, 1, 2 . . . (n-=1);

provide a difference value and an amplitude value from said four weighted sums in accordance with the following expressions;

space="preserve" listing-type="equation">D=-SUM.sub.1 SUM.sub.4 +SUM.sub.2 SUM.sub.3 +G (SUM.sub.2.sup.2 -SUM.sub.1.sup.2),

space="preserve" listing-type="equation">A=(SUM.sub.1.sup.2 +SUM.sub.2.sup.2),

wherein;

G=1/[2N sin (2π

/N)]; and

provide said frequency deviation signal in accordance with the follow expression;

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

=(6·

D)/(π

·

M·

A),

to thereby perform the approximation to the optimal mean square estimation analysis.

8. Apparatus as recited in claim 7 wherein said signal processing means is further adapted to place maximum and minimum limitations on said frequency deviation signal in accordance with the following expression:

space="preserve" listing-type="equation">-0.2≦

λ

≦

0.2,

such that said sampling period is modulated with a deviation signal within this range.

9. Apparatus as recited in claim 6 wherein N of said data signals respectively designated x_n for n=0 through N-1 are provided over M cycles of said input signal and wherein said signal processing means is adapted to:

provide four weighted sums from said data signals in accordance with the following expressions;

##EQU7## wherein;

a_n =cos [(2π

M/N)(n+1/2-N/2)];

b_n =sin [(2π

M/N)(n+1/2-N/2)];

c_n =(a_n /N).(n+1/2-N/2); and

d_n =(b_n /N).(n+1/2-N/2);

for n=0, 1, 2 . . . (N-1);

provide a difference value from said four weighted sums in accordance with the following expression;

space="preserve" listing-type="equation">D=-SUM.sub.1 SUM.sub.4 +SUM.sub.2 SUM.sub.3 +G (SUM.sub.2.sup.2 -SUM.sub.1.sup.2)

wherein;

G=1/[2N sin (2π

/N);

provide an amplitude value in accordance with the follow formula;

space="preserve" listing-type="equation">A=SUM.sub.1.sup.2 +SUM.sub.2.sup.2 ;

modulate the amplitude of said data signals in response to said amplitude value such that the amplitude of said input signal is maintained to equal a constant value I; and

provide said freuency deviation signal in accordance with the following expression;

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

=(6·

D)/(π

·

M·

I).

10. Apparatus as recited in claim 9 wherein said signal processing means is further adapted to place maximum and minimum limitations on said frequency deviation signal in accordance with the following expression:

space="preserve" listing-type="equation">-0.2≦

λ

≦

0.2,

such that said sampling period is modulated with a deviation signal within this range.