JPH02312409A - Digital/analog converter - Google Patents

Abstract

PURPOSE:To decrease the zero crossing distortion in the auditory sense by adding and subtracting a bias to and from input data and varying a variation point of an MSB from '0'. CONSTITUTION:An analog output being free from a distortion is obtained through a low-pass filter from a synthesized wave for generating repeatedly a zero crossing distortion against a digital input being below bias data determined in advance. That is, in the case data obtained by adding and subtracting the bias data is brought to D/A conversion 12, successively and repeatedly, when digital data inputted previously is below the bias data, the data generates an inversion of an MSB at a zero crossing point, a distortion based on a D/A conversion error of the MSB is contained in a switching waveform of the bias data of high repeated frequency, and accordingly, it can be eliminated easily by the low-pass filter 15, and does not remain as a distortion of an audio band. In such a way, the zero crossing distortion is generated by a high frequency in the vicinity of a zero level of the input, and it becomes scarcely audible in the auditory sense through the low-pass filter and can be attenuated.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to an improvement in a digital/analog converter (hereinafter referred to as a D/A converter), and relates to an A/A converter that is suitable for D/A conversion of audio data. [Background Art and Problems to be Solved by the Invention] Conventionally, D/A converters have been used when converting digital data such as audio signals to analog. It’s like this/
In an A converter, the MSB error has the most adverse effect on the error during D/A conversion. That is, since the MSB corresponds to the polarity bit of the output analog signal, it causes zero-crossing distortion (shown in FIG. 4) at the zero-crossing point, which has a large effect on the auditory sense, causing a problem, especially when the output level is low. The purpose of the present invention is to eliminate such zero-cross distortion, which occurs when polarity is reversed, from the auditory sense.

[Means to solve the problem]

The present invention provides an addition/subtraction means that repeatedly adds and subtracts predetermined bias data to input digital data to generate digital data, a digital-to-analog converter that sequentially converts the addition data and subtraction data into D/A, and an analog signal. An analog output corresponding to the digital data is obtained from a composite wave of the converted analog data and the switching waveform of the bias data via a low-pass filter. Further, a low-pass filter is provided after the output of the D/A converter passes through a cancellation circuit that cancels the switching waveform of bias data, and an analog output corresponding to the input digital data is obtained.

[Effect]

According to the present invention, when data obtained by sequentially adding and subtracting bias data is sequentially and repeatedly D/A converted using the above-described configuration, when the original input digital data is less than the bias data, the MSB of the data is inverted at the zero crossing point, that is, the MSB
The distortion due to the D/A conversion error is included in the switching waveform of the bias data with a high repetition frequency, and therefore can be easily removed by a low-pass filter and does not remain as distortion in the audio band. Therefore, zero-cross distortion occurs at a high frequency when the input is near zero level, and can be attenuated through a low-pass filter so that it is hardly audible.

〔Example〕

An embodiment in which the present invention is applied to a D/A converter of a CD player will be described.

FIG. 1 is a block diagram of the present invention, FIG. 2 is a diagram showing input/output, and FIG. 3 is a diagram for explaining signal waveforms of each part. A 16-bit digital signal from an optical pickup passes through a signal processing system and a digital filter, and then is input to a serial-to-parallel converter l as a 16-bit + α-bit serial data input, and is converted into parallel data A.
Convert to Preset 5 is used to detect whether the result of addition or subtraction exceeds a predetermined level in the subsequent stage.
An overflow detector 2 and an underflow detector 3 are provided to compare the data preset by preset 7 with the data preset by preset 7 through the OR circuit 4 when an overflow or underflow is detected. an adder 8 for adding or subtracting 7-bit data B set to 7 bits, for example;
The prohibition of calculation is transmitted to the control C of the subtracter 9, and the adder 8 or subtracter 9 outputs the data A as it is without adding or subtracting the preset value of the preset 7. In this case, the output E1 of the D/A converter 12 is synchronized with the clock of the canceling means 16 via the adder 14, thereby eliminating the need for a clock. Set presets 5 and 6 to 7 bits.
The adder 8 and the subtracter 9 always add and subtract the 7-bit digital value of the preset value 7 sequentially within a predetermined level at which no overflow occurs even when the digital value of the preset 7 is added. Adder 8. Output data of subtractor 9 DI, D
2 are respectively converted from parallel to serial by parallel/serial converters 10 and 11 (not necessary if the D/A converter is a parallel input type). Switching data D is input sequentially at a clock frequency of 352.8 KHz, which is twice the 4-times oversampling of 44.1 KHz.The DA converter 12 inputs data D3 biased by a preset to the D/A converter 12.
From the converted analog data, the precent value is 352.
In order to cancel the waveform repeated at 8KH2, data E2 is added to the adder 14 in reverse phase, and the zero-cross distortion at the time of a small signal and the distortion signal remaining after cancellation are removed by the low-pass filter 15. The advantages of speeding up D/A conversion and the sampling frequency of oversampling being higher than the required analog frequency and not requiring a high-order audio filter circuit are great. Therefore, it is possible to provide a digital-to-analog converter in which zero-cross distortion does not occur when the input signal is below the bias value. Furthermore, if the input is above zero cross, the zero cross distortion will be masked in the audio signal and will not be particularly noticeable.

〔Effect of the invention〕

According to the present invention, a bias is added to input data.

By subtracting, the MSB changing point can be changed from 0, and since the MSB always changes at a low level below the bias value, the zero-crossing distortion caused by the MSB inversion can be repeatedly generated at high frequencies and removed by a filter. It's easy. Further, by providing a canceling means, it is possible to reduce the power applied to the low-pass filter.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a diagram for explaining input-output, and FIG. 3 is a diagram for explaining waveforms. ) is the input (E, ) of the D/A converter, the D/A converter output (E2) is the canceling circuit output (F), and the audio output is shown in FIG. 4. FIG. 4 is a waveform diagram showing zero-cross distortion in a conventional example. ■...Serial parallel converter 2...Overflow detector 3...Underflow detector 4...OR circuit 5.6.7...Preset circuit 8...Adder 9...Subtraction Device 10.11... Parallel/serial converter 12...
D/A converter 13...Switch circuit 14...Adder 15...Low pass filter 16...Cancel circuit

Claims (2)

[Claims] (1) Addition and subtraction means that repeatedly adds and subtracts predetermined bias data to input digital data to create digital data, and digital-analog that sequentially converts the added digital data and subtracted digital data into digital-analog. Equipped with a converter and a low-pass filter that obtains analog data from a composite wave of analog data converted to an analog signal and a switching waveform of bias data, and repeatedly generates zero-cross distortion for digital inputs that are less than predetermined bias data. A digital/analog converter that features a distortion-free analog output from a synthesized wave through a low-pass filter. (2) Addition and subtraction means that repeatedly adds and subtracts predetermined bias data to input digital data to create digital data, and digital-analog that sequentially converts the added digital data and subtracted digital data into digital-analog. The converter includes a canceling means for canceling the switching waveform of the bias data from a composite wave of the switching waveform of the analog data and the bias data converted into an analog signal, and a low-pass filter, and the digital input is equal to or less than predetermined bias data. This digital/analog converter is characterized by obtaining a distortion-free analog output from a synthesized wave that repeatedly generates canceled noise and zero-cross distortion through a low-pass filter.