JPH0193211A - Signal conversion device - Google Patents

Abstract

PURPOSE:To eliminate the wave distortion of an output signal by using a coefficient to correct the group delay characteristic of an analog filter almost linearly as the filter coefficient of a digital filter. CONSTITUTION:The titled device is constituted in such a way that the group delay characteristics of the digital filters 4 and 6 are set reversely to the group delay characteristics of analog low-pass filters 2 and 8. In other words, a targeted group delay characteristic (e) can be obtained by changing the filter coefficients C0-CN-1 of the digital filter to F0-FN-1 and deforming an impulse response characteristic. Thereby, it is possible to flatten the group delay characteristics of an A/D conversion device consisting of an input terminal 1, the analog low-pass filter 2, an A/D converter 3, and the digital filter 4, etc., and a D/A conversion device consisting of the digital filter 6, a D/A converter 7, the analog low-pass filter 8, and an output terminal 9, etc., to almost linear characteristics.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to analog-to-digital conversion or digital-to-digital conversion.
The present invention relates to a signal conversion device that performs analog conversion, and is suitable for use in, for example, digital recording and reproducing devices.

[Summary of the invention]

The present invention includes an analog filter, an A/D converter or a D/A
In a signal conversion device that includes a converter and a digital filter, coefficients that approximately linearly correct the group delay characteristic of the analog filter are used as filter coefficients of the digital filter, so that high-speed signal conversion without group delay distortion can be achieved. The present invention provides a signal conversion device that performs fidelity conversion.

[Conventional technology]

In a recording and reproducing apparatus such as a digital audio tape recorder using the DAT method, as shown in FIG.
A D converter 3 converts it into a digital audio signal. In that case, the A/D converter 3 uses a frequency 2f which is twice the normal sampling frequency f.
.

I am trying to perform oversampling.

A sampling clock with a frequency of 2r is supplied to the A/D converter 3 in order to obtain the sampling rate of 2f.The digital signal obtained from the A/D converter 3 is then supplied to the digital filter 4. , where the normal frequency f
After the signal is converted to a sampling rate of , it is supplied to the digital recording/reproducing circuit 5 and recorded on the magnetic tape.

During reproduction, the digital signal reproduced from the magnetic tape is supplied from the recording/reproducing circuit 5 to the digital filter 6, sampled at a sampling rate of 5, and then supplied to the D/A converter 7. The signal is oversampled at a sampling rate of 2f and converted into an analog signal.

This analog signal passes through an analog low-pass filter 8 and is extracted from an output terminal 9 as an original signal SA.

In addition, in an actual digital recording/reproducing device, L, R
Two channel signals are recorded and reproduced.

As described above, by performing oversampling in the A/D converter 3 during recording and in the D/A converter 7 during playback, it is possible to easily design the analog low-pass filter 2.8. . For example, if the analog audio signal SA has a band from approximately 0 to f, /2, the digital filter 4.6 is used, and the input is sharply input around f, /2 as shown in filter characteristic a in Figure 6. A material with characteristics that attenuates the noise is used.

Therefore, the analog low-pass filter 2.8 may be any filter that passes the signal SA, and may have a gentle attenuation characteristic as shown by filter characteristic C in FIG. 7, for example. This facilitates the design of the analog low-chair filter 2.8, which is advantageous in terms of signal distortion, cost, etc.

Figure 8 shows F used as digital filter 4.6.
This conceptually shows an IR type digital filter, in which a digital signal S1 supplied to an input terminal 10 sequentially passes through delay elements 11 connected in series in N stages. Each delay element 1
The delay time of 1 is set to 1/2 f, which is one sampling period when double oversampling is performed.

The outputs of each delay element 11 are multiplied by predetermined filter coefficients 00 to CM-+ in a coefficient multiplier 12 and weighted, and then added in an adder 13 to produce an output signal S. It is closed and taken out. This output signal S0 has a predetermined filter characteristic a as shown in FIG. 6, for example, with respect to the input signal S+.

FIG. 9 shows an example of the impulse response characteristic of a digital filter.

This impulse response characteristic is shown in Figure 8 at input terminal 1.
Output signal S when an impulse is applied to 0. This shows the change over time. That is, after the impulse is sequentially delayed by each delay element 11, a predetermined weighting is applied to each delayed output by a coefficient unit 12, and the output is sequentially outputted by an adder 13.

The interval between each sampling point (output time of each delay element 11) in this impulse response characteristic is TS/4 (1/4
f, , Ts=1/f, ) is shown,
The length of the arrow indicating the impulse at each sampling point has a size corresponding to each coefficient 00 to CM-1 of the coefficient multiplier 12.

The group delay characteristics of the filter will be explained below.

Generally, an analog filter has a group delay characteristic in which the amount of delay changes depending on the frequency of an input signal. In the case of the low-pass filter having the filter characteristic C shown in FIG. 7 described above, it has a group delay characteristic as shown in FIG. 7 d.

On the other hand, the group delay characteristic of a digital filter is such that the amount of delay is constant regardless of the input frequency, as shown in the group delay characteristic of FIG.

[Problem that the invention seeks to solve]

In the digital recording and rewriting device shown in FIG. 5 described above,
Since the analog low-pass filter 2.8 used during A/D conversion and D/A conversion has a group delay characteristic d as shown in FIG. Waveform distortion occurred, impairing high-fidelity recording and playback.

[Means for solving problems]

In the present invention, in a signal conversion device comprising an analog filter, an A/D converter or a D/A converter, and a digital filter, the group delay characteristic of the analog filter is approximately expressed as the filter coefficient of the digital filter. Coefficients F0 to FN-1 are used for linear correction.

[For production]

The group delay characteristic of the entire signal conversion device becomes flat, and waveform distortion of the analog output signal or digital output signal can be removed when the input signal is A/D converted or D/A converted.

〔Example〕

An embodiment in which the present invention is applied to the digital recording/reproducing apparatus shown in FIG. 5 will be described below.

In this embodiment, the group delay characteristic of the digital filter 4.60 shown in FIG. 5 is made to be the inverse of the group delay characteristic d of the analog low-pass filter 2.8 shown in FIG. 7 described above. That is, in FIG. 2, the digital filter 4.
6.

As a result, an A/D converter including an input terminal l, an analog low-pass filter 2, an A/D converter 3, a digital filter 4, etc. in FIG. 5, a digital filter 6, a D/A converter 7, an analog Each group delay characteristic of the D/A converter including the low-pass filter 8, the output terminal 9, etc. can be made substantially linear and flat as shown by the characteristic f in FIG.

Incidentally, in reality, the group delay characteristic e of the digital filter 4.6 is the characteristic e shown by the thick dotted line in FIG.

As shown in FIG.

Next, an embodiment of a method for providing the digital filter 4.6 with the group delay characteristic 91 described above will be described with reference to FIG.

As mentioned above, conventional digital filters can achieve the linear group delay characteristic shown in Fig. 6 by setting the filter coefficients shown in Fig. 8 to 00 to CN-1 and creating the impulse response characteristic shown in Fig. 9. It has gained.

In this embodiment, the above filter coefficients 00 to CM-1
The target group delay characteristic e1 is obtained by changing the impulse response characteristic from F0 to FN-1 to modify the impulse response characteristic.

To this end, in FIG. 3, first, the impulse response coefficient K for correcting the group delay characteristic d of the analog low-pass filter is determined. - to 2. , -0 are determined by performing a predetermined calculation based on the characteristic d.

Next, add 0 to 2 (N-+1 and the above coefficients 00 to C) to the above coefficients.
The above coefficient F is obtained by performing an impulse synthesis calculation with G1. -FN-1 is obtained. This impulse synthesis calculation is
C0 to C) This is to calculate the convolution sum of I-1 and 0 to 2 (N-1), and the following calculation is performed.

FIG. 4 shows impulse response characteristics obtained as a result of experiments on an actual digital filter.

Figure B shows the case where the coefficient C8-CM-1 is used, and the 9th
Substantially identical to that shown.

This impulse response characteristic is bilaterally symmetrical. Figure A shows the case where the above-mentioned filter coefficients F0 to PH-1 are used, and the characteristics are asymmetrical.

FIG. 1 shows a concrete embodiment of the digital filter 4.6.

In FIG. 1, 10 is an input terminal corresponding to FIG. 8 to which a digital signal is supplied, 14 is a shift register that sequentially takes in the digital signal S1 from the input terminal 10, 15 is used to write the contents of the shift register 14, and RAM for reading, 16 a multiplier to which the output of the RAM 15 is supplied as data to one of its input terminals, 17 a ROM in which the above-mentioned filter coefficients F0 to F□I are stored in advance, 18 a multiplier output of the multiplier 16; Adder to add, 19 is adder 1
A part of the output of this accumulator 19 is fed back to the adder 16 and successively added to the multiplication output from the multiplier 16. 20 is a shift register to which the output of the accumulator 19 is supplied, and 21 is an output terminal taken out from the shift register 20.

Next, the operation of the circuit with the above configuration will be explained.

Digital signal S1 supplied from input terminal 10 is sequentially delayed by a predetermined amount in shift register 14 and RAM 15 and supplied to multiplier 16, where it is sequentially multiplied by corresponding filter coefficients F0 to FN-1 from ROM 17. The multiplication output of the multiplier 16 is then sequentially supplied to the adder 18 and added to the previous multiplication output, that is, the feedback input from the accumulator 19. When this addition output is repeated, the output of the accumulator 19 is shifted. The signal is supplied to a register 20, and the contents of this shift register 20 are taken out to an output terminal 21 as a desired digital signal S0.

〔Effect of the invention〕

According to the present invention, an analog filter, a digital filter, an A/D converter, or a D/A converter can be combined to
When configuring a D converter or a D/A converter, group delay characteristic distortion based on the group delay characteristic of an analog filter can be removed, and an output waveform with high fidelity can be obtained.

[Brief explanation of the drawing]

Fig. 1 is a professional diagram showing an embodiment of the present invention, Fig. 2 is a characteristic diagram for explaining correction of group delay characteristics, and Fig. 3 is a block diagram showing an embodiment of a method for correcting group delay characteristics. 4 is a diagram showing the vals response characteristic of the digital filter, FIG. 5 is a block diagram of a digital recording/reproducing apparatus to which the present invention can be applied, and FIG. 6 is a diagram showing the group delay characteristic and frequency characteristic of the digital filter. , FIG. 7 is a diagram showing the group delay characteristic and frequency characteristic of the analog filter, FIG. 8 is a block diagram conceptually showing the digital filter, and FIG. 9 is a diagram showing the impulse response characteristic of the digital filter. In addition, in the symbols used in the drawings, 2.8-------------Low pass filter 3----A/D converter 7-- −
・・・−・−・−−−−−−D/A converter 4.6−
・−−−1−−・−・Digital filter 12−・−・
−1−−−・−・・−・Coefficient unit 17−・−・−・−・−
------ROM.

Claims (1)

[Claims] A signal conversion device comprising an analog filter, an A/D converter or a D/A converter, and a digital filter, wherein the group delay characteristic of the analog filter is used as a filter coefficient of the digital filter. A signal conversion device characterized in that it uses a substantially linear correction coefficient.