JP3337192B2 - Encoding device, decoding device, and recorded medium of encoded information - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to an encoding device,
Regarding a decoding device and a medium on which encoded information is recorded,
In particular, the present invention relates to an apparatus and a recorded medium used to record data on a carrier, that is, a recording medium by performing data compression, and to perform data expansion during reproduction.

[0002]

2. Description of the Related Art C as an optical disc for audio reproduction
More than ten years have passed since D (Compact Disc) came on the market, and as a recording medium for audio information, it has already shown remarkable spread over conventional cassette tapes. The physical / logical format of the CD, which is a digital disk, has been established as an EFM modulation recording method for 8-bit fixed data length symbols and a data format method for subcode, audio data, CRC, etc., and various application functions are added. CD players have been developed.

A CD is also used as a data CD-ROM by discriminating it by control bits (4 bits) of the Q channel in the subcode or by discriminating the absence of TOC. Its application is expanding in the field of electronic publishing by making effective use of the large capacity and high-speed accessibility of disks. By the way, in the above CD-ROM, the audio is ADP
It is compressed by CM, and the original sound quality cannot be reproduced by the compression, so that a recording with higher fidelity has been demanded. In other words, even a compressed CD
It is expected that a disc capable of recording an audio signal equivalent to twice the bandwidth of a disc will appear. In an example of the conventional technique, the frequency band is divided into 帯 域 bands, the high frequency band is down-sampled to a サ ン プ リ ン グ sampling frequency, and the data is compressed there (Pioneer's wide band CD: AES Tokyo Convention 1995 pp 130-132 “ CD Broadband Trial-Keeping Compatibility-").

[0004]

In the conventional method, signal processing such as downsampling accompanied by FFT operation is required, and there is a disadvantage that it is difficult to secure the operation accuracy. Therefore, an object of the present invention is to provide an encoding device and a corresponding decoding device that realize a method that does not require signal processing such as downsampling and that can easily ensure calculation accuracy.

[0005]

According to the present invention, in order to achieve the above object, the present invention compresses only the band component of twice or more without compressing the band component of one time and causing no error. It was created as a device capable of providing a recording method / encoding device and a reproducing method / decoding device capable of suppressing errors in high frequency components. Here, the following will describe the 1 ×, 2 ×, and 4 × band components. One is a band component of up to about 24 kHz which is sampled at 48 kHz, which is a sampling frequency of an audio signal in the DAT standard, which is generalized in digital audio technology, and a double band component. The component is twice as high, ie up to 48 kHz sampled using 96 kHz as the sampling frequency
A band component up to the order is referred to as a quadruple band component.
It means a band component up to about 96 kHz sampled using Hz. The other is that the double band component is a band component up to about 48 kHz sampled at 96 kHz, which is the sampling frequency of the audio signal in the DVD standard, and the double band component is twice that. That is, up to 96 kHz sampled using 192 kHz as the sampling frequency.
A band component up to about z, and a quadruple band component is further doubled, that is, a maximum of 192 kHz sampled using 384 kHz as a sampling frequency.
It refers to the band component up to about Hz. Note that these frequencies are examples, and the present invention is not limited to these frequencies. The present invention provides an encoding apparatus and a decoding apparatus that can compress only a band component that is twice or more without compressing a 1-fold band component and generate an error, and that can suppress an error of a high-frequency component generated by compression. Created for the purpose.

[0006]

That is , according to the present invention, A / D conversion means for sampling an analog signal at a sampling frequency sufficiently higher than twice the highest frequency thereof and performing analog-to-digital conversion to obtain sampling code information; A low-pass filter means for passing a band frequency component of approximately 1/2 of the above, a decimation means for thinning out the output of the low-pass filter means to half,
A first difference information detecting means for obtaining difference information between every other sampling of the code information analog-to-digital converted by the / D converting means and the output information of the decimation means; and a second difference information obtaining means for obtaining difference information for every other sample. (2) An encoding device is provided that includes a difference information detecting unit and a unit for transmitting output information of the decimation unit and the difference information.

Further, according to the present invention, an analog signal is
A / D conversion means for sampling at a sampling frequency sufficiently higher than twice the highest frequency and performing analog-to-digital conversion to obtain sampling code information, and a low-pass filter for passing a band frequency component substantially half of the sampling code information Means, decimation means for thinning out the output of said low-pass filter means by half, and said A / D
A difference information detecting unit that obtains difference information between information of every other sampling of the code information analog-to-digital converted by the conversion unit and output information of the decimation unit, a unit that scales the difference information for each predetermined frame, An encoding device is provided having output information of a decimation means and means for transmitting the scaled difference information.

Further, according to the present invention, an analog signal is
A / D conversion means for sampling at a sampling frequency sufficiently higher than twice the highest frequency and performing analog-to-digital conversion to obtain sampling code information, and a low-pass filter for passing a band frequency component substantially half of the sampling code information Means, decimation means for thinning out the output of said low-pass filter means by half, and said A / D
A first difference information detecting means for obtaining difference information between information of every other sampling of the code information analog-digital converted by the converting means and output information of the decimation means;
Second difference information detecting means for obtaining difference information for each sample, means for scaling the two difference information for each predetermined frame, and transmitting the output information of the decimation means and the scaled two pieces of difference information. Means are provided.

Further, according to the present invention, an analog signal is
A / D conversion means for sampling at a sampling frequency sufficiently higher than twice the highest frequency and performing analog-to-digital conversion to obtain sampling code information, and a low-pass filter for passing a band frequency component substantially half of the sampling code information Means, decimation means for thinning out the output of said low-pass filter means by half, and said A / D
First difference information detecting means for obtaining difference information between every other sample of code information analog-to-digital converted by the converting means and output information of the decimation means, and a predicted value for every other sample from the sampling code information , A second difference information detecting means for obtaining difference information between the predicted value and every other sample, a means for scaling the two difference information for each predetermined frame, and an output information of the decimation means. And means for transmitting the scaled difference information.

Further, according to the present invention, an analog signal is
A / D conversion means for sampling at a sufficiently high sampling frequency of at least four times the highest frequency to obtain analog-to-digital conversion to obtain sampling code information, and a low-pass for passing a band frequency component of approximately 1/4 of the sampling code information Filter means; decimation means for thinning out the output of the low-pass filter means to 1/4;
3 of the code information analog-digital converted by the D conversion means
First difference information detecting means for obtaining difference information between information at every sampling and output information of the decimation means; information at every third sampling of code information analog-to-digital converted by the A / D conversion means; Second difference information detecting means for obtaining difference information from other information or a function thereof every three samplings of the code information analog-digital converted by the D converting means, and transmitting output information of the decimation means and the two difference information Means for performing the encoding.

Further, according to the present invention, an analog signal is
A / D conversion means for sampling at a sufficiently high sampling frequency of at least four times the highest frequency to obtain analog-to-digital conversion to obtain sampling code information, and a low-pass for passing a band frequency component of approximately 1/4 of the sampling code information Filter means; decimation means for thinning out the output of the low-pass filter means to 1/4;
3 of the code information analog-digital converted by the D conversion means
Difference information detecting means for obtaining difference information between information at every sampling and output information of the decimation means, encoding means for encoding information other than the information at every three samplings in the sampling code information by ADPCM, There is provided an encoding device having means for transmitting output information of a decimation unit, the difference information, and an output signal of the encoding unit.

[0013]

Further, according to the present invention, the output of the low-pass filter means for passing approximately 1/2 the band frequency component of the sampling code information in the analog-to-digital conversion is thinned out to 1/2 by the decimation means, and the analog-to-digital conversion is performed. When information indicating the difference between the information of every other sampling of the code information and the output information of the decimation means is used as difference information, an addition means for adding the transmitted sampling code information and the difference information, and 1 Means for decoding information every other sample.

Further, according to the present invention, approximately 1/2 of the sampling code information in the analog-to-digital conversion at the sampling frequency of twice or more the highest frequency of the analog signal.
The output of the low-pass filter means for passing the band frequency component of the above is thinned out to 1/2 by the decimation means, and the difference information between every other sampling of the analog-to-digital converted code information and the output information of the decimation means is converted to the first information.
As difference information, the difference information for every other sample
As the difference information, the sampling code information and the first and second
A decoding device for responding to the difference information when the difference information is transmitted, the decoding device adding the transmitted first difference information to the sampling code information, and decoding half of the sampling code information; Decoding means for decoding the remaining half of the information using the second difference information, and means for performing digital-to-analog conversion at the sampling frequency at the time of the analog-to-digital conversion using the signal decoded by the means. Is provided.

Further, according to the present invention, approximately 1/2 of the sampling code information in the analog-to-digital conversion at the sampling frequency of twice or more the highest frequency of the analog signal.
The output of the low-pass filter means for passing the band frequency component of the signal is decimated to 1/2 by the decimation means, and the information indicating the difference between every other sampling of the analog-to-digital converted code information and the output information of the decimation means is subtracted. A decoding device responsive to the difference information when the difference information is scaled and transmitted together with the sampling code information, and controls a value of the scaled difference information using the transmitted scaling information. Means, decoding means for adding the controlled difference information to the sampling code information, and decoding half of the sampling code information, and decoding means for decoding all the sampling information from a plurality of decoded information. , Using the signal decoded by the decoding means, And means for digital-to-analog conversion pulling frequency, the decoding device comprising is provided.

Further, according to the present invention, approximately 1/2 of the sampling code information in the analog-to-digital conversion at the sampling frequency of twice or more the highest frequency of the analog signal.
The output of the low-pass filter means for passing the band frequency component of the above is thinned out to 1/2 by the decimation means, and the difference information between every other sampling of the analog-to-digital converted code information and the output information of the decimation means is converted to the first information.
As difference information, the difference information for every other sample
A decoding device responsive to the first and second difference information when the first and second difference information are scaled and transmitted together with the sampling code information, wherein the first and second difference information are scaled using the transmitted scaling information; Means for controlling the values of the first and second difference information, decoding means for adding the controlled first difference information to the sampling code information, and decoding half of the sampling code information;
Decoding means for decoding the remaining half of the information using the controlled second difference information, and means for performing digital-to-analog conversion at a sampling frequency at the time of the analog-to-digital conversion using the signal decoded by the means; An apparatus is provided.

Furthermore, according to the present invention, approximately one-fourth of the sampling code information in the analog-to-digital conversion at a sampling frequency four times or more the highest frequency of the analog signal.
The output of the low-pass filter means for passing the band frequency component of the above is thinned out to 1/4 by the decimation means, and the information indicating the difference between every three samplings of the analog-to-digital converted code information and the output information of the decimation means is output. The difference information between every three samplings of the analog-to-digital converted code information and other information or its function every three samplings of the analog-to-digital converted code information is defined as second difference information. A decoding device that responds to the transmission of the code information and the first and second difference information, wherein the decoding device adds the transmitted first difference information to the sampling code information; / 4, and decodes all the sampling code information from the decoded information and the second difference information. Means for, and means for digital-to-analog conversion at a sampling frequency during the analog-to-digital conversion by using the decoded signal by said means, the decoding device comprising is provided.

Furthermore, according to the present invention, approximately one-fourth of the sampling code information in the analog-to-digital conversion at a sampling frequency four times or more the highest frequency of the analog signal.
The output of the low-pass filter means for passing the band frequency component of the signal is decimated to 1/4 by the decimation means, and the information indicating the difference between every three samplings of the analog-to-digital converted code information and the output information of the decimation means is subtracted. Information obtained by encoding the sampling code information, the difference information, and the sampling code information other than the information at every third sampling by ADPCM.
A decoding device responsive to a DPCM signal transmitted when the DPCM signal is transmitted, the difference information being added to the sampling code information, and the three-sampling information corresponding to a quarter of the sampling code information. Means for decoding information, and an ADP for decoding, using the ADPCM signal and an output signal of the decoding means, the sampling code information other than the information at every third sampling.
A CM decoder, an output signal of the decoding means, and A
A decoding means for switching and outputting a plurality of output signals of the DPCM decoder to decode the sampling code information; and a D / A conversion means for digital-to-analog conversion of the sampling code information at a sampling frequency at the time of the analog-to-digital conversion. An apparatus is provided.

[0020]

Further according to the invention, the analog signal is
With respect to the sampling code information sampled at a sampling frequency sufficiently higher than twice the highest frequency and subjected to analog-to-digital conversion by the A / D converter, a band frequency component of approximately 1/2 of the sampling code information is subjected to a low-pass filter. Means, the output of the low-pass filter means is decimated to 1/2 by the decimation means, and the information of the code information which has been analog-digital converted by the A / D conversion means at every other sampling and the output information of the decimation means The first difference information detecting means obtains the difference information from the first and second samples, and the second difference information detecting means obtains the difference information for every other sample, and transmits the output information of the decimation means and the difference information by the transmitting means. Recorded medium of encoded information in which a signal obtained by encoding the analog signal is recorded. It is provided.

Further according to the invention, the analog signal is
With respect to the sampling code information sampled at a sampling frequency sufficiently higher than twice the highest frequency and subjected to analog-to-digital conversion by the A / D converter, a band frequency component of approximately 1/2 of the sampling code information is subjected to a low-pass filter. Means, the output of the low-pass filter means is decimated to 1/2 by the decimation means, and the information of the code information which has been analog-digital converted by the A / D conversion means at every other sampling and the output information of the decimation means The difference information is obtained by difference information detection means, the difference information is scaled for each predetermined frame, and the analog signal is encoded using the output information of the decimation means and the scaled difference information. The recorded medium of the encoded information is provided.

Further according to the invention, the analog signal is
With respect to the sampling code information sampled at a sampling frequency sufficiently higher than twice the highest frequency and subjected to analog-to-digital conversion by the A / D converter, a band frequency component of approximately 1/2 of the sampling code information is subjected to a low-pass filter. Means, the output of the low-pass filter means is decimated to 1/2 by the decimation means, and the information of the code information which has been analog-digital converted by the A / D conversion means at every other sampling and the output information of the decimation means The first difference information detecting means obtains the difference information from the other, the difference information for every other sample is obtained by the second difference information detecting means, and the two difference information are scaled for each predetermined frame, and the decimation is performed. Encoding the analog signal using output information of the means and the scaled difference information Recorded medium recorded encoded information signal is provided.

Further according to the invention, the analog signal is
With respect to the sampling code information sampled at a sampling frequency sufficiently higher than twice the highest frequency and subjected to analog-to-digital conversion by the A / D converter, a band frequency component of approximately 1/2 of the sampling code information is subjected to a low-pass filter. Means, the output of the low-pass filter means is decimated to 1/2 by the decimation means, and the information of the code information which has been analog-digital converted by the A / D conversion means at every other sampling and the output information of the decimation means Is obtained by the first difference information detecting means, and a prediction value for every other sample is obtained from the sampling code information, and the difference information between the prediction value and the other every other sample is obtained as second difference information. Calculated by the detection means,
A recorded medium of encoded information in which a signal obtained by encoding the analog signal by using the output information of the decimation means and the scaled difference information, scaling the two difference information for each predetermined frame is provided. You.

Furthermore, according to the present invention, the analog signal is
For sampling code information sampled at a sufficiently high sampling frequency of four times or more the highest frequency and analog-to-digital converted by the A / D converter, a band frequency component of approximately 1/4 of the sampling code information is low-passed. The signal is passed by a filter means, the output of the low-pass filter means is thinned out to 1/4 by a decimation means, the information of the code information which is analog-digital converted by the A / D conversion means at every third sampling, and the output of the decimation means The difference information from the information is obtained by the first difference information detecting means, the information of every three samplings of the code information analog-to-digital converted by the A / D converting means, and the analog-to-digital converted information by the A / D converting means. The difference information from the other information or its function at every third sampling of the code information
A recorded medium of encoded information obtained by recording the signal obtained by encoding the analog signal using the output information of the decimation unit and the two difference information obtained by the difference information detecting unit is provided.

Further according to the invention, the analog signal is
For sampling code information sampled at a sufficiently high sampling frequency of four times or more the highest frequency and analog-to-digital converted by the A / D converter, a band frequency component of approximately 1/4 of the sampling code information is low-passed. Passing through the filter means, thinning out the output of the low-pass filter means to 1/4 by the decimation means,
Difference information between every three samplings of code information analog-to-digital converted by the A / D conversion means and output information of the decimation means is obtained by a difference information detection means. Other than the information described above, the encoding means encodes the signal by ADPCM and encodes the analog signal using the output information of the decimation means, the difference information, and the output signal of the encoding means. Is provided.

In the present invention, the reproduced data in the 1-time sampling is used without being subjected to any arithmetic processing, is used as it is, is advantageous in compatibility without the necessity of the arithmetic operation, and there is no deterioration accompanying the arithmetic operation. In addition, in reproduction at twice or more sampling, interpolation data is easily generated from accurate twice or more sampling data to decode and reproduce compressed audio data, so that complicated calculations can be avoided while satisfactorily compressing. 2
It can be suitably applied to a high bit signal such as 4 bits.

[0028]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, an encoding device according to the present invention,
An embodiment of a decoding device and a medium on which encoded information is recorded will be described in detail with reference to the drawings with reference to preferred embodiments. <First Embodiment (FIGS. 1 and 2): Encoding Device> FIG.
1 is a block diagram illustrating an encoding device according to a first embodiment of the present invention. In the block diagram of FIG. 1, the description will be made assuming that an input signal coming into the input terminal IN is a signal having a waveform as shown in FIG. 3 and is sampled at a time interval of Δt (for example, high-sampled at 96 kHz). The input audio signal is converted into a PCM signal by the A / D converter 1 at a sufficiently high frequency, for example, a high sampling frequency of 96 kHz, and sampling data # b1,..., # B2,.
Get 3. These sampling data are converted at a high resolution of, for example, 24 bits. Next, these signals are compressed by the signal processing circuit 2 using the memory 3, packed by the DVD encoding circuit 4, and
The signal is directly output from the UT 1 or, if necessary, modulated by the modulation circuit 5 and then recorded on a recording medium (not shown).

More specifically, in the signal processing circuit 2, as shown in FIG. 2, first, on a curve β in FIG. 3 which is band-limited by a low-pass filter 6 for passing a half band, for example, an FIR filter. Хc1, ..., хc2,
..., хc3, ... (curve α is the signal before band limitation)
Next, the signal хc which is decimated in half by the decimating circuit 7 as shown in the figure.
1, хc2, хc3, and хb1, thinned out by the thinning circuit 8
計算 b2, хb3 and difference calculator 9 comprising an adder (subtractor)
Then, хbi−хci = Δi is calculated (i is a natural number indicating a sample number). Here, the difference Δi is represented by, for example, 12 bits or less.
This is a fixed or variable bit.

The sampling data サ ン プ リ ン グ ci and the difference data Δi are input to the allocation circuit 10, and the allocation circuit 10 performs packing processing in a predetermined arrangement and outputs the result. Here, when the user data (packet) is 2034 bytes like a DVD, 456 pieces of sampling data хci and 456 pieces of difference data Δi are stored. The subheader is 2 bytes.

<Second Embodiment (FIGS. 6 and 7): Decoding Device> FIG. 6 is a block diagram showing a decoding device as a second embodiment of the present invention. This will be described assuming that the input signal (encoded data) coming into the left input terminal in the block diagram of FIG. 6 is decoded into a signal having a waveform as shown in FIG. 6, an input signal is demodulated by a demodulation circuit 11, is unpacked and decoded by a DVD decoding circuit 12, generates interpolation data by a signal processing circuit 13 and a memory 14, and is converted into an analog signal by a D / A converter 15. Is output.

More specifically, the signal processing circuit 13 is configured as shown in FIG. 7, and uses the input signals хci and Δi supplied to the input terminals of the signal processing circuit 13 to adder 16
Then, Δi + хci = хbi is calculated. Here, the output хbi is represented by the same 24 bits as the original digital data (хbi in FIG. 3). The interpolation processing circuit 17 uses a plurality of the output data хbi, and outputs inter-sampling data хa1, хa2,.
i, ... are calculated. For example, in the up-sampling method, the value хai can be obtained by a low-pass filter by filling each data with 0 data. Alternatively, a curve approximation or a prediction approximation can be used.

Then, the data is expressed as хb1, хa1,.
The signals are output in the order of i, хai,... and supplied to the D / A converter 15 to be converted into an analog signal at 96 kHz. In FIG. 6, the 1-times data хci is directly supplied from the DVD decoding circuit 12 as an unpacking decoder to the D / A converter 15 without passing through the signal processing circuit 13, and is converted with 24-bit accuracy at, for example, 48 kHz. You can do it.

Here, considering the processing in the signal processing circuit 13 of FIG. 6, one of them is only the operation of the addition (16), the original data хbi is reproduced without deterioration, and the other processing is the interpolation processing. (17) only, for example, can be dealt with by one filter operation of finite coefficients (acyclic) (as is well known, the QMF filter includes the FFT operation as in the related art,
In the FFT operation, since multiplication is used very often, the accuracy is deteriorated and the accuracy of 24 bits is difficult.) Therefore, since the approximation can be satisfactorily performed without requiring the processing accuracy, the processing accuracy can be remarkably improved. It is possible to easily cope with 24 bits of high bits, and the degree of output deterioration falls within several bits in a high frequency range.

Further, a recording medium (carrier) can be manufactured according to the above embodiment by manufacturing a master from the signal of the output terminal OUT2 in FIG. The recording medium can be an optical disk such as a DVD. Note that the interpolation processing circuit 17 shown in FIG. 7 uses a plurality of the above хbi, and uses inter-sampling data хa1, хa2,..., Хai,.
Has been described, for example, in the up-sampling method, the hold data (that is, хb1 between хb1 and хb2) is buried in each without filling in 0, and the signal x1 in FIG. x2, x3, ...
・ You can ask for That is, intermediate interpolation values xa1, xa2, xa3,.
It is not limited to what seeks.

<Third Embodiment (FIGS. 1 and 4): Encoding Apparatus> The encoding apparatus according to the third embodiment has the same overall block diagram as shown in FIG. Only the middle signal processing circuit 2 differs as shown in FIG. Therefore, the third embodiment will be described focusing on the differences from the first embodiment. In the signal processing circuit 2, FIG.
As shown in FIG. 5, first, signals хc1,..., Хc2,..., 上 の c on the curve β of FIG. 5 band-limited by a low-pass filter 6 that passes a half band, for example, an FIR filter.
, And then the signals хc1, хc2, хc3,..., Which are decimated in half by the decimating circuit 7, and the signals хb1, хb2, хb3 is calculated by the difference calculator 9, and the signals хb1, хb2, хb3 ·
.. Every other signal хa1, хa2,
The difference between хa3... and the signals хc1, хc2, 計算 c3,... is similarly calculated by the difference calculator 9, and -Obtain the second difference. Here, the first and second differences Δ1i and Δ2i are represented by, for example, 12 bits or less. This is a fixed or variable bit.

The allocation circuit 10 stores the sampling data хci and the first and second difference data Δ1i, Δ1
2i is input, the allocation circuit 10 performs packing processing in a predetermined arrangement, and outputs the result. Here, when the user data (packet) is 2034 bytes like a DVD, 406 pieces of the sampling data Δci and 406 pieces of the first and second difference data Δ1i and Δ2i are stored. The subheader is 4 bytes.

<Fourth Embodiment (FIGS. 6 and 9): Decoding Apparatus> The decoding apparatus according to the fourth embodiment
As in the embodiment, the configuration is as shown in FIG. 6. In the block diagram of FIG. 6, an input signal (encoded data) input to the left input terminal is decoded into a signal having a waveform as shown in FIG. It will be described as being performed. 6, an input signal is demodulated by a demodulation circuit 11, is unpacked and decoded by a DVD decoding circuit 12, generates interpolation data by a signal processing circuit 13 and a memory 14, and is converted into an analog signal by a D / A converter 15. Is output.

The signal processing circuit 13 in the fourth embodiment is
As shown in FIG. 9, the adder 16 calculates Δ1i + хci = хbi using the input signals хci and Δ1i supplied to the left input terminal. Further, using the input signals Δci and Δ2i similarly supplied, the adder 18 calculates Δ2i + хci = хai. Here, the outputs хbi and хai are represented by the same 24 bits as the original digital data (хbi and хai in FIG. 5).

Then, the data is expressed as хb1, хa1,.
The signals are output in the order of i, хai,... and supplied to the D / A converter 15 to be converted into an analog signal at 96 kHz. In FIG. 6, one-time data хci is directly supplied from the DVD decoding circuit 12 which is an unpacking decoder to the D / A converter 15 without passing through the signal processing circuit 13 so that the data хci is converted at, for example, 48 kHz with 24-bit accuracy. Can be. Here, the signal processing circuit 1 shown in FIG.
Considering the processing in No. 3, only the operation of addition (16) is performed, and the original data хbi and хai are reproduced without deterioration (if
When the QMF filter is applied as in the conventional case, the FFT operation is included. In the FFT operation, the multiplication is used very often, so that the accuracy deteriorates and the accuracy of 24 bits becomes difficult.)
No processing accuracy is required, that is, the processing accuracy can be remarkably improved, the 24 bit high bit can be easily coped with, and the degree of output deterioration falls within several bits in a high frequency range.

<Fifth Embodiment (FIGS. 1 and 2): Encoding Apparatus> The encoding apparatus according to the fifth embodiment has the same overall block diagram as that of the first embodiment shown in FIG. The signal processing circuit 2 is as shown in FIG. The differences from the first embodiment are as follows. That is, in the fifth embodiment, difference information is scaled for each predetermined frame (data section), and errors caused by compression are more effectively suppressed. In the signal processing circuit 2, as shown in FIG. 2, first, signals хc1,..., Хc2, on the curve β of FIG. 3 band-limited by a low-pass filter 6, for example, an FIR filter, which passes a half band.・ 、 Хc3 、 ・
・ Get The characteristic of the FIR filter is such that it rolls off sharply as shown in FIG. Next, the difference calculator 9 calculates хbi−хci = Δi. Here, the difference Δi is represented as 12-bit normalized scaled data by the allocation circuit 10 at the subsequent stage. This is a fixed or variable bit.

The above-mentioned sampling data хci and the difference data Δi are inputted to the allocation circuit 10, and the allocation circuit 10 respectively receives the peak value Sp in the frame.
After normalizing from, packing processing is performed in a predetermined arrangement and output. Here, when the user data (packet) is 2034 bytes like a DVD, 455 pieces of sampling data Δci and 455 pieces of difference data Δi are stored. The subheader is 7 bytes. Three bytes are allocated to a peak value Sp as a scaling factor (scaling information) used for scaling.

<Sixth Embodiment (FIGS. 6 and 10): Decoding Apparatus> The decoding apparatus of the sixth embodiment differs from the second embodiment in the following points. That is, the signal processing circuit 13 of FIG.
The input signal 入 力 ci and Δi supplied to the input terminal of the signal processing circuit 13 are used to calculate Δi · Sp + ・ ci = хbi in the adder 16. Note that Sp is a peak value used for scaling during encoding. Here, the output хbi is represented by the same 24 bits as the original digital data (хbi in FIG. 3). The interpolation processing circuit 17 uses a plurality of the output data хbi to obtain inter-sampling data хa1, хa2,
..., хai, ... are calculated.

<Seventh Embodiment (FIGS. 1 and 4): Encoding Apparatus> The seventh embodiment is a modification of the third embodiment, and will be described focusing on the differences. In the signal processing circuit 2, FIG. As shown in (1), first, signals хc1,..., Хc2,..., Хc3,... On the curve β whose band is limited by a low-pass filter 6 that passes a half band, such as an FIR filter, are obtained. The characteristic of the FIR filter is such that it sharply rolls off as shown in FIG. Next, the signals хc1, хc2,
差分 c3,.. and the signals хb1, хb2, хb3,..., and the difference calculator 9 calculates хbi−хci = Δ1i. Here, the difference Δ1i is represented by, for example, a 6-bit normalized value in the allocation circuit 10 at the subsequent stage.
Next, signals хa1, хa2,..., Хai,... Are selected, and the difference calculator 9 calculates хai−хci = Δ2i. Here, the difference Δ2i is represented as normalized scaled data of, for example, 6 bits in the allocation circuit 10 at the subsequent stage. This is a fixed or variable bit. Note that normalization (performed for similar peak value P _R in the R channel) that carried out for the peak value P _L of L channel combined difference Δ1i and differential Deruta2i.

The allocation circuit 10 receives the sampling data хbi and the difference data Δ1i, Δ2i,
The allocation circuit 10 normalizes the peak value Sp in the frame, performs packing processing in a predetermined arrangement, and outputs the result. Here, when the user data (packet) is 2034 bytes like a DVD, 225 pieces of the sampling data Δci and 225 pieces of the difference data Δ1i and Δ2i are stored respectively. The subheader is 9 bytes. That is, the difference data Δ1i, Δ2i
Are allocated as 6-bit difference data Δ1i and Δ2i for the L and R channels, respectively, and are treated as the third channel. Each of the sub-header 3 bytes of the peak value P _L, to place the P _R.

The DVD encoding circuit 4, which is a packing encoder, receives the one-time sampling data хai and the difference data Δ1i, Δ2i, performs a packing process in an arrangement as shown in FIG. 13, and outputs it to an output terminal.

<Eighth Embodiment (FIGS. 6 and 12): Decoding Device> The eighth embodiment is a modification of the fourth embodiment, and the decoding device differs from the fourth embodiment in the following points. That is, the signal processing circuit 13 of FIG. 6 is configured as shown in FIG.
In the block diagram of FIG. 5, the description will be made assuming that an input signal (encoded data) input to the left input terminal is decoded into a signal having a waveform as shown in FIG.

In FIG. 6, an input signal is supplied to a demodulation circuit 11
The signal is demodulated by the DVD decoding circuit 12, is unpacked and decoded by the DVD decoding circuit 12, generates interpolation data by the signal processing circuit 13 and the memory 14, is converted into an analog signal by the D / A converter 15, and is output. In the signal processing circuit 13, as shown in FIG. 12, the input signal supplied to the input terminal is multiplied by the difference between the peak value and the adder 16 to calculate Δ1i · Sp1 + хci = хbi. Here, the output хbi is represented by 24 bits. Next, the adder 18 calculates Δ2i · Sp2 + ci = хai. Here, the output хai is represented by 24 bits. Therefore, the data is хb1, хa1, ..., хbi, хai,
D / A converter 1 by outputting in the order
5 into an analog signal at 96 kHz.

<Ninth Embodiment (FIGS. 1 and 14): Encoding Apparatus> The ninth embodiment differs from the third embodiment in the following points. That is, the second difference is set as xai-Prediction [i] = Δ2i. Here, xai-Prediction [i] is the Nth-order linear prediction value, and when N = 3, Prediction [i] = A1 · xa, i−1 + A2 · xbi + A3 · x
a, i + 1 (A1, A2, A3 are prediction coefficients) and three points xa, i-
It can be predicted as a point on a curve passing through 1, xbi, xa, i + 1. Such linear prediction is performed by the signal processing circuit 2 shown in FIG.
This is performed by the linear prediction circuit 19 in FIG. The output signal Prediction [i] of the linear prediction circuit 19 is input to the adder 9B which operates as a subtractor, and is subtracted from the input signal xa. In this embodiment, the A / D converter 1 uses a high sampling frequency of 192 kHz, and the characteristics of the LPF 6 are as shown in FIG.

<Tenth Embodiment (FIGS. 6 and 15): Decoding Device> The tenth embodiment differs from the fourth embodiment in the following points.
That is, the adder 18 calculates Δ2i + Prediction [i] = xai. Here, Prediction [i] is an Nth-order linear prediction value, and when N = 3, Prediction [i] = A1 · xa, i−1 + A2 · xbi + A3 · x
It is represented by a, i + 1 and can be predicted as a point on a curve passing through three points xa, i-1, xbi, xa, i + 1. Such linear prediction is performed by the linear prediction circuit 20 in the signal processing circuit 6 shown in FIG. In this embodiment, a high sampling frequency of 192 kHz is used in the D / A converter 15, while 96 kHz can be reproduced from the 1 × data. As shown in the above embodiment, any data may be used as a reference for detecting the second difference information.

<Eleventh Embodiment (FIGS. 1 and 17): Encoding Apparatus> The encoding apparatus according to the eleventh embodiment has the same overall block diagram as shown in FIG. Only the inside signal processing circuit 2 is different as shown in FIG. Therefore, the eleventh embodiment will be described focusing on the differences from the first embodiment. In the block diagram of FIG. 1, an input signal (audio signal) coming into the input terminal IN is sampled at a time interval of Δt with a waveform as shown in FIG. 18 (for example, sampled at a high rate of 192 kHz). It will be described as. The input audio signal is converted into a PCM signal by the A / D converter 1 at a sampling frequency four times as high as that of a normal PCM.
From the output terminal of the converter 1, the sampling data {b1,
хd1, хa1, хe1, хb2, хd2, хa2, хe2, хb3, х
d3, хa3, хe3 ... These sampling data are converted at a high resolution of, for example, 20 bits. Next, these signals are compressed by the signal processing circuit 2 using the memory 3 and packed by the DVD encoding circuit 4.
The signal is directly output from the output terminal OUT1 or, if necessary, modulated by the modulation circuit 5 and then recorded on a recording medium (not shown).

In the signal processing circuit 2, as shown in FIG. 17, first, a band is limited by a low-pass filter 6 that passes a quarter band, for example, an FIR filter.
, 1c2,..., Хc
,... Are obtained, and then signals хc1, хc2, хc3,...
The output signal of the thinning circuit 7 is provided to the minus input terminal of the first difference calculator 9A. On the other hand, the input signal of the low-pass filter 6, that is, the output signal of the A / D converter 1 is given to the switch 21 acting as the distribution / selection means, and the curve α
Хbi and хai at the above intervals of 4Δt are distributed and selected, хbi is given to the + input terminal of the first difference calculator 9A and the average value calculator 22, and хai is given to the + input terminal of the second difference calculator 9B. . The average value calculator 22 adds two successive хbi and then divides the result to obtain an average value (хbi + хb, i + 1) of two consecutive data for each 4Δt on the curve α.
/ 2 is given to the minus input terminal of the second difference calculator 9B.

The first difference calculator 9A calculates the хci at the 4Δt interval on the curve β, which is thinned out to the above quarter, and the 分配 ci at the 4αt interval distributed and selected on the curve α, at the same timing as the хci. Is calculated, and хbi−хci = Δ1i... First difference is obtained. Here, the first difference Δ1i is represented by, for example, 10 bits or less. This is a fixed or variable bit. In addition, the second difference calculator 9B calculates the ４ai at a 4Δt interval thinned to a quarter on the curve α and the average value (хbi + хb, i + 1) / 2 of two consecutive data for each 4Δt on the curve α.
Хai− (хbi + хb, i + 1) / 2 = Δ2i... The second difference is obtained. Here, the second difference Δ2i is represented by, for example, 14 bits or less. This is a fixed or variable bit.

The allocation circuit 10 stores the sampling data хci and the first and second difference data Δ1i, Δ2i
Is input, the allocation circuit 10 performs packing processing in a predetermined arrangement, and outputs the result. Here, when the user data (packet) is 2034 bytes like a DVD, 406 pieces of the sampling data Δci and 406 pieces of difference data Δ1i and Δ2i are stored. The subheader is 4 bytes.

<Twelfth Embodiment (FIGS. 6 and 19): Decoding Apparatus> The decoding apparatus according to the twelfth embodiment has the same configuration as that of the second embodiment as shown in FIG. The input signal (encoded data) coming into the left input terminal in the block diagram is shown in FIG.
Description will be made on the assumption that the signal is decoded into a signal having a waveform as shown in FIG. In FIG. 6, the input signal is a demodulation circuit 1
The signal is demodulated by 1 and is unpacked and decoded by the DVD decoding circuit 12, generates interpolation data by the signal processing circuit 13 and the memory 14, is converted into an analog signal by the D / A converter 15, and is output.

The signal processing circuit 13 in the twelfth embodiment
As shown in FIG. 19, the first adder 16A calculates Δ1i + хci = хbi using the input signal Δci supplied to the left input terminal and the first difference Δ1i. Here, the output хbi is represented by the same 24 bits as the original digital data (хbi in FIG. 18). The output data хbi of the first adder 16A is provided to the output terminal, the interpolation processing circuit 17 and the average calculator 23. The average value calculator 23 adds two successive хbi and then divides it to obtain an average value (хbi + хb, i + 1) / 2 of two continuous data for each 4Δt, and obtains the second adder 16B To the input terminal. The second difference data Δ2i is input to the second adder 16B, and Δ2i + (хbi + хb, i + 1) / 2 = хai is calculated. Here, the output хai is represented by the same 24 bits as the original digital data (хai in FIG. 18).

The output data xai of the second adder 16B is supplied to an output terminal and the interpolation processing circuit 17. The interpolation processing circuit 17 uses a plurality of data xai and xbi to sample data xdl, xe1, xd2, xe as shown in FIG.
2,..., X di , x e i,. For example, in the up-sampling method, each value is filled with 0 data, and the value xai
Can be requested. Alternatively, a curve approximation or a prediction approximation can be used.

Then, the data is converted into xbl, xdl, xal,
xel, ···, xbi, xdi, xai, xei, ·
.. The D / A converter 15 is output in the order as shown in FIG.
Is converted into an analog signal at 192 kHz. In FIG. 6, D which is an unpacking decoder is used.
1 from the VD decoding circuit 12 without passing through the signal processing circuit 13
The double data xci is directly supplied to the D / A converter 15 and can be converted at a 24-bit accuracy at, for example, 48 kHz.

Here, considering the processing in the signal processing circuit 13 shown in FIG. 6, one of them is only the operation of addition (16A, 16B), the original data #bi and #ai are reproduced without deterioration, and the other processing is performed. Is only an interpolation process (17), and can be handled by one filter operation of, for example, a finite coefficient (acyclic) (as is well known, the QMF filter is
Including FFT operation, multiplication is very often used in FFT operation, so the accuracy deteriorates and the accuracy of 24 bits becomes difficult.) The processing accuracy is remarkable because it can be approximated well without requiring processing accuracy. To 24 bits of high bits, and the degree of output degradation falls within several bits in the high frequency range.

Also, a recording medium (carrier) can be manufactured according to the above embodiment by manufacturing a master from the signal of the output terminal OUT2 in FIG. FIG.
The interpolation processing circuit 17 of No. 9 uses a plurality of хbi and хai, and uses inter-sampling data хd1, хe1,..., Хdi, хe
Although it has been described that i,... are calculated, for example, in the up-sampling method, hold data (ie, хb1 between хb1 and хb2) is buried in each without filling in 0, and FIG. Signals x1, x2,
x3,... That is, intermediate interpolation values xa1, xa2, x
It is not limited to those seeking a3, ...

<Thirteenth Embodiment (FIGS. 1 and 21): Encoding Apparatus> The encoding apparatus according to the thirteenth embodiment has the same overall block diagram as that shown in FIG. Only the inside signal processing circuit 2 is different as shown in FIG. Therefore, the thirteenth embodiment will be described focusing on the differences from the first embodiment. In the block diagram of FIG. 1, an input signal (audio signal) coming into the input terminal IN is sampled at a time interval of Δt with a waveform as shown in FIG. 18 (for example, sampled at a high rate of 192 kHz). It will be described as. The input audio signal is converted into a PCM signal by the A / D converter 1 at a sampling frequency four times as high as that of a normal PCM.
From the output terminal of the converter 1, the sampling data {b1,
хd1, хa1, хe1, хb2, хd2, хa2, хe2, хb3, х
d3, хa3, хe3 ... These sampling data are converted at a high resolution of, for example, 20 bits. Next, these signals are compressed by the signal processing circuit 2 using the memory 3 and packed by the DVD encoding circuit 4.
The signal is directly output from the output terminal OUT1 or, if necessary, modulated by the modulation circuit 5 and then recorded on a recording medium (not shown).

In the signal processing circuit 2, as shown in FIG. 21, first, a band is limited by a low-pass filter 6 that passes a quarter band, for example, an FIR filter.
, 1c2,..., Хc
,... Are obtained, and then signals хc1, хc2, хc3,...
The output signal of the thinning circuit 7 is provided to the minus input terminal of the difference calculator 9A. On the other hand, the input signal of the low-pass filter 6,
That is, the output signal of the A / D converter 1 is given to the switch 21 acting as distribution / selection means, and хbi, хdi, хai, and хei at Δt intervals on the curve α are distributed / selected, and are sent to the ADPCM encoder 26. Given. Also,
хbi is also given to the + input terminal of the difference calculator 9A.

The difference calculator 9A calculates хbi−хci = Δi to obtain a difference Δi. Here, the difference Δi is represented by, for example, 11 bits or less. This is a fixed or variable bit. On the other hand, ス イ ッ チ bi, хdi, хa
.. are selected and the ADPCM encoder 26 calculates three differences as で di-хbi = Di1 хai-хdi = Di2 хei-хai = Di3. Note that Di1, Di2, and Di3 are represented by the general formula Dik.
(K is an integer of 1 to 3). Here, ADPCM as each difference which is the output signal of the ADPCM encoder 26
The data Dik is represented by, for example, 3 bits or less.

The allocation circuit 10 includes a thinning circuit 7
, The difference Δi from the difference calculator 9A, and the ADPC from the ADPCM encoder 26.
The M data (difference data) Dik is input, the allocation circuit 10 performs packing processing in a predetermined arrangement, and outputs it to an output terminal. Here, when the user data (packet) is 2034 bytes like a DVD, 812 pieces of sampling data хci and 812 pieces of difference data Δi, Dik are stored respectively. The subheader is 4 bytes.

<Fourteenth Embodiment (FIGS. 6 and 22): Decoding Apparatus> The decoding apparatus according to the fourteenth embodiment has the same configuration as that of the second embodiment as shown in FIG. The input signal (encoded data) coming into the left input terminal in the block diagram is shown in FIG.
Description will be made on the assumption that the signal is decoded into a signal having a waveform as shown in FIG. In FIG. 6, the input signal is a demodulation circuit 1
The signal is demodulated by 1 and is unpacked and decoded by the DVD decoding circuit 12, generates interpolation data by the signal processing circuit 13 and the memory 14, is converted into an analog signal by the D / A converter 15, and is output.

In the signal processing circuit 13 of the fourteenth embodiment, as shown in FIG. 22, the adder 16A calculates Δi + хci = хbi using the input signal хci supplied to the left input terminal and the difference Δi. You. Here, the output хbi of the adder 16A is represented by the same 20 bits as the original digital data (хbi in FIG. 18). The output data $ bi of the adder 16A is supplied to one input terminal of a switch 28 as a selection means and the ADPCM decoder 27. The ADPCM data Dik supplied to the input terminal is also input to the ADPCM decoder 27. A
The DPCM decoder 27 outputs the output хbi of the adder 16A and A
Using DPCM data Dik, data between samplings
Calculate and obtain di, хai, хei. Where the outputs хdi, хa
i and хei are represented by the same 20 bits as the original digital data (хbi in FIG. 18).

Therefore, these data, that is, adder 1
6A output хbi and ADPCM decoder 27 output хd
i, хai, хei, хb1, хd1, хa1, хe1, ...,
The digital signal is converted into an analog signal at 192 kHz by sending it out of the switch 28 and supplying it to the D / A converter 15 so that the order is like $ bi, $ di, $ ai, $ ei.

In the configuration shown in FIG. 6, one-time data Δci is directly supplied to the D / A converter 15 without passing through the signal processing circuit 13 from the DVD decoding circuit 12 which is an unpacking decoder. It can be configured to convert with bit precision. Here, considering the processing in the signal processing circuit 13 of FIG. 6, one of them is added (16A)
, The original data хbi is reproduced without deterioration, and the other processing is only the adaptive difference processing (26).
For example, a single filter operation of a finite coefficient (acyclic) can be used (as is well known, the QMF filter is F
Including the FT operation, the multiplication is used very often in the FFT operation, so the accuracy deteriorates and the accuracy of 20 bits becomes difficult.) Since the processing accuracy can be performed without requiring, the processing accuracy can be remarkably improved. It is possible to easily cope with 20 bits of high bits, and the degree of deterioration of the output falls within several bits in a high frequency band.

In the description of the first embodiment, the output terminal OU shown in FIG.
Although it has been described that the recording medium (carrier) can be manufactured according to the first embodiment by manufacturing a master from the signal of T2, etc., it is also possible to manufacture the recording medium according to the other embodiments of the encoding device. It goes without saying that a medium can be manufactured.

[0070]

As described above, since the encoding device, the decoding device and the medium on which the encoded information is recorded according to the present invention are constructed as described above, they have the following effects. That is, it is possible to further suppress signal deterioration due to signal processing when compressing high bit data of a high sampling frequency. In addition, when high-fidelity data at a high sampling frequency is reproduced at a normal sampling rate (48 kHz) and compressed data is decoded, signal processing requiring high precision can be avoided, and provided at low cost. . Further, compatibility with the normal sampling rate is obtained, and the compression ratio can be set relatively freely by the data length of the difference.

[Brief description of the drawings]

FIG. 1 is a block diagram of an encoding device as a first embodiment, a third embodiment, a fifth embodiment, a seventh embodiment, a ninth embodiment, an eleventh embodiment, and a thirteenth embodiment of the present invention. .

FIG. 2 is a block diagram illustrating a configuration of a signal processing circuit in FIG. 1 according to a first embodiment and a fifth embodiment.

FIG. 3 is a signal waveform diagram for explaining the operation of the encoding device of the present invention.

FIG. 4 is a block diagram showing a configuration of a signal processing circuit in a third embodiment and a seventh embodiment in FIG. 1;

FIG. 5 is a signal waveform diagram for explaining the operation of the encoding device of the present invention.

FIG. 6 shows a second embodiment, a fourth embodiment, a sixth embodiment, an eighth embodiment, a tenth embodiment, a twelfth embodiment, and a fourteenth embodiment of the present invention.
It is a block diagram of a decoding device as an example.

FIG. 7 is a block diagram showing a configuration of a signal processing circuit in FIG. 6 according to a second embodiment.

FIG. 8 is a signal waveform diagram for explaining the operation of the decoding device of the present invention.

FIG. 9 is a block diagram showing a configuration of a signal processing circuit in FIG. 6 in a case of a fourth embodiment;

FIG. 10 is a block diagram showing a configuration of a signal processing circuit in FIG. 6 in a case of a sixth embodiment;

FIG. 11 is a graph showing frequency characteristics of an FIR filter that can be used as the LPF in FIGS. 2 and 4;

FIG. 12 is a block diagram showing a configuration of an eighth embodiment of the signal processing circuit in FIG. 6;

FIG. 13 is a format diagram schematically showing a packing process in a DVD encoding circuit which is a packing encoder.

FIG. 14 is a block diagram showing a configuration of a ninth embodiment of the signal processing circuit in FIG. 1;

FIG. 15 is a block diagram showing a configuration of the signal processing circuit in FIG. 6 in a case of a tenth embodiment;

FIG. 16 is a graph showing a frequency characteristic of the LPF of FIG. 14;

FIG. 17 is a block diagram showing a configuration of the signal processing circuit in FIG. 1 in the case of an eleventh embodiment.

FIG. 18 is a signal waveform diagram for explaining the operation of the encoding device of the present invention.

FIG. 19 is a block diagram showing a configuration of the signal processing circuit in FIG. 6 in the case of a twelfth embodiment.

FIG. 20 is a signal waveform diagram for explaining the operation of the decoding device of the present invention.

FIG. 21 is a block diagram showing a configuration of the signal processing circuit in FIG. 1 in the case of a thirteenth embodiment.

FIG. 22 is a block diagram illustrating a configuration of a signal processing circuit in FIG. 6 in a case of a fourteenth embodiment;

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 A / D converter (A / D conversion means) 2, 13 Signal processing circuit 3, 14 Memory 4 DVD encoding circuit 5 Modulation circuit 6 LPF (Low-pass filter means) 7, 8 Decimation circuit (Decimation means) 9, 9A 9B Adder (subtractor: difference calculator: difference detecting means) 10 Allocation circuit (means for transmission) 11 Demodulation circuit 12 DVD decoding circuit 15 D / A converter 16, 16A, 16B, 18 Adder 17 Interpolation processing circuit 19, 20 Linear prediction circuit 21 Switch (distribution / selection means) 22, 23 Average value calculator 26 ADPCM encoder 27 ADPCM decoder 28 Switch (selection means)

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-7-143443 (JP, A) JP-A-5-243906 (JP, A) JP-A-6-268527 (JP, A) (58) Field (Int.Cl. ⁷ , DB name) H03M 3/04 G11B 20/10 301 H03M 1/06

Claims (18)

(57) [Claims] An analog signal is sampled at a sampling frequency sufficiently higher than twice the maximum frequency of the analog signal and is converted from analog to digital to obtain sampling code information.
/ D conversion means, low-pass filter means for passing a band frequency component of approximately 1/2 of the sampling code information, decimation means for thinning out the output of the low-pass filter means to 1/2, and A / D conversion means. First difference information detecting means for obtaining difference information between every other sampling information of the analog-to-digital converted code information and output information of the decimation means, and second difference information detecting means for obtaining difference information for every other sample. An encoding device comprising: a unit for transmitting output information of the decimation unit and the difference information. 2. An analog signal obtained by sampling an analog signal at a sampling frequency sufficiently higher than twice the highest frequency thereof and performing analog-to-digital conversion to obtain sampling code information.
/ D conversion means, low-pass filter means for passing a band frequency component of approximately 1/2 of the sampling code information, decimation means for thinning out the output of the low-pass filter means to 1/2, and A / D conversion means. A difference information detecting means for obtaining difference information between information of every other sampling of the analog-digital converted code information and output information of the decimation means; a means for scaling the difference information for each predetermined frame; Means for transmitting output information and means for transmitting said scaled difference information. 3. An analog signal obtained by sampling an analog signal at a sampling frequency sufficiently higher than twice the maximum frequency thereof and performing analog-to-digital conversion to obtain sampling code information.
/ D conversion means, low-pass filter means for passing a band frequency component of approximately 1/2 of the sampling code information, decimation means for thinning out the output of the low-pass filter means to 1/2, and A / D conversion means. First difference information detecting means for obtaining difference information between every other sampling information of the analog-to-digital converted code information and output information of the decimation means, and second difference information detecting means for obtaining difference information for every other sample. An encoding device comprising: means for scaling the two pieces of difference information for each predetermined frame; and means for transmitting output information of the decimation means and the scaled two pieces of difference information. 4. An analog signal obtained by sampling an analog signal at a sampling frequency sufficiently higher than twice the highest frequency thereof and converting the analog signal into a digital signal to obtain sampling code information.
/ D conversion means, low-pass filter means for passing a band frequency component of approximately 1/2 of the sampling code information, decimation means for thinning out the output of the low-pass filter means to 1/2, and A / D conversion means. First difference information detecting means for obtaining difference information between information of every other sample of analog-to-digital converted code information and output information of the decimation means; prediction for obtaining a predicted value for every other sample from the sampling code information Means, second difference information detecting means for obtaining difference information between the predicted value and every other sample, means for scaling the two difference information for each predetermined frame, output information of the decimation means, and the scaling Means for transmitting the obtained two difference information. 5. An A / D converter for sampling an analog signal at a sufficiently high sampling frequency of four times or more the highest frequency thereof and performing analog-to-digital conversion to obtain sampling code information; Low-pass filter means for passing a band frequency component of No. 4; decimation means for thinning out the output of the low-pass filter means to 1/4; information for every three samplings of code information analog-to-digital converted by the A / D conversion means; First difference information detecting means for obtaining difference information from output information of the decimation means; information for every three samplings of code information analog-to-digital converted by the A / D conversion means; Difference between digitally converted code information and other information every three samplings or its function A second differential information detecting means for obtaining, and the means for transmitting output information and the two difference data of the decimation means comprises encoding apparatus. 6. A / D conversion means for sampling an analog signal at a sufficiently high sampling frequency of at least four times the highest frequency thereof and performing analog-to-digital conversion to obtain sampling code information; Low-pass filter means for passing a band frequency component of No. 4; decimation means for thinning out the output of the low-pass filter means to 1/4; information for every three samplings of code information analog-to-digital converted by the A / D conversion means; Difference information detecting means for obtaining difference information from output information of the decimation means; encoding means for encoding information other than the information at every third sampling in the sampling code information by ADPCM; output information of the decimation means And transmitting the difference information and the output signal of the encoding means. And means, having the encoding device. 7. An output of a low-pass filter means for passing a substantially half band frequency component of the sampling code information in the analog-to-digital conversion is decimated to 1/2 by a decimation means, and one sampling of the analog-to-digital converted code information is performed. When information indicating a difference between every other information and the output information of the decimation means is set as difference information, an adding means for adding the transmitted sampling code information and the difference information, and information on every other sample is obtained from the addition result. Means for decoding. 8. The decimation means outputs the output of a low-pass filter means for passing a band frequency component of approximately 1/2 of sampling code information in an analog-to-digital conversion at a sampling frequency twice or more the highest frequency of an analog signal. The difference information between every other sampling of the code information which has been decimated to 2 and converted from analog to digital and the output information of the decimation means is defined as first difference information, and the difference information for every other sample is defined as second difference information. A decoding device responsive to code information and said first and second difference information when they are transmitted, said decoding apparatus adding said transmitted first difference information to said sampling code information, and halving said sampling code information. Decoding means for decoding the remaining half of the information using the transmitted second difference information And stage, and means for digital-to-analog conversion at a sampling frequency during the analog-to-digital conversion by using the decoded signal by said means comprises decoding device. 9. An output of a low-pass filter means for passing a band frequency component of approximately 1/2 of sampling code information in an analog-to-digital conversion at a sampling frequency of twice or more of a maximum frequency of an analog signal, is output to a decimation means. The information indicating the difference between every other sampling of the analog-to-digital converted code information and the output information of the decimation means is used as difference information, and the difference information is scaled and transmitted together with the sampling code information. A decoding device responsive to these, a means for controlling the value of the scaled difference information using transmitted scaling information, and adding the controlled difference information to the sampling code information; Decoding means for decoding half of the sampling code information; A decoding device, comprising: decoding means for decoding all the sampling information from the decoded information, and digital-to-analog conversion at a sampling frequency at the time of the analog-to-digital conversion using a signal decoded by the decoding means. 10. An output of a low-pass filter means for passing a band frequency component of approximately 1/2 of sampling code information in an analog-to-digital conversion at a sampling frequency of twice or more of a maximum frequency of an analog signal is decimation by a decimation means. 2, the difference information between the information of every other sampling of the code information which has been decimated and converted from analog to digital and the output information of the decimation means is defined as the first difference information.
A decoding device that responds to the difference information for every other sample as second difference information, wherein the first and second difference information are scaled and transmitted together with the sampling code information. Means for controlling the values of the scaled first and second difference information using: a decoding means for adding the controlled first difference information to the sampling code information and decoding half of the sampling code information Means for decoding the remaining half of the information using the controlled second difference information; and means for performing digital-to-analog conversion at a sampling frequency at the time of the analog-to-digital conversion using the signal decoded by the means. Decoding device. 11. An output of a low-pass filter means for passing a band frequency component of approximately one-fourth of sampling code information in an analog-to-digital conversion at a sampling frequency of four times or more of a maximum frequency of an analog signal is decimation by a decimation means. 4, information indicating the difference between the information obtained at every third sampling of the code information subjected to the analog-to-digital conversion and the output information of the decimation means is defined as first difference information;
The difference information between every three samplings of the analog-to-digital converted code information and other information or its function every three samplings of the analog-to-digital converted code information is defined as second difference information. A decoding device responding to the first and second difference information when the first and second difference information are transmitted, wherein the transmitted first difference information is added to the sampling code information to decode 1/4 of the sampling code information. Means for decoding all of the sampling code information from a plurality of pieces of decoded information and the second difference information, and means for performing digital-to-analog conversion at a sampling frequency at the time of the analog-to-digital conversion using a signal decoded by the means. And a decoding device comprising: 12. An output of a low-pass filter means for passing a band frequency component of approximately one-fourth of sampling code information in an analog-to-digital conversion at a sampling frequency of four times or more of a maximum frequency of an analog signal is decimation by a decimation means. 4, the information indicating the difference between the information of every three samplings of the code information which has been decimated and converted from analog to digital and the output information of the decimation means is defined as difference information, and the sampling code information, the difference information, and the sampling code information When an ADPCM signal obtained by encoding information other than the information at every third sampling by ADPCM is transmitted, a decoding device responding to the signal, wherein the transmitted difference information is added to the sampling code information, The three samplers corresponding to a quarter of the sampling code information. Means for decoding every other information; an ADPCM decoder for decoding the sampling code information other than the information every three samplings using the ADPCM signal and an output signal of the decoding means; and an output of the decoding means. Means for switching and outputting a signal and a plurality of output signals of the ADPCM decoder to decode the sampling code information; and D for performing digital-to-analog conversion on the sampling code information at a sampling frequency at the time of the analog-to-digital conversion.
/ A conversion means. 13. An analog signal having the highest frequency of 2
With respect to the sampling code information sampled at a sampling frequency sufficiently higher than twice and analog-to-digital converted by the A / D conversion means, a band frequency component of approximately 1/2 of the sampling code information is passed through the low-pass filter means. The output of the low-pass filter is decimated to 1/2 by the decimation means, and the difference information between the information of every other sampling of the code information analog-to-digital converted by the A / D conversion means and the output information of the decimation means is calculated. The analog signal is obtained by the first difference information detecting means, the difference information for every other sample is obtained by the second difference information detecting means, and the output information of the decimation means and the difference information are transmitted by the transmitting means. A recorded medium of encoded information in which a signal obtained by encoding is recorded. 14. An analog signal having its highest frequency of 2
With respect to the sampling code information sampled at a sampling frequency sufficiently higher than twice and analog-to-digital converted by the A / D conversion means, a band frequency component of approximately 1/2 of the sampling code information is passed through the low-pass filter means. The output of the low-pass filter is decimated to 1/2 by the decimation means, and the difference information between the information of every other sampling of the code information analog-to-digital converted by the A / D conversion means and the output information of the decimation means is calculated. The difference information is obtained by the difference information detection means, the difference information is scaled for each predetermined frame, and the output information of the decimation means and the encoded information of the signal obtained by encoding the analog signal using the scaled difference information are recorded. Recorded media. 15. An analog signal having the highest frequency of 2
With respect to the sampling code information sampled at a sampling frequency sufficiently higher than twice and analog-to-digital converted by the A / D conversion means, a band frequency component of approximately 1/2 of the sampling code information is passed through the low-pass filter means. The output of the low-pass filter is decimated to 1/2 by the decimation means, and the difference information between the information of every other sampling of the code information analog-to-digital converted by the A / D conversion means and the output information of the decimation means is calculated. The first difference information detection means obtains the difference information for every other sample by the second difference information detection means. The two difference information are scaled for each predetermined frame, and the output information of the decimation means and Recording a signal obtained by encoding the analog signal using the scaled difference information Recorded medium encoded information. 16. An analog signal having its highest frequency of 2
With respect to the sampling code information sampled at a sampling frequency sufficiently higher than twice and analog-to-digital converted by the A / D conversion means, a band frequency component of approximately 1/2 of the sampling code information is passed through the low-pass filter means. The output of the low-pass filter is decimated to 1/2 by the decimation means, and the difference information between the information of every other sampling of the code information analog-to-digital converted by the A / D conversion means and the output information of the decimation means is calculated. A first difference information detecting means obtains a predicted value for every other sample from the sampling code information, and a difference information for the predicted value and the other every other sample is obtained by a second difference information detecting means. Scaling the two pieces of difference information for each predetermined frame; and outputting the output of the decimation means. A recorded medium of encoded information in which a signal obtained by encoding the analog signal using information and the scaled difference information is recorded. 17. An analog signal having the highest frequency of 4
With respect to the sampling code information sampled at a sufficiently high sampling frequency twice or more and analog-to-digital converted by the A / D conversion means, a band frequency component of approximately 1/4 of the sampling code information is passed by the low-pass filter means. The output of the low-pass filter means is thinned out to 1/4 by the decimation means, and the difference information between the information of every three samplings of the code information analog-to-digital converted by the A / D conversion means and the output information of the decimation means Is obtained by the first difference information detecting means, and information of every three samplings of the code information analog-digital converted by the A / D converting means and three samplings of the code information analog-digital converted by the A / D converting means are obtained. Second difference information detecting means for obtaining difference information from every other information or its function, Output information and said two difference information and recorded medium recorded encoded information signal encoding the analog signal using the Shon means. 18. An analog signal having the highest frequency of 4
With respect to the sampling code information sampled at a sufficiently high sampling frequency twice or more and analog-to-digital converted by the A / D conversion means, a band frequency component of approximately 1/4 of the sampling code information is passed by the low-pass filter means. The output of the low-pass filter means is thinned out to 1/4 by the decimation means, and the difference information between the information of every three samplings of the code information analog-to-digital converted by the A / D conversion means and the output information of the decimation means In the sampling code information, information other than the information at every third sampling is encoded by the encoding means by ADPCM, and the output information of the decimation means, the difference information, and the And a recording of a signal obtained by encoding the analog signal using an output signal. Recorded medium of encoding information.