JP3209397B2 - Floating A / D converter - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to a floating A
The present invention relates to a / D conversion device, which is applied to a gain difference detection and gain adjustment processing section between signals and suppresses switching distortion due to a gain difference at the time of signal switching.

[0002]

2. Description of the Related Art Conventionally, in a floating A / D converter, a gain difference between signals has been a signal distortion at the time of switching, and a gain accuracy in an analog amplifier is increased to suppress the gain difference distortion. The effect of distortion on image quality and sound quality has been reduced by securing the accuracy during correction and devising switching timing. However, in such a method, it is difficult to respond to a gain change due to a temperature change or the like, and the switching frequency becomes coarse, so that the switching that follows the signal level change efficiently cannot be performed, and the advantage of the floating method is fully utilized. It wasn't cut.

[0003]

SUMMARY OF THE INVENTION According to the present invention, when a plurality of gain signals are switched and output as in a floating A / D converter, the output gain of each signal is unified to the same gain, and the signal at the time of switching is changed. Suppresses distortion due to gain difference between the two, enables adjustment corresponding to gain deviation in the analog section caused by temperature change, etc., and is less affected by noise, etc., and detects gain differences between signals stably And
It is an object of the present invention to shorten the time required for the gain deviation at the time of starting the system by shortening the required detection time, and to prevent deterioration of a reproduced signal (image quality and sound quality) by enabling high-performance gain correction as needed.

[0004]

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and comprises the following means 1) to 5). 1) In a floating A / D converter having n A / D converters and adjusting the gain of an output signal with an amplification gain corresponding to the level of an input signal, Amplifying means having n different amplifying sections, A / D converting means for quantizing these amplified signals,
Gain data ratio calculating means for calculating a gain data ratio (D1 / D2) with respect to a reference gain by converting each of the data quantized by the / D conversion means into a gain data ratio; Gain correction coefficient detecting means for averaging the gain data ratio; and, based on each gain difference correction coefficient obtained by the gain correction coefficient detecting means, each data quantized by the A / D conversion means. A floating A / D converter comprising signal level correcting means for adjusting an output level, and signal switching means for selectively outputting a signal obtained by the signal level correcting means in accordance with an input level.

[0005] 2) In a floating A / D converter having n A / D converters, each of which adjusts the gain of an output signal with an amplification gain corresponding to the level of an input signal. , An A / D converter for quantizing the amplified signals, and each of the data converted to quantized data by the A / D converter. , Gain data ratio with reference gain (D1 /
D2), a gain correction coefficient detecting means for compressing the calculated amount of fluctuation of the predetermined number of gain data ratios, and gains obtained by the gain correction coefficient detecting means. Based on the difference correction coefficient,
Signal level correcting means for adjusting the output level of each data quantized by the A / D converting means, and signal switching for selectively outputting a signal obtained by the signal level correcting means in accordance with the input level Means for floating A / D conversion.

3) A floating A / D converter having n A / D converters for adjusting the gain of an output signal with an amplification gain corresponding to the level of an input signal. , An A / D converter for quantizing the amplified signals, and each of the data converted to quantized data by the A / D converter. , Gain data ratio with reference gain (D1 /
D2), a gain correction coefficient detecting means for filtering the calculated predetermined number of gain data ratios, and each gain difference correction coefficient obtained by the gain correction coefficient detecting means. Signal level correction means for adjusting the output level of each data quantized by the A / D conversion means based on
A floating A / D converter comprising signal switching means for selectively outputting a signal obtained by the signal level correcting means in accordance with an input level.

4) The floating A / D according to claim 1
A floating A / D converter, wherein the gain correction coefficient detecting means performs a filtering process after averaging a predetermined number of gain data ratios.

5) The floating A / D according to claim 2
In the converter, the gain correction coefficient detecting means may be configured to perform a filtering process after compressing a predetermined amount of variation of the gain data ratio.
Ting A / D converter.

[0009]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described. FIG. 1 is a schematic block diagram of a floating A / D converter according to the embodiment. In the figure, an input signal Si is divided into A1 × Si, A2 × by the gains A1, A2,.
.., An × Si. The signals amplified with these amplification gains are respectively quantized by an A / D (analog / digital) converter 1-2, and the data D1,.
D2,..., Dn.

For the quantized data, a ratio between output reference gain data (D1 / Dn) is calculated for each sample by a gain data ratio calculation circuit 1-3, and the calculation result is used as a coefficient Ct for each sample to obtain a gain. Correction coefficient detection circuit 1
-4.

In the gain correction coefficient detecting circuit 1-4, each data ratio Ct sent for each sample is averaged by an averaging process shown in each embodiment to be described later in order to suppress variation between samples. Processing such as restriction is performed, and it is obtained as a gain difference correction coefficient Ci.

The gain difference correction coefficient Ci stably obtained by the gain correction coefficient detection circuit 1-4 is adjusted to the output level of each signal in the next signal level correction circuit 1-5. Floating signal switch 1-6
Thus, signal data quantized at an appropriate resolution according to the input signal level is selectively switched.

In the apparatus according to the above-described embodiment of the present invention, the following points are particularly taken into consideration.
That is, the quantized data D1 obtained from the A / D converter
(n), D2 (n)) should be D1 (n) / D2 (n) = A1 / A2 if the data ratio is constant in an ideal state. However, in an actual device, the S /
A constant data ratio cannot be obtained due to a difference in N, a quantization error, or an arithmetic error. For this reason, it is not appropriate to perform signal gain correction only on the data ratio in a specific sampling, and the correction coefficient derived from one sampling result is not stable whether the value indicates a true gain ratio and is stable. It is difficult to perform the correction.

In view of this, the apparatus of the present invention takes into account such a point, and obtains an average data ratio from the ratio of a plurality of N pieces of sampling data, or obtains a data ratio with a small amount of variation by using a correction coefficient. To obtain the accuracy of In the above-described configuration, for example, the above-described gain correction coefficient detection circuit 1-4 includes a DSP (Digital Signal Process).
or), but in the case of realization with this DSP, it is desirable to change the method of the detection processing according to the processing content to be executed. That is, the processing capability differs depending on the type of the DSP, and there are a number of executable instructions within 1 / fs (fs: sampling frequency) and an arithmetic process of which the user is not good. The detection method by the device of the present invention is particularly effective for a DSP in which a load on the division processing is not so large.

In order to obtain a high-accuracy average data ratio, it is necessary to increase the sampling number N.
This may be difficult to achieve depending on the capabilities of S. In such a case, the data ratio obtained for each sample is passed through a low-pass filter (LPF) to suppress a variation in the calculation result for each data, thereby obtaining a relatively stable gain difference correction coefficient. Can be.

[0016]

Next, preferred embodiments of the present invention will be described with reference to preferred embodiments. FIG. 2 is a detailed block diagram of the gain correction coefficient detection circuit 1-4.
For simplicity, the description will be made on the assumption that the data supplied from the A / D converter 1-2 is D1 and D2. In the figure, data D1 and D2 quantized by an A / D converter 1-2 are output to a reference data gain ratio (D1 / D2) for each sample in a gain data ratio calculation circuit 1-3. ), And the data ratio is supplied to the correction coefficient detection circuit 1-4 as a coefficient Ct for each sample.

In the gain correction coefficient detection circuit 1-4, M coefficients C obtained from m to n samples (nm = M) at any time are used.
The average of t is calculated, and the gain difference correction coefficient Ci is detected. Here, the memory 1-4a in this circuit is for holding the data ratio Ct calculated by the data ratio calculation circuit 1-3 for M samples, and the latest data ratio Ct is obtained. Then, the oldest data ratio Ct is discarded. The M data ratios Ct held in the memory 1-4a are added up by the adder 1-4b, and then the multiplier 1-4c
Is multiplied by 1 / M and the data ratio ratio Ct for M data
Is obtained. That is, a stable gain difference correction coefficient Ci is obtained by this averaging process.

The signal level correction circuit 1-5 calculates each signal level using the gain difference correction coefficient Ci obtained by the stabilization processing (output level is D1 level, Ct =
If D1 / D2, the D2 signal is adjusted to the output reference level by the operation of D2.times.Ci. Each signal unified to the output reference level has a configuration in which signal data quantized with an appropriate resolution according to the input signal level is selectively switched by the floating switching circuit 1-6.

Next, similarly, the second embodiment will be described with reference to FIG. The ratio of the quantized signal data to the output reference gain data (D1 / D2) is calculated for each sample by the data ratio calculation circuit 1-3, and the ratio between these data is calculated for each sample. It is supplied to the gain correction coefficient detection circuit 2-4 as the coefficient Ct. In the gain correction coefficient detection circuit 2-4, the result of the data ratio Ct obtained by the gain data ratio calculation circuit 1-3 is held in the hold memory 2-4d by the processing of the subtracter 2-4a. The difference ΔCn from the correction coefficient Cn-1 of the previous sample is calculated, and the result is calculated by the multiplier 2-4.
b, multiply ΔCn by 1 / m (1> m), and subtracter 2-4c
, The ΔCn / m is added to or subtracted from Cn−1 to detect a correction coefficient Ci (= Cn) as a gain difference correction coefficient. That is, in the gain correction coefficient detection circuit 2-4, a stable coefficient is obtained by performing a compression process on the amount of change in the gain ratio.

According to this embodiment, in the configuration of the above-mentioned first embodiment, the amount of memory for holding the arithmetic processing result obtained from the previous sample is M if the average of M samples is required. However, in the present embodiment, a memory capacity that holds only the operation result obtained one sample before is sufficient. Further, according to this embodiment, the change of the calculation result in each sample can be reflected for each sample. That is, when the value of the obtained data ratio (D1 / D2) is larger than Ci obtained by the operation result of the previous sample, the operation of Ci obtained by this sample increases, and when it is smaller, the operation of decreasing Ci is realized. This is an effective configuration for a system with an unstable gain setting.

Next, a third embodiment will be described with reference to FIG. In the gain correction coefficient detection circuit 3-4 in FIG.
Data ratio C obtained by gain data ratio calculation circuit 1-3
The result of t is subjected to band limiting processing by a low-pass filter (LPF) 3-4a to suppress variation in the data ratio Ct,
The gain difference correction coefficient Ci is detected. By this processing, the gain difference correction coefficient Ci is obtained as a stable value. According to the configuration of this embodiment, it is possible to suppress the influence of the variation of Ct with respect to the calculated Ci, particularly the effect of spike-like variation (error generated for each sample due to an arithmetic error or the like).

FIG. 5 shows a fourth embodiment. This embodiment is a combination of the configurations shown in FIGS. That is, the gain data ratio calculation circuit 1
The data ratio Ct obtained in step -3 is stored in the memory 4-4a and the adder 4-4 in the gain correction coefficient detecting circuit 4-4.
b, an averaging process is performed by the multiplier 4-4c, and the coefficient Ca for each sample obtained by the averaging is band-limited by a low-pass filter (LPF) 4-4d, so that more stable correction is performed. The configuration is such that the coefficient Ci can be obtained.

FIG. 6 shows a fifth embodiment. The device of this embodiment is a combination of the components shown in FIGS. That is, the gain data ratio calculation circuit 1
The data ratio Ct obtained in step S-3 is subtracted from the subtracter 5-4a and the multiplier 5-4 in the gain correction coefficient detection circuit 5-4.
b, a subtractor 2-4c, and a stabilizing process through a hold memory 5-4d, and then a low-pass filter (LP
F) The band is limited by 5-4e to obtain the gain difference correction coefficient Ci.

As described above, the following effects can be verified by verifying the effects of the devices of the above-described embodiments with reference to the waveform diagrams. That is, as shown in FIG.
The signals amplified with different gains are quantized (D1, D2
) Also has a level difference, and in an ideal state, D1 / D2 = A1 / A2 = constant. However, actually, as shown in FIG. 7B, D2 / D1 is not constant, and the division result for each sample varies and becomes an unstable value. In particular, according to the configuration shown in FIGS. 2 and 4 (first and third embodiments), as shown in FIG. 7D, the variation of D1 / D2 is reduced, and at the same time, the variation of Ci is reduced gradually. Can be obtained.

Further, according to the configuration of FIG. 3 (second embodiment), as shown in FIG. 7C, the variation of D1 / D2 is reduced and stabilized, so that it is always close to A1 / A2. A value correction coefficient is obtained.

Further, according to the configuration of FIGS. 5 and 6 (the fourth and fifth embodiments), the value with increased stability is further band-limited, as shown in FIG. The gain difference correction coefficient Ci having the above stability can be obtained.

[0027]

According to the floating A / D converter of the present invention, when switching and outputting signals of a plurality of gains, the output gain of each signal is unified to the same gain, and the gain difference between the signals at the time of switching is obtained. To suppress distortion caused by noise, to enable adjustment corresponding to the gain shift in the analog section caused by temperature change, etc., to detect the gain difference between signals without being affected by noise, etc. By shortening the time, the time during which the gain shift occurs when the system is started is shortened, and at the same time, it is possible to perform high-performance gain correction at any time, so that the reproduced signal (image quality and sound quality) can be prevented from deteriorating. .

[Brief description of the drawings]

FIG. 1 shows a floating A / according to an embodiment of the present invention.
It is a schematic block diagram of a D conversion device.

FIG. 2 shows a floating A / according to a first embodiment of the present invention;
It is a schematic block diagram of a D conversion device.

FIG. 3 illustrates a floating A / according to a second embodiment of the present invention;
It is a schematic block diagram of a D conversion device.

FIG. 4 shows a floating A / according to a third embodiment of the present invention;
It is a schematic block diagram of a D conversion device.

FIG. 5 shows a floating A / according to a fourth embodiment of the present invention;
It is a schematic block diagram of a D conversion device.

FIG. 6 shows a floating A / according to a fifth embodiment of the present invention.
It is a schematic block diagram of a D conversion device.

FIG. 7 is a waveform chart for explaining the effect of each embodiment.

[Explanation of symbols]

 1-1 Amplifier 1-2 A / D (analog / digital) converter 1-3 Gain data ratio calculation circuit 1-4 Gain correction coefficient detection circuit 1-5 Signal level correction circuit 1-6 Floating signal switch

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-2-27813 (JP, A) JP-A-6-97824 (JP, A) JP-A-5-29939 (JP, A) JP-A-3-297 32227 (JP, A) JP-A-57-203322 (JP, A) JP-A-2-27813 (JP, A) JP-A-57-46330 (JP, U) (58) Fields investigated (Int. Cl. ⁷ , DB name) H03M 1/00-1/88

Claims (5)

(57) [Claims] 1. A floating A / D converter having n A / D conversion means for adjusting the gain of an output signal with an amplification gain corresponding to the level of an input signal, wherein: Amplifying means having n different amplifying sections, A / D converting means for quantizing the amplified signals, and each data which has been quantized by the A / D converting means. , Gain data ratio with reference gain (D1 / D2)
A gain data ratio calculating means for calculating the gain data ratio, a gain correction coefficient detecting means for averaging the calculated predetermined number of gain data ratios, and a gain difference correction coefficient obtained by the gain correction coefficient detecting means. A signal level correcting means for adjusting an output level of each data quantized by the A / D converting means; and a signal obtained by the signal level correcting means is selectively output according to an input level. A floating A / D converter comprising signal switching means. 2. A floating A / D converter having n A / D conversion means for adjusting the gain of an output signal with an amplification gain corresponding to the level of an input signal. Amplifying means having n different amplifying sections, A / D converting means for quantizing the amplified signals, and each data which has been quantized by the A / D converting means. , Gain data ratio with reference gain (D1 / D2)
Gain data ratio calculating means for calculating the gain data ratio, a gain correction coefficient detecting means for compressing the calculated amount of fluctuation of the predetermined number of gain data ratios, and each gain difference correction obtained by the gain correction coefficient detecting means. Signal level correction means for adjusting the output level of each data quantized by the A / D conversion means based on the coefficient; and a signal obtained by the signal level correction means according to the input level. A floating A / D converter comprising signal switching means for selectively outputting. 3. A floating A / D converter having n A / D conversion means for adjusting the gain of an output signal with an amplification gain corresponding to the level of an input signal. Amplifying means having n different amplifying sections, A / D converting means for quantizing the amplified signals, and each data which has been quantized by the A / D converting means. , Gain data ratio with reference gain (D1 / D2)
, A gain correction coefficient detecting means for filtering the calculated predetermined number of gain data ratios, and a gain difference correction coefficient obtained by the gain correction coefficient detecting means. A signal level correcting means for adjusting an output level of each data quantized by the A / D converting means; and a signal obtained by the signal level correcting means is selectively output according to an input level. A floating A / D converter comprising signal switching means. 4. The floating A / D converter according to claim 1, wherein the gain correction coefficient detecting means performs a filtering process after averaging a predetermined number of gain data ratios. Floating A characterized by
/ D converter. 5. The floating A / D converter according to claim 2, wherein the gain correction coefficient detecting means performs a filtering process after compressing a predetermined number of gain data ratio fluctuations. A floating A / D converter.