Method and device relating to supervision and control of an oscillator signal

• US 6,147,560 A
• Filed: 01/26/1998
• Issued: 11/14/2000
• Est. Priority Date: 01/28/1997
• Status: Expired due to Term

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Rizell, Svenolov, Erikmats, Osten E., Erhage, Lars I., Karlsson, H.ang.kan L.
• Primary Examiner(s): KINKEAD, ARNOLD M
• Application Number:
  US09/013,220
• Time in Patent Office: 1,023 Days
• Field of Search: 331/74, 331/1 R, 331/12, 331/4, 342/194, 342/199, 342/200, 375/375, 455/260, 455/255, 329/304, 329/308, 329/307
• US Class Current: 331/1.00R
• CPC Class Codes: G01S 13/282 using a frequency modulated carrier wave G01S13/286 takes precedence
• CPC Class Codes: G01S 13/34 using transmission of continuous, frequency-modulated waves while heterodyning the received signal, or a signal derived therefrom, with a locally-generated signal related to the contemporaneously transmitted signal
• CPC Class Codes: G01S 7/4008 of transmitters
• CPC Class Codes: H03B 23/00 Generation of oscillations periodically swept over a predetermined frequency range angle-modulating circuits in general H03C3/00
• CPC Class Codes: H03C 3/08 Modifications of modulator to linearise modulation, e.g. by feedback, and clearly applicable to more than one type of modulator
• CPC Class Codes: H03L 7/02 using a frequency discriminator comprising a passive frequency-determining element

Abstract

The present invention relates to methods and devices for such control and supervision of an oscillator signal from a controllable oscillator that is done mainly to control the frequency variation of the oscillator signal. According to the invention, the controllable oscillator is controlled by a controlling voltage, which in turn is modified by a correction signal, generated in a control loop. A time discrete representation of a secondary phase is generated in the control loop, the secondary phase corresponding to a frequency being the difference between the frequency of the oscillator signal and a constant frequency. A time discrete approximation signal is generated in dependence of the time discrete representation of the secondary phase. A time discrete error signal is generated in dependence of the time discrete approximation signal, the time discrete error signal indicating the difference between the actual frequency slope of the oscillator signal and a desired frequency slope. The correction signal is generated in dependence of the time discrete error signal. The control loop can also be adaptive, meaning that data from one control sequence is being used in a later control sequence.

Claims

1. Method for control of an oscillator signal (cos(φ

(t))) from a controllable oscillator (13) during a time period (τ

) when the frequency (f(t)) of the oscillator signal (cos(φ

(t))) is intended to vary linearly with time corresponding to a predetermined frequency slope (μ

₀), whereat the oscillator (13) is controlled by a control signal (V(t)) which is modified during the time period (τ

) in dependence of a correction signal (K(t)) characterised by;

a) quadrature demodulation of the oscillator signal (cos(φ

(t))) with respect to a constant frequency (ω

₀ /2π

), whereby a first quadrature signal (I(t)) (in-phase) exhibiting a secondary phase (θ

(t)) and a second quadrature signal (Q(t)) (quadrature phase) exhibiting a phase difference π

/2 to the secondary phase (θ

(t)) are generated;

b) A/D-conversion (31,29) of the quadrature signals (the in-phase and the quadrature phase signals) (I(t), Q(t)) at predetermined points of time (t_k ;

k=0,1, . . . , N) during the time period (τ

), whereby time discrete quadrature signals (I_k, Q_k ;

k=0,1, . . . , N) are generated;

c) generation of a time discrete approximation signal (Z_k ;

k=0, 1, . . . N) in dependence of the time discrete quadrature signals (I_k, Q_k), such that the signal value (Z_n) of the time discrete approximations signal, corresponding to given point of time (t_n) represents an approximation of the second time derivative (θ

(t_n)) of the secondary phase (On) at the given point of time (t_n);

d) generation of a time discrete error indication signal (e_k ;

k=0,1, . . . , N) in dependence of the time discrete approximation signal (Z_k), such that the signal value (e_n) of the time discrete error indication signal, corresponding to a given point of time (t_n), indicates the deviation of the frequency slope (μ

(t_n)) of the oscillator signal (cos(φ

(t))) at the given point of time (t_n) from the predetermined frequency slope (μ

₀);

e) generation of the correction signal (K(t)) in dependence of the time discrete error indication signal (e_k).

2. Method according to claim 1, characterised in that:

f) step c) comprises generation of a time discrete differential signal (Y_k ;

k=0,1, . . . , N) in dependence of the time discrete quadrature signals (I_k, Q_k), such that signal value (Y_n) of the time discrete differential signal, corresponding to a given point of time (t_n), represents a difference (Δ

^- θ

_n n) between the secondary phase (θ

_n) at the given point of time (t_n) and the secondary phase (θ

_n-1) at the immediately preceding point of time (t_n-1);

g) and that step c) comprises that the generation of the time discrete approximation signal (Z_k) is performed in dependence of the time discrete differential signal (Y_k), such that the signal value (Z_n) of the time discrete approximation signal, corresponding to a given point of time (t_n), represents a difference (Δ

^2- θ

_n) between the difference (Δ

^- θ

_n) which is represented by the signal value (Y_n) of the time discrete differential signal corresponding to the given point of time (t_n), and the difference (Δ

^- θ

_n-1), which is represented by the signal value (Y_n-1) of the time discrete differential signal corresponding to the immediately preceding point of time (t_n-1).

3. Method according to claim 2 characterised in that:

h) step f) comprises generation of a first time discrete complex signal (X_k ;

k=0,1, . . . , N), such that the signal value (X_n) of the first time discrete complex signal corresponding to a given point of time (t_n), exhibits a real part and an imaginary part, resp., corresponding to said first and said second time discrete quadrature signals (I_n, Q_n) at the given point in time (t_n);

i) and that step f) comprises that the generation of the time discrete differential signal (Y_k) is performed such that its signal value (Y_n), corresponding to a given point of time (t_n), corresponds to the value of a multiplication (35) of the signal value (X_n) of the first time discrete complex signal corresponding to the given point of time (t_n) and the signal value (X*_n-1) of the complex conjugate of the first time discrete complex signal corresponding to the immediately preceding point of time (t_n-1).

4. Method according to claim 3, characterised in that:

j) step g) comprises that the generation of the time discrete approximation signal (Z_k) is performed such that its signal value (Z_n), corresponding to a given point of time (t_n), corresponds to the value of a multiplication (39) of the signal value (Y_n) of the time discrete differential signal, corresponding to the given point of time (t_n), and the signal value (Y_n-1 *) of the complex conjugate of the time discrete differential signal, corresponding to the immediately preceding point of time (t_n-1).

5. Method according to claim 4 characterised in:

k) storing of stored correction values (K_k ;

k=0,1, . . . , N) corresponding to each one of the points of time ((t_k)), whereby the storing is performed preceding the start of the time period (τ

);

l) and that step e) comprises that the generation of the correction signal (K(t)) is performed also in dependence of the stored correction values (K_k).

6. Method according to claim 5 characterised in that:

m) step l) comprises time discrete low pass filtering (49) of the time discrete error indication signal (e_k),

whereby a first time discrete low pass signal (LP1(e_k);

k=0,1, . . . , N) is generated;

n) step l) comprises time discrete low pass filtering (51) of the stored correction values (K_k), whereat a second time discrete low pass signal (LP2(K_k);

k=0,1, . . . , N) is generated;

o) step l) comprises generation of new correction values (K_k ;

k=0,1, . . . , N) in dependence of the time discrete first and second low pass signals (LP1(e_k), LP2(K(t));

p) and step l) comprises generation of the correction signal by a D/A-conversion (55) of the new correction values (K_k).

7. Method according to claim 6, characterised in that:

q) the time discrete low pass filtering (51) in step m) is a causal FIR-filtering.

8. Method according to claim 7, characterised in that:

r) the time discrete low pass filtering (51) in step n) is a FIR-filtering.

9. Method according to claim 8 characterised in that:

s) step o) comprises that the new correction value (K_n), corresponding to a given point of time (t_n), is generated so as to correspond to the value of the addition (53) of the signal values (LP1(e_n), LP2(K_n)) of the first and the second time discrete low pass signals corresponding to the given point of time (t_n).

10. Method according to claim 9, characterised in that:

t) modification of the stored correction values (K_k), such that the stored correction value (K_n), corresponding to a given point of time (t_n), is modified such as to correspond to the new correction value (K_n), corresponding to the given point of time (t_n).

11. A device for generation of an of an oscillator signal (cos(φ

(t))) having a predetermined frequency variation (f_D (t)), whereby the frequency variation (f_D (t)) comprises at least one time period (τ

) during which the frequency (f(t)) is intended to vary linearly with time corresponding to one for the time period (τ

) predetermined frequency slope (μ

₀), comprising;

a controllable oscillator (71), having a control signal input (75) and an oscillator signal output (73), whereby the controllable oscillator (71) is arranged to emit, via the oscillator signal output(73), an oscillator signal (cos(φ

(t))), the frequency of which (f(t)) depends on a control signal (V(t)) received via the control signal input (75);

a control signal generator (77), which in turn comprises a control signal output (79) and a correction signal input (81), the control signal output (79) being connected to the control signal input (75) of the controllable oscillator (71), whereby the control signal generator (77) is adapted to emit via the control signal output (79) the control signal (V(t)) which is modified in dependence of a via the correction signal input (81) received correction signal (K(t)), characterised in that;

the device comprises means (87,89) for quadrature demodulation of the oscillator signal (cos(φ

(t))) with respect to a constant frequency ω

₀ /2π

), the means for quadrature modulation (87,89), comprising a first and a second quadrature signal output (93,95), whereby the means for quadrature modulation (87,89) are adapted to emit a first quadrature signal (I(t)) (an in-phase) having a secondary phase (θ

_n) and a second quadrature signal (Q(t)) (a quadrature phase), having a phase separated Π

/2 from the secondary phase (ν

_n), via the first and a second quadrature signal output (93,95);

the device comprises means (97,99) for A/D-conversion of the quadrature signals (I(t), Q(t)), whereby the means (97,99) for A/D-conversion are adapted to A/D-conversion at a number of points of time (t_k ;

k=0,1, . . . , N) during the time period (τ

) and to thereby generate time discrete quadrature signals (I_k, Q_k ;

k=0,1, . . . , N);

the device comprises means (97) for generating a time discrete approximation signal (Z_k ;

k=0,1, . . . , N) in dependence of the time discrete quadrature signals (I_k, Q_k), such that the signal value (Z_n) of the time discrete approximations signal, corresponding to given point of time (t_n), represents an approximation of the second time derivative (θ

(t)) of the secondary phase at the given point of time (t_n);

the device comprises means (97) for generation of a time discrete error indication signal (e_k ;

k=0,1, . . . , N) in dependence of the time discrete approximation signals (Z_k), such that the signal value (e_n) of the time discrete error indication signal, corresponding to a given point of time (t_n) indicates the deviation of the frequency slope (μ

(t_n)) of the oscillator signal (cos(φ

(t))) at the given point of time (t_n) from the predetermined frequency slope (μ

₀); and

that,

the device comprises means (97,99) for generation of the correction signal (K(t)) in dependence of the time discrete error indication signal (e_k), whereby the means (97,99) for generation of the correction signal (K(t)), are connected to the correction signal input (81).

12. The device according to claim 11, characterised in that:

the device comprises a control unit (97), having memory means, processor means, clock means, and program; and

that the device comprises an analog input/output unit (99), having a first and a second analog input port (101,103) and an analog output port (105), wherein the first and the second analog input port (101,103), respectively, are connected to said first and said second quadrature signal output (93,95) and the output port (105) is connected to the correction signal input (81), whereat the input/output unit (99) is connected to the control unit (97).

13. The device according to claim 12, characterised in that the means (97,99) for A/D-conversion of the quadrature signals (I(t), Q(t)) consist of the control unit (97) in combination with the input/output unit (99).

14. The device according to claim 12, characterised in that the means (97) for generation of the time discrete approximation signal (Z_k) consist of the control unit (97).

15. The device according to claim 12, characterised in that the means (97) for generation of time discrete error indication signal (e_k) consist of the control unit (97).

16. The device according to claim 12, characterised in that the means (97,99) for generation of the correction signal (K(t)) consist of the control unit in combination with the input/output unit (99).

17. The device according to claim 11, characterised in that the means (87,89) for quadrature demodulation of the oscillator signal (cos(φ

(t))) comprises a quadratur. demodulator (89), having a signal input (91) and the quadrature signal outputs (93,95); and

that the means (87,89) for quadrature demodulation of the oscillator signal (cos(φ

(t))) comprises a measurement means (87) for the measurement of the oscillator signal (cos(φ

(t))), whereby the measurement means are connected to the oscillator signal output (73) and also connected to the quadrature demodulator (89) signal input (91).

18. The device according to claim 17, characterised in that the measurement means (87) is a coupler (87).

19. A method for establishing of an oscillator signal frequency slope characterized in:

a) quadrature demodulation of the oscillator signal (cos(φ

(t))) with respect to a constant frequency (ω

₀ /2π

), whereby a first quadrature signal (I(t)) (in-phase) exhibiting a secondary phase (θ

(t)) and a second quadrature signal (Q(t)) (quadrature phase) exhibiting a phase difference π

/2 to the secondary phase (θ

(t)) are generated;

b) A/D-conversion (31,29) of the quadrature signals (I(t), Q(t)) at a number of points of time (t_k), whereby time discrete quadrature signals (I_k, Q_k) are generated;

c) generation of a time discrete approximation signal (Z_k), in dependence of the time discrete quadrature signals (I_k, Q_k) , such that the signal value (Z_n) of the time discrete approximation signal, corresponding to a given time position (t_n), represents an approximation of the second time derivative (θ

(t_n)) of the secondary phase (θ

_n) at the given point of time (t_n);

whereby the approximation signal (Z_k) establishes the frequency slope (μ

(t)) of the oscillator signal (cos(φ

(t))), since-this frequency slope (μ

(t)) is directly related to the second time derivative θ

(t)) of the secondary phase (θ

(t)).

20. The method according to claim 19, characterised in that:

d) step c) comprises generation of a time discrete differential signal (Y_k) in dependence of the time discrete quadrature signals (I_k, Q_k) such that the signal value (Y_n) of the time discrete differential signal, corresponding to a given point of time (t_n) represents a difference (Δ

^- θ

_n) between the secondary phase (θ

_n) at the given point of time (t_n) and the secondary phase (θ

_n-1) at the immediately preceding point of time (t_n-1)

e) and that step c) comprises that the generation of the time discrete approximation signal (Z_k) is performed in dependence of the time discrete differential signal (Y_k), such that the signal value (Z_n) of the time discrete approximation signal, corresponding to a given point of time (t_n), represents a difference ((Δ

^-)² θ

_n) between the difference (Δ

^- θ

_n) , which is represented by the signal value (Y_n) of the time discrete differential signal, corresponding to the given point of time (t_n), and the difference (Δ

^- θ

_n-1), which is represented by the signal value (Y_n-1) of the time discrete differential signal corresponding to the immediately preceding point of time (t_n-1).

21. The method according to claim 20, characterised in that:

.

f) step d) comprises generation of a first time discrete complex signal (X_k), such that the the signal value (X_n) of the first time discrete complex signal corresponding to a given point of time (t_n) exhibites a real and an imaginary part corresponding to the time discrete first and second quadrature signal values (I_n,Q_n) at the given point of time (t_n);

g) and that step d) comprises that the generation of the time discrete differential signal (Y_k) is performed such that its signal value (Y_n), corresponding to a given point of (t_n), corresponds to the value of a multiplication (35) of the signal value (X_n) of the first time discrete complex signal, corresponding to given point of time (t_n), and the signal value (X*_n-1) of the complex conjugate of the first time discrete complex signal corresponding to the immediately preceding point of time (t_n-1).

22. The method according to claim 21, characterized in that:

h) step e) comprises that the generation of the time discrete approximation signal (Z_k) is performed such that its signal value (Z_n), corresponding to a given point of time (t_n), corresponds to the value of a multiplication (39) between the signal value (Y_n) of the time discrete differential signal, corresponding to the given point of time (t_n), and the complex conjugate of the signal value (Y_n-1 *) of the time discrete complex signal, corresponding to the immediately preceeding point of time (t_n-1).