JPH0444849B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a digital-to-analog converter that converts a digital signal into an analog signal.

[Technical background of the invention and its problems]

The basic circuit of a digital-to-analog converter (hereinafter referred to as a D/A converter) that converts a digital signal such as a binary code into an analog signal corresponding to the digital signal is explained in the first part for the case where, for example, a 4-bit digital signal is converted to an analog signal. This will be explained with reference to the figures. In the figure, _S3 to _S0 are analog switches that are connected to the power source _E3 side if the state of each bit of the binary code is 1, and are connected to the ground side if the state is 0. _R3 ～ _R0
are resistors connected in series to each analog switch S ₃ to S ₀ , and if the resistance value of resistor R ₃ is R, the resistance values of resistors R ₂ , R ₁ , and R ₀ are 2R, respectively.
2 ² R, 2 ³ R. The arithmetic circuit 1 adds the outputs from each resistance value to obtain an output V ₀ as an analog signal.

In a normal D/A converter that is not intended for special nonlinear input/output characteristics, the input/output characteristics of an analog output signal corresponding to a 4-bit digital input signal are ideally linear as shown in Figure 2. must be.

However, in an actual D/A converter, the input/output characteristics are not perfectly linear due to the resistance value accuracy of the resistor ladder circuit that constitutes the resistors R ₃ to _{R 0} and variations in the on-resistance of the analog switch.
This results in non-linear input/output characteristics that are somewhat away from a straight line as shown in FIG. There are two types of nonlinearity characteristics, and in the regular catalogs/performance tables of D/A converter manufacturers, they are called integral (or total) nonlinearity, and
DIFFERENTIAL
NONLINEARITY). Integral nonlinearity refers to whether the actual output signal deviates from the ideal output signal within a certain range, regardless of the input signal.For example, integral linearity ±1/2 LSB (LEAST
SIGNIFICANT BIT (that is, the average analog output difference with respect to the minimum digital input difference) guarantees that the output signal falls within the range of straight lines a and b in FIG. On the other hand, differential nonlinearity refers to continuous values 0-1, 1-2,...n-(n+1)
Normally, the difference between the analog output signal value and the digital input signal must be 1 LSB, but this is a characteristic that shows how much the difference actually occurs.

If we consider the case where the X and Y input digital signals are converted into analog signals through such a D/A converter, and these are input to the X and Y deflection systems of a cathode ray tube for imaging, the differential straight line A D/A converter with poor performance has the disadvantage that, even for an input that should produce a uniform mesh, line-shaped artifacts run vertically and horizontally due to the regular proximity/separation of the meshes, resulting in an extremely unsightly appearance.

[Purpose of the invention]

The present invention has been made in view of the above circumstances, and even if a plurality of D/A converters with poor differential linearity are used, one digital signal is subjected to multiple D/A conversions, and one of the repetitions is It is an object of the present invention to provide a digital-to-analog converter in which the differential linearity is statistically improved and the effects of differential nonlinearity, such as the linear artifacts, can be reduced.

[Summary of the invention]

In order to achieve the above object, the present invention comprises: a random number forming means for generating a random number having the same number of bits as an input digital signal to be converted into analog; a digital subtraction means for subtracting the random number from the digital signal; a first digital-to-analog conversion means for converting the output of the digital subtraction means from digital to analog; a second digital-to-analog conversion means for directly converting the random number from digital to analog; the first digital-to-analog conversion means and the second analog conversion means; The apparatus is characterized in that it includes an analog addition means for adding the outputs of the output and the output means in an analog manner.

[Embodiments of the invention]

Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

FIG. 4 is a block diagram showing an embodiment of the D/A converter of the present invention. In the figure, an n-bit digital signal input from an input terminal 2 and to be converted into an analog signal is transferred to a first latch circuit 3. The pseudo-random number generator 4 generates a pseudo-random number with the same number of bits as the digital signal, for example n bits, and transfers it to the second latch circuit 5. Note that this pseudo-random number generator 4 can be realized by a well-known shift register, exclusive OR, OR circuit, etc. The strobe signal output from the strobe input terminal 6 is input to the clock terminals of the first and second latch circuits 3 and 5, and each time this strobe signal is input, the strobe signal is input to the first and second latch circuits 3 and 5. The latched digital signal and random number are transferred to digital subtraction means 7. This digital subtraction means 7 subtracts a random number from the digital signal, and as a result, the first D/A, which is the first digital to analog conversion means, has (n+1) bits.
Transfer to A converter 8. Similarly, the second latch circuit 5 is activated every time the strobe signal is input.
transfers the random number to the second D/A converter 9, which is the second digital-to-analog conversion means, having an n-bit number. The difference between the digital signal converted to analog by the first D/A converter 8 and the random number and the random number converted to analog by the second D/A converter 9 are added in an analog manner by analog adding means 10, Output as a signal. still,
The pseudo random number generator 4 and the second latch circuit 5 constitute a random number generating means 15.

The operation of the present invention constructed as described above will be explained in detail.

A digital signal from the input terminal 2 to be converted into an analog signal is latched by a first latch circuit 3, and its output is designated as I. A pseudorandom number having the same number of bits as the digital signal is also latched by the second latch circuit 5, and its output is designated as R. The strobe signal is a signal that controls whether or not to D/A convert the input digital signal, and each time this strobe signal is input to the first and second latch circuits 3 and 5, the first latch circuit 3 outputs I, and the second latch circuit 5 outputs R. Digital subtraction means 7 digitally subtracts random number R from digital signal I. At this time, the digital signal I and random number R are n
Since it is a bit, the value of (IR) is -(2 ⁿ -1)
The value is up to (2 ⁿ −1). This is (n+1)
Since it is expressed as a number of bits, the output of the digital subtraction means 7 is converted into analog by a first D/A converter 8 having a number of (n+1) bits. The output of this first D/A converter 8 is basically an output proportional to (I-R) except for nonlinear characteristics, and the output _O1 varies depending on the type of D/A converter, but usually , gain α ₁ and offset voltage β ₁ , it is expressed as O ₁ =α ₁ (IR)+β ₁ (1).

On the other hand, through the second latch circuit 5, the pseudo-random number R
The second D/A converter 9, which receives only the input voltage, converts R into analog, and basically outputs an output proportional to R, and its output O ₂ has a gain α ₂ and an offset voltage β ₂ , so that O ₂ = α ₂ It is expressed as R+β ₂ ...(2).

These outputs O ₁ and O ₂ are added by the analog addition means 10, and from the gain α and equations (1) and (2), the output O ₃ of the analog addition means 10 is O ₃ = (α/α ₁ ) O ₁ +(α/α ₂ )O ₂ −(α/α ₁ β ₁ +α/α ₂ β ₂ )=αI...(3).

From this equation (3), the digital signal to be analog converted is subtracted by the random number generated from the random number forming means 15, the output after analog conversion by the first D/A converter 8 is intentionally shifted, and the second By adding the random numbers of the same value converted into analog by the D/A converter 9, it is possible to obtain an analog signal that does not include the processed random numbers. Therefore, if multiple random numbers are processed for the same value digital signal and analog conversion is performed multiple times, even if the first and second D/A converters 8 and 9 are not necessarily ideal, If the average value of repeated analog conversions is taken, the value approaches the ideal point, and it becomes possible to statistically reduce the influence of differential nonlinearity.

It goes without saying that the present invention is not limited to the embodiments described above, and includes various modifications within the scope of the gist of the present invention.

For example, the configuration shown in FIG. 5 may be used. That is, in place of the random number forming means 15, the following random number forming means 15' is constructed. That is, the analog random number generator 11 is used instead of the pseudo random number generator 4.
Alternatively, use a triangular wave generator, etc. to generate analog random numbers, and send the random numbers to sample and hold device 1.
2, and each time the strobe signal is input to the sample-and-hold device 12 via the timing circuit 13, this analog random number is converted into a digital value by the A/D converter 14, which is the analog-to-digital converter. It is transferred to the digital subtraction means 7. Further, the analog random number output from the sample and hold device 12 is directly transferred to the analog addition means 10 and added to the output of the first D/A converter 8.
In this way, an analog circuit can also be used as a random number generating means.

〔Effect of the invention〕

As detailed above, according to the present invention, even if a plurality of D/A converters with poor differential linearity are used, one digital signal can be subjected to multiple D/A converters.
It is possible to provide a digital-to-analog converter in which the differential linearity is statistically improved through repetition of the conversion, thereby reducing the effects of differential nonlinearity, such as the linear artifacts.

[Brief explanation of drawings]

Figure 1 is a basic D/A converter circuit diagram.
FIG. 2 is an input/output characteristic diagram of an ideal D/A converter, FIG. 3 is an input/output characteristic diagram of a conventional D/A converter, and FIG. Block Diagram FIG. 5 is a block diagram showing another embodiment of the present invention. 2... Input terminal, 4... Pseudo random number generator, 7... Digital subtraction means, 8... First digital to analog conversion means (first D/A converter), 9... Second digital to analog conversion means (second D/A of
converter), 10... analog addition means, 11...
Analog random number generator, 12... Sample hold device, 14... Analog-digital converter (A/D converter), 15, 15'... Random number forming means.

Claims (1)

[Scope of Claims] 1. Random number forming means for generating a random number having the same number of bits as the input digital signal to be converted into analog, digital subtraction means for subtracting the random number from the digital signal, and this digital subtraction means. a first digital-to-analog conversion means for digital-to-analog conversion of the output of the random number; a second digital-to-analog conversion means for directly converting the random number from digital to analog; A digital-to-analog converter comprising an analog addition means for adding outputs in an analog manner. 2. Claim 1, wherein the random number generating means includes an analog random number generator that generates analog random numbers, and an analog-digital converter that samples and holds the random numbers and then converts them into analog-digital data. Digital-to-analog converter as described in section.