JPH09261068A - Data compression/decoding/transmission/reception/ recording/reproduction method and device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain the data compression method/decoding method without any loss and to obtain the transmission and recording method for a broad band digital signal and the transmission and recording device while keeping compatibility with a format processing a digital signal with a narrow band. SOLUTION: A sub-band analysis filter 11 is used to separate an inputted digital signal into a low frequency signal and a high frequency signal, the low frequency signal is inputted to a formatter 13, from which the resulting signal is transmitted and the high frequency signal is compressed by a loss-less coder 12, inputted to the formatter, from which the resulting signal is transmitted. A decoder side receives the transmission signal, reproduces the low frequency signal as it is and the high frequency signal is compression-decoded by a loss- less decoder 22, and the decoded signal and the low frequency signal are synthesized by a sub-band synthesis filter 23, from which a substantial broad-band reproduction signal is obtained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital signal compression and decoding method and apparatus. Further, the present invention utilizes this digital signal compression / decoding method to transmit or record and receive or reproduce a high band digital signal while maintaining compatibility with a format handling a narrow band digital signal. Effective digital signal transmitting method and device, digital signal receiving method and device, digital signal recording method and device, digital signal reproducing method and device, digital signal lossless compression method, and the digital signal being recorded Recording medium.

[0002]

2. Description of the Related Art Since the advent of CD, digital audio signals have been widely accepted, and the image of CD = high fidelity has been established, but it has been over ten years since its release, and due to the progress of audio equipment, the sound quality of CD is insufficient as high fidelity. The recognition that it is is spreading.

By the way, recently, a disc having a higher density than that of a CD has been developed, and a digital video disc capable of recording digital video and digital audio has been developed and standardized. For digital video discs,
In order to record a long time of the audio signal together with the video, both signals are compressed and recorded even if the disc has a high density. On the other hand, in order to emphasize the sound quality, it is also considered to perform recording with a sound quality improved from that of the CD. More specifically, it is possible to handle uncompressed digital audio in addition to compressed audio. This uncompressed digital audio has a sampling frequency of 48 kHz, a sampling frequency of 16 bits, and a sampling frequency of 96 kHz and 1 sample. It can handle up to 24 bits. This is C
Compared with D, the reproduction band is 20kHz to 48kHz 2
More than double, dynamic range from 98dB to 136dB
It can be dramatically improved.

[0004]

On the other hand, on the other hand, since the audio handled by the digital video disc is treated as the audio accompanying the video signal, this high-density disc is used to treat the audio mainly.
Furthermore, there is a growing demand for ultra-high-fidelity audio discs with extremely high quality sound. A specific demand for high sound quality is to expand the audio band up to about 100 kHz in consideration of the bands of various musical instruments and the bands that can be handled by recording equipment.

By the way, when making a disc dedicated to music,
Compatibility with the previously standardized digital video discs becomes a problem. The newly standardized player for music-only discs can play back the audio of the preceding digital video discs (forward compatibility), as well as the digital music discs that preceded the new music-only discs. It is preferable that the player also has backward compatibility (backward compatibility) in terms of function and sound quality so that it can be reproduced within the performance range of the preceding standard.

Specifically, a new disc has a frequency of 100 kHz.
48k with a digital video disc player, even if a band digital audio signal was recorded
This means that components up to the Hz band can be reproduced. Forward compatibility is more preferable if there is a commonality in the disc formats, but basically it is possible to design so that the player can play both formats. On the other hand, backward compatibility requires that the format of the new disc be carefully created because players of the previous standard will not recognize the format of the new standard.

Further, if a signal having a wide band and a large bit width is recorded, the required bit capacity is increased, and even a high density disc cannot be recorded for a long time.
Moreover, the bit rate to be handled increases, which is not preferable. Specifically, the bit rate of the sampling frequency of 192 kHz and the sample bit width of 24 bits needs to be 6 times the bit rate of the sampling frequency of 48 kHz and 16 bits like CD. Therefore, some compression is needed to save the number of bits. However, from the standpoint of aiming for ultra-high-fidelity, it is necessary to use lossless coding (lossless coding) that can completely reproduce uncompressed data without any deterioration of information in the compression coding process, rather than compression accompanied by sound quality deterioration. To be done.

An object of the present invention is to provide a lossless data compression method and apparatus and a decoding method and apparatus thereof. Another object of the present invention is to provide a wideband digital signal transmission / recording method and transmission / recording apparatus using the lossless compression / decoding method while maintaining compatibility with a format handling a narrow band digital signal. To do.

Another object of the present invention is to provide a reproducing method and a reproducing apparatus for receiving and reproducing the signal transmitted as described above or the recorded signal. Another object of the present invention is to provide a recording medium on which the signal formatted as described above is recorded and the recording medium.

[0010]

In the data compression method and apparatus of the present invention, a block is composed of a plurality of data samples of a digital signal, and the block within the block is formed according to a signal level that can be shared by the data samples within the block. All the data samples of (1) are compressed blocks by reducing a certain number of bits from the code bit side, and the information of the reduced number of bits is added to the compressed block.

Further, in the data compression method and apparatus of the present invention, the compressed block is divided into smaller sub-blocks,
Depending on the signal level that the data samples in the sub-block can be shared for each sub-block, all the data samples in the sub-block are further reduced by a certain number of bits from the sign bit side to make a compressed sub-block. Information is added to either or both of the compressed block and the compressed sub-block.

In the data compression method and apparatus of the present invention, the size of the sub-block can be changed. Further, the size of the sub-block can be adaptively changed according to the content of the signal.

Further, in the data compression method and apparatus of the present invention, the representation bit number of the reduced bit number information of the sub-block can be changed for each block. Further, the number of expressed bits of the information of the reduced number of bits of the sub-block can be adaptively changed according to the content of the signal.

Further, the data compression method and apparatus of the present invention are characterized in that the difference between before and after the data sample is taken and the difference data is processed. Further, in the compressed data decoding method and apparatus of the present invention, a block is composed of a plurality of data samples of a digital signal, and all the data samples in the block are classified according to the signal level that can be shared by the data samples in the block. A compressed block is formed by reducing a certain number of bits from the code bit side, and the information on the reduced bit number captures the encoded data added to the compressed block, and the information on the reduced bit number is used. The reduced bits are then decoded and added to each data sample of the compressed block.

Further, in the digital signal transmission / reception / recording / reproducing method and apparatus of the present invention, on the transmission or recording side, a sub-band analysis filter for dividing the input digital signal into a low frequency signal and a high frequency signal is provided. The low-frequency signal is transmitted or recorded as it is, the high-frequency signal is compressed or data is transmitted or recorded by any of the methods described above, and the transmission or recording is performed on the receiving or reproducing side. Received signal, the low-frequency signal is reproduced as it is, the high-frequency signal is compressed and decoded, and the decoded signal and the reproduced low-frequency signal are combined by a combining subband filter to obtain a reproduced signal having an original band. The digital signal is generated and reproduced before compression.

Further, according to the transmission / recording / reception / reproduction method and apparatus of the present invention, the low-frequency signal is transmitted or recorded as it is through a sub-band analysis filter for dividing the input digital signal into a low-frequency signal and a high-frequency signal. , The high-frequency signal takes the difference from the previous adjacent data sample, the data is compressed and transmitted or recorded, and the transmitted or recorded signal is received, while the low-frequency signal can be reproduced as it is, while the high-frequency signal can be reproduced. It is characterized by a lossless compression reproduction method in which a signal is compression-decoded and a reproduction signal having an original band is generated by a combined subband filter of the decoded signal and a low-frequency signal.

Further, the transmission / recording / reception / reproduction method and apparatus of the present invention is characterized in that the sign inversion is performed every other data sample of the high frequency signal before the above difference is obtained.

Further, in the transmission / recording / reception / reproduction method and apparatus of the present invention, the high-frequency signal constitutes a block including a plurality of data samples, and the blocks within the block are formed in accordance with the signal level of the data samples. It is characterized by a block compression method in which all data samples are reduced by a fixed number of bits from the code bit side, and the reduced number of bits is added to a block.

Further, in the transmission / recording / reception / reproduction method and apparatus of the present invention, the block-compressed signal is divided into smaller sub-blocks, and each sub-block is divided into sub-blocks according to the signal level in the sub-blocks. Is a block sub-block compression method in which a fixed number of bits is further reduced from the sign bit side of all the data samples and the reduced number of bits is added.

In the transmission / recording / reception / reproduction method and apparatus of the present invention, the block-compressed signal is reduced in number from the code bit side by a certain number of bits according to the signal level of each data sample. It is characterized by a compression method in which a reduced number of bits is added.

Further, in the transmission / recording / reception / reproduction method and apparatus of the present invention, in the compression method for adding the reduced bit number, the information of the reduced bit number is displayed as a difference from the reduced bit number of the previous sample. It is characterized by

The transmission / recording / reception / reproduction method and apparatus of the present invention is characterized in that the size of the sub-block is adaptively changed according to the content of the signal.

Further, the transmission / recording / reception / reproduction method and apparatus of the present invention are characterized in that the representation bit number of the reduced bit number of the sub-block can be changed for each block.

Further, the transmission / recording / reception / reproduction method and apparatus of the present invention are characterized in that the number of expressed bits of the reduced number of bits of the sub-block is adaptively changed according to the content of the signal.

Further, in the recording medium of the present invention, the low-frequency signal is transmitted or recorded as it is, and the high-frequency signal is passed through the sub-band analysis filter which divides the input digital signal into the low-frequency signal and the high-frequency signal, and the high-frequency signal is sampled before the adjacent data And the data is compressed and recorded, and the high-frequency signal constitutes a block including a plurality of data samples,
Depending on the signal level of the data samples in the block, all the data samples in the block are compressed by the block compression method that reduces a certain number of bits from the sign bit side and adds the reduced number of bits to the block. The compressed signal is divided into smaller sub-blocks, and all the data samples in the sub-blocks are further reduced by a certain number of bits from the sign bit side according to the signal level in each sub-block. Is recorded by being compressed by a block sub-block compression method in which

[0026]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a basic block diagram of the present invention. The wideband digital audio input signal is input to the subband analysis filter 11. Here, the sampling frequency is 192 kHz. Although audio data is taken as an example here, various kinds of signals are possible as data types.

The sub-band analysis filter 11 divides the band of the input signal into two bands, a low band and a high band, and thins each output sample to obtain a sampling frequency 9
It becomes a low-frequency signal and high-frequency signal of 6 kHz. Further, the high-frequency signal is data-compressed by the lossless encoder (lossless encoder) 12 to become a high-frequency signal stream, and the low-frequency signal becomes a low-frequency signal stream as it is. The low-frequency signal stream and the high-frequency compressed signal stream are converted by the formatter 13 so as to match the transmission or recording format and become the output stream.

The reproducing side receives or reproduces the transmitted or recorded signal, and the deformatter 21 takes out the low-frequency signal stream and the high-frequency signal stream. The high frequency signal stream is returned to the original uncompressed high frequency signal by the lossless decoder 22, and the uncompressed high frequency signal and the low frequency signal are
The signal is input to the sub-band synthesis filter 23 and becomes an output signal having the original wide band and a sampling frequency of 192 kHz.

Subband analysis filter Filter 11
The signal that is band-divided by and the signal reproduced through the sub-band synthesis filter 23 must completely match the original signal. Such conditions for perfect matching and conditions for complete recomposition are imposed on the characteristics of the subband analysis filter 11 and the subband synthesis filter 23. As such a filter, a quadrature mirror filter (QMF: Quadrature Mir filter) is used.
ror Filter) is known. However, in reality, since there is an error because the calculation accuracy of the filter or the bit width per sample of the band-divided signal is generally limited, perfect re-synthesis is not performed and a certain error is included. This error can be set to about 1 LSB (lest significant bit), and it can be said that the accuracy is sufficient for sound quality.

FIG. 2 shows a spectrum to show the characteristics of the subband filter. FIG. 2A shows the spectrum of the input signal. Here, the sampling frequency of the input signal is set to 192 kHz. This spectrum also shows the general property of audio signals, that the signal energy becomes smaller as the frequency gets higher. FIG. 2 (b) shows the low band signal obtained by sub-band division, and FIG. 2 (c) shows the high band signal. Here, the sampling frequency of both the low-frequency signal and the high-frequency signal is 96 kHz, which is half of 192 KHz, and the sampling frequency is 96 kHz according to the Nyquist principle.
The component of half of 48 kHz or more of Hz is a signal which is folded and copied between 0 and 48 kHz.

Especially, the high frequency signal is changed from the original 48 kHz to 9 kHz.
Signal components up to 6 kHz (Fig. 2 (a)) are from 0 Hz to 4
It folds back to the band up to 8 kHz and, for example, even if this signal is reproduced as it is as a PCM signal and heard, it cannot be heard as an audio signal. On the other hand, the low frequency signal can be reproduced and heard as it is as a PCM signal limited to half the band of the original input signal.

From this, it becomes possible to provide compatibility between the wide band signal format and the narrow band signal format. More specifically, digital video
It is possible to have compatibility between the disc and the new audio-only disc. That is, in the digital video disc, sampling frequencies of 48 kHz and 96 kHz are available as formats for handling non-compressed PCM signals other than high-efficiency compression using the psychoacoustic model, and bit widths are 16, 20, and 24 bits. Therefore, if a low-frequency signal obtained by dividing the above wideband signal having a sampling frequency of 192 kHz by the subband analysis filter 11 is recorded on a disc in a digital video disc format, the digital video disc player will have a sampling frequency of 96 kHz. It can be played back as audio without any problems.

On the other hand, since the high frequency signal cannot be processed by the digital video disk at all,
Video discs must be formatted to ignore this high frequency signal. Digital video
In order to be ignored by the disc, specifically, there are flags and I in the format of the digital video disc.
It may be newly set by D (definition data: Identification Data) or the like, but there are many methods. As a simple example, a new value may be assigned to the high frequency signal stream as an ID that specifies the type of stream such as compressed audio or uncompressed PCM signal.

On the other hand, the new audio-only disc player recognizes both the ID of the uncompressed PCM signal stream and the ID assigned to the new high-frequency signal, extracts the low-frequency signal stream and the high-frequency signal stream, and outputs the high-frequency signal stream. If the original wideband signal is reproduced by the sub-band synthesizing filter 23 by compressing and decoding the range signal, the digital video
You can enjoy super high-fidelity audio with higher sound quality than discs.

The present invention will be further described below. Lossless encoding / decoding (lossless compression reproduction) can completely reproduce the original data bit by bit when compressed and decoded, and is an essential requirement for achieving super high fidelity.

FIG. 3 shows a specific example of the lossless compression reproduction method of the present invention. FIG. 3A shows a sample sequence of one block when the input sample sequence is divided into blocks composed of a plurality of samples. The samples are arranged from top to bottom, and each sample has 24 bits, and the left side is shown as MSB (sign bit) by "M" or "*" to the right. "M" indicates the MSB, and when the same bit value as the MSB continues, the value 2SB or less is also indicated by M, and after the bit value is changed, the LSB is indicated by "*".
This series of "*" can be considered to directly represent the signal level. That is, for large signals the MSB is 2
Change from SB, "*" will extend to 2SB,
Smaller levels will only extend closer to the LSB. In other words, the lower the level of the signal, the “M”
Extends. This "M" sequence is a sequence of the same bit values as the MSB, and therefore can be deleted if the continuous number is known.

In the example of the block shown in FIG. 3A, the bits on the left side of the vertical line can be reduced. The result is the sample sequence shown in FIG. The information on the number of bits reduced here is necessary at the time of decoding, and is added to the sample sequence for each block as a block scale factor. At the time of decoding, if the number of bits removed from the block scale factor is obtained and the reduced number of bits is copied to the MSB side of each sample, the original sample sequence is completely restored. Various methods of transmitting and recording the block scale factor are possible, and each block may be added as an information section, or a plurality of blocks may be collectively transmitted or recorded as a control information block. You may

As described with reference to FIG. 2, generally, the signal energy at the high frequency signal level gradually decreases. In particular, 5
It seems that the level becomes considerably smaller in the band exceeding 0 kHz. Then, the high-frequency signal sample is a sample in which “*” is short, in other words, “M” is continuous, as shown here. Therefore, the method shown in FIG. 3 can significantly reduce the amount of data. The example of FIG. 3 shows an example in which the scale factor as the block side information of the compressed block is 5, that is, the reduction bit number is 5.

FIG. 4 shows an actual processing flow chart for generating the compressed block. Read a sample processing block. Next, the number of consecutive MSBs of each sample is checked to find the minimum number. Determine the number of reduction bits. Each sample is compressed according to the reduced number of bits. Block side information (scale factor) is added and a compressed block is output.

The present invention is not limited to the above embodiment. FIG. 5A shows another embodiment.
Here, the compressed block compressed in FIG. 3 (scale factor as block side information is 5 is further divided into sub blocks, and the MSB is further reduced for each sub block. The side information for each block is the block scale. In addition to the factor, the sub-block size is added to the block as additional information, that is, the block scale factor 5 and the sub-block size 4 are added as side information of the compressed block, and the sub-block size 4 is composed of 4 samples. Means that.
Further, a sub-block scale factor is added as additional information for each sub-block. In this case, in each sample, the sum of the reduced bit number indicated by the block scale factor and the reduced bit number indicated by the sub block scale factor is reduced.

At the time of decoding, each additional information is read, the number of reduced bits is calculated, and the MSB is extended by the number of reduced bits of each sample according to the result, so that the decoding can be performed. This method is used to increase the efficiency of reduction. That is,
If you increase the block size, 1
Even if there is a sample with a large level, the number of reduction bits is determined by it, and the reduction efficiency does not increase. On the contrary, if the block size is made small, it can be locally reduced significantly, but the number of block scale factors that indicate the number of reduction bits is increased, and the bit length must be increased. The efficiency of data compression including it does not increase.

Therefore, as in the present invention, first, the block is reduced in a large size, and the sub-block is used to perform the compression corresponding to the local signal change, whereby the more efficient compression becomes possible.

FIG. 5B shows an example of an actual processing flow. Read the processing block sample block by block. Next, the MSB continuity of each sample is examined to find the minimum number, and the number of reduced bits is determined. Next, it is divided into sub-blocks and the number of reduction bits of each sub-block is determined. Next, each sample is compressed according to the block and sub-block reduction bit numbers. Next, the side information of the sub-block and the side information of the block are added, and the compressed block is output.

FIG. 6A shows an example of a flowchart of another embodiment for determining the size of a sub block. It is not always necessary to uniformly determine the size of the sub-block.

Even if the signal level changes slowly,
In some cases, the size of the sub-block may change drastically, so that the size of the sub-block can be adaptively varied according to the characteristics of the signal, thereby enabling more efficient and efficient compression of all types of signals.

That is, as shown in FIG. 6A, after performing block compression, an arbitrary size of the sub-block is set. Next, after determining the scale factor for each sub-sub-block, the total number of bits is counted after compression including side information at each sub-block size. Next, the block size is changed and the same compression process is performed on each sub-block. Then, select the block size that gives the smallest total number of bits, and restart with that block size.
The sub-block scale factor is determined, the sample is compressed, and side information is added.

Further, although not shown here, more efficient compression can be performed by determining the number of bits expressing the sub-block scale factor for each block. FIG.
In the special case where the sub-block size is 1 (1 sample), (b) shows an example of using the sample scale factor instead of the sub-block scale factor.

FIG. 7 shows another embodiment. FIG. 6 (b)
Similar to the case of, the block compression is performed for each sample, and the scale factor of each sample indicates the difference in the number of reduced bits of the previous sample.

In the example of FIG. 7, 1 bit is given to each sample, and when "1", the reduction bit number of the previous sample + 1 bit is shown, and when "0", the reduction bit number of the previous sample is shown. It indicates that the reduction of -1 bit is performed. Naturally, it is also necessary to limit the reduction bit number if it is negative or is less than the reduction bit number indicated by the block scale factor, or if it exceeds a certain value. In addition, the rule that the beginning of a block starts from the number of reductions indicated by the block scale factor (for example, 5 in the above example) is also necessary. Further, as shown in FIG. 7A, the number of reduced bits may exceed the number of reducible bits. In such a case, it is necessary to go back through the processed sample and modify it to determine the scale factor so as not to cause a failure. By doing so, it is possible to efficiently perform compression with side information having a smaller number of bits.

Similar to FIG. 7, FIG. 8 shows the case where the scale factor of each sample indicates a difference with 2 bits. In this case, the scale factors S are 1, 0, 2, 3 respectively.
In case of, +1, +-with respect to the number of reduced bits of the previous sample
Indicates 0, -1, and -2. In this case as well, for the failure, it is necessary to go back and correct the process as in the previous example. Although the number of bits of the side information increases as compared to the case of 1 bit, it can be compensated for by the increase of the number of reduced bits, which also enables efficient reduction.

In FIG. 9, the scale factor represents the difference by 2 bits, but the scale factor S is 1, 0, respectively.
When it is 2 or 3, + the reduction bit number of the previous sample
Represents 1, + -0, -1, and reset. This reset indicates resetting the reduced bit number to the reduced bit number (5 in this example) indicated by the block scale factor when a failure occurs. In this way, the failure does not occur, the process does not have to be traced back, the encoding process is simplified, and the processing time is shortened.

FIG. 10 shows an actual processing flow chart for realizing the processing of FIGS. 8 and 9. That is, when the block scale factor is obtained, the process of sub-blocks starts. First, an initial value sample and an initial sample bit number are set in order to obtain a difference sample value. In this case, the sample at the beginning of the block and the block scale factor are the targets. Initially, the difference from the block scale factor will be obtained. The number of reduction bits is determined by this difference. It is judged whether or not there is a failure in the number of reduced bits. When there is a failure, the reduced bit number of the previous sample is corrected so that the failure does not occur. Then, go back and make corrections until the failure is resolved.

If there is no failure, the target sample is compressed according to the reduced number of bits, and the differential scale factor is added. This process is performed for all samples in the block. When all the samples in the block are processed, block side information is added and all block information is output.

FIG. 11 shows another embodiment. The same parts as those in the embodiment of FIG. 1 are designated by the same reference numerals. In this embodiment, the encoder on the transmission or recording side is provided with the differential encoder 14 in front of the lossless encoder 12, and the differential decoder 24 is provided at the next stage of the lossless decoder 22 of the decoder on the receiving or reproducing side. Shows.

In order to perform the lossless encoding of the high frequency signal more efficiently, the differential encoder 14 performs the differential encoding before the lossless encoding. The difference encoding transmits / records the difference from the previous sample value instead of transmitting / recording the sample value. The slower the sample value changes, the smaller the amplitude of the difference value. Therefore,
It can be compressed very efficiently in lossless coding.

By the way, as shown in FIG. 2C, the high band signal is converted to the low band signal. At this time, in a general subband filter, the high and low frequencies are inverted. Therefore, as shown in FIG. 2A, although the signal energy decreases as the frequency increases, the signal energy increases as the frequency increases. In differential encoding, the slower the signal changes, that is, the lower the frequency, the more the differential value decreases.
In the high frequency signal shown in FIG. 2C, the amplitude of the difference value cannot be expected to be small. Therefore, it is necessary to invert the height of the frequency component or to shift it by half the sampling frequency to obtain a spectrum as shown in FIG. 2D due to the nature of discrete sample values. As a method, the samples of the high frequency signal may be alternately multiplied by +1 and -1. This is frequency fs / 2 (sampling frequency: fs = 96 kHz
This is equivalent to the multiplication of the carrier of z), and the frequency shift of 48 kHz is performed.

Of course, if the subband filter is provided with this frequency inversion or shift, the frequency inversion or shift here becomes unnecessary. 12A and FIG.
(B) shows specific blocks of the differential encoder and differential decoder 24.

The differential encoder 14 includes a sign inverter 141, a switch 142, a delay register 143, and a subtractor 14
4 Further, the differential decoder 24 is the adder 24
1, a delay register 242, a sign inverter 243, and a switch 244.

In the encoder 14, first, the sample of the input high frequency signal and the signal whose sign is inverted are alternately taken in as the signal S '(n), and S' (n) is stored in the 1-sample delay register 143. To be S from switch 142
′ (N) and the output of the delay register 143 are taken and output as a difference signal.

The differential signal is decoded by summing the differentially decoded signal with the output of register 242 of the previous sample. The decoded signal is stored in delay register 242 for decoding the next sample. When a frequency shift of 48 kHz is performed at the time of encoding, the differentially decoded signal is further output as a code signal by alternately taking a sample as it is and a sample whose sign is inverted as in the case of encoding. .

By the way, in the case of differential encoding, since the differential value is added to the previous sample, an initial value must be given at the start of differential decoding. Therefore, there are drawbacks such that the decoding can be started only when the initial value is obtained, and if an error occurs in the sample value during the decoding, all the decoded samples after that become errors (propagation of error). is there.

Therefore, a plurality of difference samples are collected to form a block, and the first sample of the block is not the difference sample but the sample value itself, which is used as the initial sample, and the subsequent samples may be used as the difference sample value. In this way, the decoding can arbitrarily select the beginning of each block as the decoding starting point, and even if an error occurs in the difference sample, the error propagation can be stopped within the block. For signals differentially encoded in this way,
Each of the lossless compression methods described above can be applied.

FIG. 13 shows a case where block scaling is performed on a differentially encoded signal. A sample that is not a difference is placed at the beginning of the block as a block initial sample. A difference sample is arranged after the second sample, a block scale is obtained for the difference sample, and scaling is performed.

Furthermore, it is obvious that sub-block scaling, sample scaling, and differential scaling can be applied to block scaling to achieve more efficient compression.

In the above embodiments, the division into the low band and the high band by the sub-band is explained with the bandwidth of 1: 1.
For example, it is obvious that even a division such as 1: 3 has no change. Also, after dividing into 1: 1
It is obvious that the effect of the present invention can be exerted even if the low frequency band is further divided into 1: 1 and the lowest frequency band is left uncompressed and the remaining two bands are losslessly compressed.

FIG. 14 shows still another embodiment of the present invention. In the case of this embodiment, in the low-frequency signal system of the subband analysis filter 11 and the formatter 13,
Further, a subband analysis filter 15 and a lossless encoder 16 are added. Subband analysis filter 15
Divides a low frequency signal of 0 to 24 KHz and a high frequency signal of 24 to 48 KHz. Corresponding to this encoder, the decoder has a configuration in which a lossless decoder 25 and a subband synthesis filter 26 are added to the output side of the deformatter 21.

As described above, according to the present invention, it is possible to perform mutual compatibility between a wideband signal transmission / recording system and a narrower band signal transmission / recording system, and to reduce the quality of signal contents. Without, the data capacity can be compressed. According to the present invention, it is possible to make a reproducing decoder according to the required quality and cost while using the same transmission medium and recording medium, and the user can select.

Next, an optical disc system to which the present invention is applied will be briefly described. FIG. 15 shows an optical disk reproducing apparatus, FIG. 16 shows a basic configuration of a disk drive unit 30 for driving the optical disk 100 on which the above audio stream is recorded, and FIG.
A diagram for explaining a configuration example of the optical disc 100 is shown in FIG.

The optical disk reproducing apparatus of FIG. 15 will be described.
The optical disk reproducing device has a key operation / display unit 500. A monitor and a speaker are connected to the optical disk reproducing device. The pickup data read from the optical disc 100 is sent to the system processing unit 504 via the disc drive unit 501. The pickup data read from the optical disc 100 is, for example, video data,
It includes sub-picture data and audio data, and these data are separated by the system processing unit 504. The separated video data is supplied to the video decoder 508 via the video buffer 506, the sub video data is supplied to the sub video decoder 509 via the sub video buffer 507, and the audio data is audio decoder via the audio buffer 512. 513 is supplied. The video signal decoded by the video decoder 508 and the sub video signal decoded by the sub video decoder 509 are combined by the combining unit 510 and output as an analog video signal by the D / A converter 511 and supplied to the monitor. The audio signal decoded by the audio decoder 513 becomes an analog audio signal by the D / A converter 514 and is supplied to the speaker.

Reference numeral 502 denotes a system CPU, and the entire reproducing apparatus is managed by this system CPU 502. Therefore, the system CPU 502 can exchange control signals, timing signals, and the like with the disk drive unit 501, the system processing unit 504, and the key operation / display unit 500. A system ROM / RAM 503 is connected to the system CPU 502. The system ROM / RAM 503 stores a fixed program for the system CPU 502 to perform data processing, and stores management data reproduced from the optical disc 100. It can also be stored.

The data RAM 505 is used by the system processor 5
04, and is used as a buffer when performing the above-described data separation, error correction, and the like. The disk drive unit 501 of FIG. 16 will be described.

The disk motor drive circuit 531 rotationally drives the spindle motor 532. Spindle motor 5
When 32 rotates, the optical disk 10 rotates, and the optical head unit 533 can pick up recorded data recorded on the optical disk. Optical head part 5
The signal read by 33 is supplied to the head amplifier 534, and the output of the head amplifier 534 is input to the system processing unit 504.

The feed motor 535 is driven by the feed motor drive circuit 536. Feed motor 53
5 drives the optical head unit 533 in the radial direction of the optical disk 10. The optical head unit 533 is provided with a focus mechanism and a tracking mechanism, and these mechanisms include a focus circuit 537 and a tracking circuit 53, respectively.
8 is provided.

Control signals are input from the servo processing unit 539 to the disk motor drive circuit 531, the feed motor drive circuit 536, the focus circuit 537, and the tracking circuit 538. Accordingly, the disc motor 532 controls the rotation of the optical disc 100 so that the frequency of the pickup signal is a predetermined frequency, and the focus circuit 537 causes the optical beam of the optical head unit 533 to have the best focus on the optical disc 100. To control the focus mechanism of the optical system, and to control the tracking circuit 53
Reference numeral 8 controls the tracking mechanism so that the optical beam is applied to the center of a desired recording track.

The structure of the optical disc 100 shown in FIG. 17 will be described. The optical disc 100 has information recording areas 102 around the clamp areas 101 on both sides thereof. The information recording area 102 has a lead-out area 103 in which information is not recorded on the outer periphery, and the clamp area 101
Lead-in area 10 in which no information is recorded at the boundary
4 A data recording area 105 is between the lead-out area 103 and the lead-in area 104.

Tracks are continuously formed in a spiral shape in the data recording area 105. This track is divided into a plurality of physical sectors, and the sectors are numbered consecutively. The signal trace of the track is formed as a pit. In a read-only optical disc, a pit row is formed by a stamper on a transparent substrate, and a reflective film is formed on the pit row forming surface to form a recording layer. The two-disk bonded type optical disk is a composite disk in which two disks are combined with an adhesive layer so that the recording layers face each other.

Next, the logical format of the above optical disc 100 will be described. FIG. 18 shows a logical format which is an information division of the information recording area 105. This logical format is defined in accordance with a specific standard such as Micro UDF and ISO9660. In the following description, the logical address is the micro UDF.
And the logical sector number (LS defined by ISO 9660)
N) means that one logical sector, which is the same size as the preceding physical sector, has 2048 bytes, and the logical sector number (LSN) is given in ascending order of physical sector numbers and consecutive numbers. And

The logical format has a hierarchical structure and includes a volume and file structure area, a video manager,
It has at least one or more video title sets and other recording areas. These areas are partitioned on logical sector boundaries. One logical sector is 2048 bytes. One logical block is also 2048 bytes, and one logical sector is defined as one logical block.

The file structure area corresponds to the management area defined by the micro UDF and ISO9660, and the data of the video manager is stored in the system ROM / RAM section 52 through the description of this area. The video manager describes information for managing a video title set, and is composed of a plurality of files starting from file # 0. In the video title set, compressed video data, sub-picture data, audio data, and reproduction control information for reproducing these are recorded.

In the other recording area, information used when using the information of the video title set or information used independently is recorded. The video manager will be described with reference to FIG.

The video manager is a video manager information (VMGI), a video object set for the video manager information menu (VMGM). VOB
S) 76 and backup of video manager information (VMGI BUP).

VMGM The VOBS stores video data, audio data, and sub-picture data for a menu relating to the volume of the optical disc managed by the video manager. That is, it is possible to obtain the description information by the audio and the sub-picture for each title in the volume and the selective display of the title. For example, if the optical disc contains English conversation for language learning, the title of the English conversation and the lesson example will be played back and displayed, the theme song will be played back as audio, and the level of the teaching material in the sub-video will be displayed. Is displayed. As a selection item, selection of a lesson number (level) is displayed, and the operation of the viewer is awaited. VMG for such use
M VOBS is used.

FIG. 20 shows a video object set (V
OBS) is shown. As the video object set (VOBS), there are two types for the menu and one type for the title for the video, but both have the same structure.

Video Object Set (VOBS)
Is defined as a set of one or more video objects (VOBs), and VOBs are used for the same purpose. Usually a video object set (VOB) for a menu
S) is configured as a video object (VOB) for displaying a plurality of menu screens, and a video object set (VOBS) for a video title set is
Video object (V
OB).

The video object (VOB) has an identification number (VOB). IDN # j), and the identification number (VOB A video object (VOB) can be specified using IDN # j). One video object (VOB) is composed of one or more cells. Similarly, the identification number (C IDN #
j) is attached, and this identification number (C IDN # j)
Can be used to specify the cell. A video object for a menu may be composed of one cell.

Further, one cell is composed of one or a plurality of video object units (VOBU).
And one video object unit (VOBU)
Is defined as a pack sequence having one navigation pack (NV pack) at the top. One video object unit (VOBU) is defined as a set of all packs recorded from the NV pack to immediately before the next NV pack.

Video Object Unit (VOBU)
Playback time corresponds to the playback time of video data composed of a single or multiple GOPs (groups of pictures) contained in this VOBU. The playback time is approximately 0.4 seconds or more and within 1 second. It is set. According to the MPEG standard, one GOP compresses image data corresponding to a reproduction time of about 0.5 seconds. Therefore,
According to the MPEG standard, information for about 0.5 seconds is arranged for both audio and video.

One video object unit (VO
BU), the video pack (V pack), the sub-picture pack (SP pack),
Audio packs (A packs) are arranged. Therefore, a plurality of V packs in one VOBU have one GOP or a plurality of GO packs of compressed image data whose playback time is within one second.
The audio signal corresponding to the reproduction time is compressed and arranged as an A pack. Also, the sub-picture data used during this playback time is compressed and arranged as an SP pack. However, the audio signal is recorded by packing data of, for example, eight streams and sub-pictures of, for example, 32 streams. One stream of an audio signal is data encoded in one type of encoding format, for example, linear P
It is composed of 8 channels of CM and 20-bit quantized data.

Returning to FIG. 19, description will be made. As the volume manager information (VMGI), information for searching a video title is described, and at least three tables are included.

Volume management information management table (VMGI
MAT is the size of the video manager (VMG), the start address of each information in the video manager, and the video object set (VMGM) for the video manager menu. VOBS) is described.

Title search pointer table (TT
SRPT) describes an entry program chain (EPGC) of a video title included in the volume of the optical disc, which can be selected according to the key operation of the apparatus and the input of the title number from the display unit 500.

The program chain will be described with reference to FIG. A program chain is a set of programs for reproducing a story of a certain title, and a story chapter or story of a certain title is completed by continuously reproducing the program. Further, one program is composed of a plurality of cells. The cell identification number can identify the cell in the VOBS.

Video title set attribute table (VT
S ART) describes the attribute information defined in the video title set (VTS) in the volume of the optical disc. As the attribute information, the number of VTSs, the number, the video compression method, the audio coding mode, the sub-picture display type, etc. are described in this table.

In the apparatus for performing the reproduction process as described above, the measures as described in FIGS. 1 to 14 are taken for the audio, and the audio processing system is the A pack (shown in FIG. 20). The desired data will be reproduced. Here, in the new playback device,
The ID is identified and both the low frequency band and the high frequency band are reproduced. However, if the high frequency signal cannot be processed, the A pack is not recognized and only the low frequency signal is reproduced. Even if only a low frequency signal is reproduced, there is a band of 0 to 48 KHz, and high quality can be obtained as a normal audio reproduction.

[0095]

As described above, the present invention provides a lossless data compression method and apparatus and a decoding method and apparatus thereof. Further, the present invention provides a method and apparatus for transmitting and recording a wideband digital signal while maintaining compatibility with a format for handling a digital signal having a narrow band by using the lossless compression and decoding method. Further, the present invention provides a reproducing method and a reproducing apparatus for receiving and reproducing the signal transmitted as described above or the recorded signal. Furthermore, the present invention provides a recording medium on which a signal formatted as described above is recorded and the recording medium.

[Brief description of drawings]

FIG. 1 is a block diagram for explaining the basic configuration of the present invention.

FIG. 2 is a frequency spectrum diagram of a signal for explaining the operation of the present invention.

FIG. 3 is a diagram for explaining the operation of the first embodiment of the present invention.

FIG. 4 is a flowchart for explaining the operation of the first embodiment of the present invention.

FIG. 5 is a diagram for explaining the operation of another embodiment of the present invention and a flowchart for explaining the encoding operation.

FIG. 6 is a diagram for explaining the operation of another embodiment of the present invention.

FIG. 7 is a diagram for explaining the operation of another embodiment of the present invention.

FIG. 8 is a diagram for explaining the operation of still another embodiment of the present invention.

FIG. 9 is a diagram for explaining the operation of another embodiment of the present invention.

FIG. 10 is a flowchart for explaining the operation of the embodiment of FIGS. 8 and 9.

FIG. 11 is a block diagram for explaining the basic configuration of another embodiment of the present invention.

12 is a diagram showing the differential encoder and differential decoder of FIG. 11. FIG.

FIG. 13 is a view for explaining the operation of still another embodiment of the present invention.

FIG. 14 is a block diagram for explaining the basic configuration of still another embodiment of the present invention.

FIG. 15 is a block configuration diagram of a disc reproducing device.

FIG. 16 is an explanatory diagram of a disc drive unit.

FIG. 17 is an explanatory diagram of an optical disc.

FIG. 18 is an explanatory diagram showing a logical format of an optical disc.

19 is an explanatory diagram of the video manager of FIG.

20 is an explanatory diagram of the video object shown in FIG.

FIG. 21 is an explanatory diagram of a program chain.

[Explanation of symbols]

 11 ... Subband analysis filter 12 ... Lossless encoder 13 ... Formatter 21 ... Deformatter 22 ... Lossless decoder 23 ... Subband synthesis filter.

Claims (8)

[Claims] 1. A block is constituted by a plurality of data samples of a digital signal, and all the data samples in the block have a constant value from the sign bit side thereof in accordance with a common signal level of the data samples in the block. A data compression method, wherein the number of bits is reduced to form a compressed block, and information on the reduced number of bits is added to the compressed block. 2. The compressed block is divided into smaller sub-blocks, and all the data samples in the sub-block are further divided from the code bit side in accordance with the signal level of the data samples in the sub-block that can be shared for each sub-block. Reduce a certain number of bits to make a compressed sub-block,
The data compression method according to claim 1, wherein the information on the number of reduced bits is added to either or both of the compressed block and the compressed sub-block. 3. A block is composed of a plurality of data samples of a digital signal, and all the data samples in the block are constant from the sign bit side thereof in accordance with the signal level that can be shared by the data samples in the block. The number of bits is reduced to form a compressed block, and the information on the reduced number of bits captures the encoded data added to the compressed block, and the information on the reduced number of bits is used to decode the reduced bits. Then, the data decoding method is characterized in that it is added to each data sample of the compressed block. 4. The compressed block is divided into smaller sub-blocks, and all the data samples in the sub-blocks are further divided into code bit sides according to the signal level of the data samples in the sub-blocks that can be shared for each sub-block. It is a compressed sub-block by reducing a more constant number of bits, the information of the reduced bit number is added to either or both of the compressed block or the compressed sub-block, using the information of the reduced bit number, 2. The data decoding method according to claim 1, wherein the reduced bits are decoded and added to each data sample of the compressed sub-block. 5. A subband analysis filter for dividing an input digital signal into a low-frequency signal and a high-frequency signal, the low-frequency signal is transmitted or recorded as it is, and the high-frequency signal is taken as a difference from an adjacent data sample. The data is compressed and transmitted, the transmitted signal is received, the low-frequency signal can be reproduced as it is on the one hand, and the high-frequency signal is compressed and decoded on the other hand, and the decoded signal and the low-frequency signal are sub-band combined. A data transmission / reception apparatus having lossless compression reproduction system means for synthesizing by a filter to generate a reproduction signal having an original band. 6. The data transmission / reception apparatus according to claim 4, further comprising means for inverting the sign of every other data sample of the high frequency signal before taking the difference. 7. The high-frequency signal constitutes a block including a plurality of data samples, and all the data samples in the block are reduced by a certain number of bits from the sign bit side according to the signal level of the data samples in the block. Then, the block compression method for adding the reduced number of bits to the block is used, and the block-compressed signal is further divided into smaller sub-blocks, and each sub-block is divided into sub-blocks according to the signal level in the sub-blocks. 5. The data according to claim 4, further comprising means for further compressing all the data samples of 4) by a fixed number of bits from the code bit side and compressing them by a block sub-block compression method for adding the reduced number of bits. Transmission / reception device. 8. A subband analysis filter that divides an input digital signal into a low-frequency signal and a high-frequency signal, the low-frequency signal is transmitted or recorded as it is, and the high-frequency signal is taken as a difference from a pre-adjacent data sample, The data is compressed and recorded, and the high frequency signal constitutes a block including a plurality of data samples, and all the data samples in the block are constant from the sign bit side according to the signal level of the data samples in the block. The number of bits is reduced and the reduced number of bits is added to the block for compression by a block compression method. Furthermore, this block-compressed signal is divided into smaller sub-blocks, and each sub-block is divided according to the signal level in the sub-block. , All data samples in the sub-block are further reduced by a certain number of bits from the sign bit side, Recording medium, characterized in that it is compressed in the block sub-block compression scheme adds a few are recorded.