Method and device for compensating phase delays

• US 6,721,375 B1
• Filed: 07/17/2000
• Issued: 04/13/2004
• Est. Priority Date: 11/23/1998
• Status: Expired due to Fees

• Current Assignee: Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Hammel, Christof
• Primary Examiner(s): Vo, Don N.
• Application Number:
  US09/600,466
• Time in Patent Office: 1,366 Days
• Field of Search: 375/350, 375/316, 375/355, 708/313, 341/126, 341/155
• US Class Current: 375/350
• CPC Class Codes: G01D 3/02 with provision for altering or correcting the law of variation
• CPC Class Codes: H03H 17/0219 Compensation of undesirable effects, e.g. quantisation noise, overflow stability problems H03H17/0461

Abstract

A method for compensation of phase delays that occur in discrete sampling of an analog output signal (U_S) has the steps of subjecting the output signal (U_S) to signal processing by an anti-aliasing filter and then by a sampling member to obtain a sampling signal (U_A) that has a sampling period (T), compensating a phase delay of the sampling signal caused by the sampling member, taking into account the sampling signal of a preceding sampling period to obtain a signal (U_F,K), and compensating of the phase delay of the sampling signal caused by the anti-aliasing filter, taking into account a filter time constant (T_F) of the anti-aliasing filter, a sampling period (T_A) of the sampling member, and the sampling signal of the preceding sampling period, to obtain a signal (U_SK).

Claims

1. A method for compensation of phase delays that occur in discrete sampling of an analog output signal (U_S), comprising the steps of subjecting the output signal (U_S) to signal processing by an anti-aliasing filter and then by a sampling member to obtain a sampling signal (U_A) that has a sampling period (T);

compensating a phase delay of the sampling signal caused by the sampling member;

taking into account the sampling signal of a preceding sampling period to obtain a signal (U_F,K); and

compensating the phase delay of the sampling signal caused by the anti-aliasing filter, taking into account a filter time constant (T_F) of the anti-aliasing filter, a sampling period (T_A) of the sampling member, and the sampling signal of the preceding sampling period, to obtain a signal (U_SK).

2. A method as defined in claim 1;

and further comprising, in the case of a further phase delay of the sampling signal (U_A) caused by an idle time (T_T) which occurs in further processing of the sampling signal and by which an idle-time-delayed sampling signal (U_T) is generated, performing a compensation of the phase delay of the sampling signal (U_T) caused by the idle time, taking into account the idle time (T_T) the sampling period of the sampling member, and the sampling signal of the preceding sampling period, to obtain an idle-time-compensated sampling signal (U_A,K) and subjecting the idle-time-compensated sampling signal (U_A,K) to the phase shift compensation of claim 1 for the sake of further phase compensation.

3. A method as defined in claim 1;

and further comprising correcting the phase delay caused by the sampling member, which delay can be represented by the equation ${\mathrm{\varphi <br><br>}}_A=\mathrm{\omega <br><br>}·<br><br>\frac{1}{2}·<br><br>T_A=-2·<br><br>\mathrm{\pi <br><br>}·<br><br>f·<br><br>\frac{T_A}{2}=-\mathrm{\pi <br><br>}·<br><br>f·<br><br>T_A$

which f is the signal frequency and T_A is the sampling period of the sampling member, by means of a linear extrapolation using a correction value of the form −

y_k,2=X_k+½

·

(x_k−

x<sub>k−

1</sub>), in which x<sub>k−

1</sub>, x_k are values of the sampling period in successive sampling periods T<sub>T,k−

1</sub>, T_T,k.

4. A method as defined in claim 1;

and further comprising correcting a phase delay which is caused by the anti-aliasing filter and can be represented in the form φ

_F=arctan(2·

π

·

f·

T_F), in which T_F is the filter time constant of the anti-aliasing filter, by a correction value of the form $y_{K,3}=x_K+\frac{T_F}{T_A}·<br><br>\left(x_k-x_{k-1}\right).$

5. A method as defined in claim 1;

and further comprising correcting the phase delay which is caused by the idle time (T_T) and can be represented in the form φ

_V=−

arctan(2π

ƒ

_F)−

π

fT_A−

2π

ƒ

T_T, by a correction value of the form ${\mathrm{\gamma <br><br>}}_{k,1}=x_k<br><br>\left(1+\frac{T_T}{T_A}\right)-x_{k-1}·<br><br>\frac{T_T}{T_A}.$

6. A method as defined in claim 1;

and further comprising performing the individual phase delay compensations successively, so that for correcting a sampling signal (U_A, U_T) during one sampling period, sampling signals of two or three preceding sampling periods are taken into account.

7. A method as defined in claim 1;

and further comprising performing the individual phase delay compensations in reverse order from the order in which they occurred.

8. An arrangement or compensation of phase delays that occur in discrete-time sampling of an output signal (U_S) in which the output signal (U_S) is subjected to signal processing by an anti-aliasing filter to obtain a signal (U_F) and then by a sampling member for obtaining a sampling signal (U_A), the arrangement comprising means for compensation for a phase delay of the sampling signal caused by the sampling member, taking into account the sampling signal of a preceding sampling period;

and means for compensation of the phase delay of the sampling signal caused by the anti-aliasing filter, taking into account a filter time constant (T_F) of the anti-aliasing filter, the sampling period (T_T) of the sampling member, and the sampling signal of preceding sampling period.

9. An arrangement as defined in claim 8;

and further comprising means for compensation of a sampling signal phase delay which is caused by an idle time (T_T) occurring in a further processing of the sampling signal by which a signal (U_T) is obtained, taking into account the idle time (T_T), the sampling period (T_A) and the sampling signal of the preceding sampling period, to obtain an idle-time-compensated sampling signal (U_A,K); and

means for delivering this idle-time-compensated sampling signal (U_A,K) to the phase delay mean for further time compensation.

10. An arrangement as defined in claim 8, wherein said means for compensation of the phase delay caused by the sampling member, the anti-aliasing filter or the further processing idle time (T_T), each have a number of parallel signal transmission channels, in which a first signal transmission channel serves to transmit an unaltered sampling signal, a second signal transmission channel having a multiplier member serves to transmit a sampling signal multiplied by a factor, and a third signal transmitting channel having a delay member and a multiplier member serves to transmit a sampling signal delayed by one sampling period and multiplied by factor, to a summation member.

11. An arrangement as defined in claim 10, wherein said means for compensation of the phase delay caused by the sampling member, the anti-aliasing filter, or the further processing idle time (T_T) are connected in series with one another in such a way that a phase-compensated output signal (U_S,K) is obtained, taking into account the uncompensated sampling signals (U_T) of the preceding three sampling periods.

12. An arrangement as defined in claim 8, wherein a number of series-connected delay members, multiplier members connected to them and parallel to one another, and a summation member for adding the output signals of the multiplier members to achieve a phase compensation are provided, to achieve a phase compensation for obtaining a phase-compensated sampling signal (U_S,k) of the form U_S,K(k)=A·

U_T(k)+B·

U_T(k−

1)+C·

U_T(k−

1)+D·

U_T(k−

3), in which U_T(k−

i), where l=0, 1, 2, 3 represents the uncompensated sampling signal (U_T) delayed by i sampling periods, and A, B, C, D are system constants of a signal sampling path.

13. An arrangement as defined in claim 12, and further comprising means for turning off the phase compensation in the event that a change in the input signal within one sampling period exceeds a threshold value (anwdUBTmax).