JPH0442873B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a recording and reproducing method for a video format signal, and more particularly to a method for recording and reproducing image information and audio information as a video format signal on the same recording track of a recording medium.

When audio information corresponding to image information is recorded on the same track on a recording medium together with image information, the audio information is compressed in time and inserted into one part of the video format signal, and the image information is inserted into the remaining part. There is a way to do it. In such conventional methods, the insertion pattern of audio and image information on a video format in one recording medium is fixed, and it is therefore impossible to obtain various playback operations during playback. There is.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a video format signal recording and reproducing system that enables various reproducing operations when reproducing image information.

The video format signal recording and reproducing method according to the present invention divides a two-dimensional screen obtained by a signal corresponding to one field (one frame) of the video format signal into a plurality of blocks, inserts audio information into a predetermined block, and converts the image into the remaining blocks. In addition to inserting the information, control information is also inserted and recorded on the recording medium, and this control information corresponds to the image information by including at least address information indicating the address of the block where the audio information is inserted. The present invention is characterized in that the voice information can be inserted at a desired location.

Audio information is recorded with time axis compression, and when playing back, this time axis compressed reproduced audio information is extracted by expanding the time axis using a memory or the like so that it becomes real time. The signals thus reproduced become moving images of reproduced images and audio during trick play (steal, slow, step, etc.).

Audio information is digitized and recorded, but at this time, redundant bits are added for error correction so that errors can be detected and corrected within each block, so bit errors and burst errors are unavoidable. Efforts are being made to improve the reliability of information transmitted through recording media. In this way, by dividing audio information into blocks and enabling error detection and correction within each block, the length of audio information per field, that is, the real-time audio time, can be arbitrarily set for each block. Become. This will become clear in the explanation below, but on the playback screen that includes audio information, while playing the audio, the audio is played back sequentially at a time equal to the length of the audio information for each field (frame). This means that the speed at which the playback screen is sent when outputting data can be set arbitrarily. Furthermore, since the audio information can be inserted into any part of the screen, it is possible to selectively insert audio information into a less important part of the reproduced image, which is advantageous.

The reproduced image portion corresponding to the audio information insertion point is forcibly clamped to a certain level such as the pedestal level or average video level (APL) on the playback side, or controlled to be replaced with another image from a computer etc. In this case, an adverse visual effect on the reproduced image is prevented.

Furthermore, by providing a decoder for reading control information and audio information, it becomes possible to control the operation of a playback device (such as a video disc player) from this decoder. Therefore, there is an advantage that the operating procedure (software) for the reproducing device is stored in advance in the recording medium, which is equivalent to the method, and various reproducing operation controls can be performed without providing a special circuit on the reproducing device side.

The present invention will be explained below using the drawings.

FIG. 1 is a principle diagram for explaining the outline of the present invention, and shows a two-dimensional screen obtained by a signal corresponding to one field (one frame) of a video format signal at the time of recording. The upper edge, lower edge and both side edge parts are vertical and horizontal planking parts,
The part surrounded by the dashed line is the part that appears on the actual screen of the monitor television. In the two-dimensional screen excluding the vertical and horizontal blanking portions, the first number H (H means the horizontal scanning period) following the vertical blanking portion at the upper edge is used as a control signal insertion block. It is indicated by diagonal lines in the figure. The remaining part of the screen consists of m horizontally and n vertically.
Each block is divided into m×n blocks. The block location where the time-axis compressed audio information is inserted is specified by the control signal, and the audio information insertion position on the screen is determined. Conversely, it is of course equivalent to specifying the inserting position and size of the image information to determine the inserting block of the audio information.

FIG. 2 is a diagram showing an example of inserting audio information, in which audio information is inserted into the diagonally shaded block at the bottom right, and image information is inserted into the central area surrounded by this audio information insertion block. There will be. A part of the audio information insertion block appears on the actual monitor screen surrounded by a dashed line,
This portion is clamped to a certain level on the reproduction side or replaced with other image information from the outside.

Although FIG. 3 shows the video format signal waveform during recording in accordance with the NTSC system, other systems such as the PAL system are also possible. Figure A shows the waveform before voice information is inserted, and Figure B shows the waveform after voice information is inserted. a and b are test signals such as color bars, c is control information, and d is audio information, respectively. In the figure, vertical blanking (V-
Although neither control information nor audio information is inserted in the BLK, it goes without saying that they may be inserted. Among the control information, at least those related to the audio information insertion position and the image processing method must be inserted at a predetermined portion at a position preceding the corresponding image.
In some cases, audio information is recorded over several frames, in which case it is sufficient to insert the control information in a predetermined part of the first frame, but there is no problem in inserting it in other frames as well. do not have.

FIG. 4 is a block diagram of an encoder for obtaining a video format signal according to the recording method of the present invention. Analog audio signal is ADM
(Adaptive Delta Modulation) and ADPCM
The signal is digitized by an A/D (analog-digital) converter 1 using a modulation method with a high compression effect, such as (Adaptive Differential PCM). This digital signal is input to the error correction encoding circuit 2, where it is rearranged on the time axis, that is, interleaved, and then redundant bits are added so that an error correction code is completed for each block.
The purpose of interleaving is to disperse errors in digital signals on the time axis when signal dropouts occur due to scratches, dust, etc. on the video disk, which is a recording medium. The signal thus subjected to error correction encoding processing is written to the time axis compression memory 3 at a sampling frequency of ₁ . By reading data from the memory 3 at a frequency ₂ higher than frequency ₁ , the time axis is compressed.

Control signals, which are control information, can be classified into those that are extremely important to the system and require error detection and correction, and those that are not so important and do not require error detection and correction. The former control signal includes a block number indicating the audio information insertion block, the aforementioned sampling frequency designation, audio information channel number designation (monaural, stereo, etc.), VDP (video disk player) playback operation designation, There are designations for continuous and discontinuous reproduction of audio information, group numbers for dividing audio information insertion blocks into groups, and these important control signals are input to the error correction encoding circuit 4.

This control signal is also encoded using a block-based error correction coding system, like the audio information, but unlike the audio information, it is preferable that it be a powerful correction code. This is because control information is critical to system operation.

On the other hand, control signals such as flags indicating the presence or absence of audio information in individual fields (frames) and Phillips codes (this is a stop code, and when this code is played, VDP automatically switches to still playback) are less important. These signals are directly input to the switching circuit 5, and are selected and output by the switching circuit 5 together with the output of the time axis compression memory 3, the output of the error correction encoding circuit 4, and the image signal (including the synchronization signal). That will happen. The selection operation of the switching circuit 5 is controlled by a timing signal generation circuit 6, and each write and read control of the memory 3 is also controlled by the timing signal generation circuit 6.
In the timing signal generation circuit 6, an internal oscillator is synchronized with the synchronization signal of the input image signal, and various timing signals are generated in accordance with external control signals and the like. From the output of the switching circuit 5, a video format signal is obtained in which time-base compressed audio information is inserted into the desired block and line information is inserted into the remaining blocks.

FIG. 5 is a block diagram of a decoder for reproducing a video disc on which the video format signal thus obtained is recorded. Signal separation circuit 7
The audio information and control information separated from the reproduced signal are error corrected by error correction circuits 8 and 9, respectively. After that, the audio information is transferred to high frequency ₂
is written into the time axis expansion memory 10.
The readout is performed using a low-speed clock with a sampling frequency of ₁ , and the time axis is expanded to become real-time audio data. Thereafter, the D/A converter 11 demodulates the signal into the original analog signal. Note that the error correction may be configured to be performed together with the time axis extension when reading from the memory 10.

Both the error-corrected control signal and the control signal that does not require error correction are input to the system control circuit 12. This circuit 12 controls the entire decoder system and further controls the VDP using a control signal separated from the reproduction signal and an external control signal. Since the reading of audio information from the memory 10 is not necessarily performed immediately after writing,
It is necessary to store the control signals necessary for the readout once, so the memory 1 for control signals is
3 is provided for this purpose. The timing signal generation circuit 14 has an internal oscillator synchronized with the synchronization signal separated in the signal separation circuit 7.
Various timing signals are generated according to the control signal and supplied to each block within the decoder.

The block portion into which the audio information has been inserted is replaced by another video signal obtained from a computer or the like in the video synthesis (or masking) circuit 15, or output after being clamped to a certain level such as the pedestal level or APL. . The external audio signal is selected as the output audio signal when audio information is not being output or when the switch 16 is forcibly switched. This external audio signal is, for example, an FM-processed two-channel audio signal obtained within a VDP (not shown). Note that the system control circuit 12 generates a control signal for the VDP playback operation included in the played control signal, and outputs it to the VDP (not shown) via the player controller 17.

The part indicated by the dotted line in the figure is necessary when the decoder is equipped with a so-called LL (Language Laboratory) function, in which the learner's microphone input voice is digitized by the A/D converter 18 and the time axis is expanded. Write to memory 10. This memory 10 will take the place of a tape recorder, and if the output of this memory 10 is read out at readout frequency ₁ , the voice information that is considered to be the model pronunciation and the pronunciation input by one's own microphone will be combined. comparison becomes possible.

Sampling frequency ₁ for writing to memory 3 and reading from memory 10 in FIGS. 4 and 5
It is necessary to switch and use the information depending on the content of the audio information in order to efficiently utilize the information transmission medium.
Now, according to the sampling theorem, the sampling frequency needs to be more than twice the signal frequency band,
If the audio information is music, it has a wide frequency band and therefore requires a high sampling frequency. On the other hand, if it is a human voice, a low sampling frequency is sufficient. Therefore, in order to switch and use the sampling frequency according to the audio information, it is preferable to include the designation of the sampling frequency in the control information. Also, if it's music information, it's stereo (2 channels), and if it's a human voice, it's generally monaural (1 channel).
Similarly, it is also useful to include the designation of the number of channels in the control information. Therefore, it is necessary to specify the sampling frequency and the number of audio channels.
As mentioned above, it is included in the control information.

Figure 6 shows an example of processing of reproduced images, and A
is a reproduced image into which audio information has been inserted (indicated by diagonal lines), and this reproduced image is converted into an image such as B or C in the video synthesis (or masking) circuit 15 in the decoder shown in FIG. be. Diagram B
In Figure C, the audio information insertion block is replaced by an image from an external source such as a computer, and in Figure C, for example, the pedestal level is replaced. The method shown in Figure B is useful when used as an educational device, and allows interaction between the user and the computer. For example, a certain question appears in an image, the corresponding audio information is output, and then the image, i.e., the question, remains as is, and the answer is output as text or an image from the computer and inserted as shown in Figure B. be able to. It is also possible to display characters that instruct the next operation.

Image processing is easily and reliably performed by inserting a control signal (or flag) indicating insertion of audio information into every frame (field) in which audio information is inserted and referring to this control signal. be able to. That is, by detecting this control signal, it becomes possible to input necessary images from the outside into the audio information insertion section independently of the state of the video signal and the system operation.

FIG. 7 is a diagram showing the relationship between the audio information insertion mode and the reproduced image, with two cases A and B shown. First, referring to Figure A, when the audio information of each frame (field) is continuously played back while transmitting the playback image, the audio information (shown as SWS data, Still picture
WithSound) is inserted and recorded in each block at the top and bottom edges of each frame (which may also be a field). To reproduce such a signal, the SWS data in the first frame is written to the first area of the time axis expansion memory (memory 10 in FIG. 5, and the same shall apply hereinafter), and this writing is performed at the start of reproduction of the second frame. Read data while expanding the time axis. At this time, the SWS data in the second frame is simultaneously written to the second area of the time axis expansion memory,
Following the completion of reading the SWS data in the first frame, this written data is read out while expanding the time axis, and playback of the third frame is started, and this SWS data is read out while expanding the time axis.
Write data to the first area. These operations are repeated in sequence, and the audio information is output continuously on the real time axis, and the corresponding images are played back. In this case, the mode of the reproduced image is that only the first frame is a moving image, and from the third frame onwards, images are sent in a time equal to the length of each frame SWS data, so normal playback (moving image), depending on the length of the data, Slow, step (so-called frame-by-frame forwarding), etc. are possible. In this example, writing and reading are performed alternately in two areas of the time axis expansion memory 10, and there may be some overlap between the two areas, so the memory capacity should be less than or equal to two frames of SWS data. I can do it.

Referring to FIG. 7B, this is an example in which SWS data is inserted into all blocks of, for example, six preceding frames (fields may be used), and only image information is recorded in each of the remaining frames. To reproduce this signal, the SWS data of these six preceding frames are sequentially written to the time axis expansion memory while the VDP is set to normal reproduction, and the audio information is read from the memory in real time from the start of the reproduction of the 7th frame image. It is read out. In this case, the mode of the reproduced image can be moving image, slow motion, step, still, etc. In this example, since it is necessary to store all SWS data within six frames, a larger memory capacity is required compared to the memory capacity of the example shown in Figure A. On the other hand, in the example of FIG. A, if still playback is performed, the audio time per frame is short, so many identical screens must be sent in steps, which has the disadvantage that the number of still images that can be recorded is reduced. However, the example shown in FIG. B has the advantage that the number of still images that can be recorded increases at the cost of an increase in memory capacity. Furthermore, as shown in Figure B, all SWS
Even when data is stored in memory, the data may be inserted only at the upper and lower edges of the image as shown in Figure A.

In this way, various VDP operations are possible depending on the relationship between the SWS data insertion mode and the image. Such various operations, ie, FIG. 7A
In the operations of B and B, the control of writing and reading of memory is different, so it is necessary to appropriately switch and instruct these using control information. Also, in VDP, depending on the image content, slow, step, steal, etc. Since different operations may be required, all switching instructions for these operations are performed by signals included in the control information.

FIG. 8 is a diagram showing another relationship between the audio information insertion mode and the reproduced video signal, and is an example of the method shown in FIG. 7A. VDP is configured to advance frame by frame depending on how many frames of SWS data corresponds to one frame.
control. In the example shown in the figure, the frames are advanced every four frames. First, SWS data a is written to memory, and from the start of playback of the next frame, this data a is expanded on the time axis and read out.
SWS data b is written to memory. data a
After the reading of data b is completed, data b is similarly expanded on the time axis and read out, and the next data c is written into the memory portion that has been storing data a during this time.
In this way, data a, b, c, d, . . . are each written once. Since each of the data a to d corresponds to the length of an integer frame, if the capacity of the time axis expansion memory 10 is set to two frames of data, reading is continuous and continuous sound reproduction becomes possible.

The insertion position of each of the data a to d in each frame in the figure can be arbitrary, and continuous audio can be obtained by setting the reading start and end positions of each of the data a to d constant regardless of the data insertion position. Images during this time are still reproduced every four frames, resulting in a frame-by-frame reproduced image.

Here, it is impossible to distinguish between normal playback and frame-by-frame playback using only the playback signal from the VDP, so in the case of Figure 8, if the decoder determines that normal playback has continued, SWS data b, c, or d will be written to memory many times. In order to prevent such a situation, it is necessary to determine whether the contents of the SWS data are the same in the previous frame and the subsequent frame, and control information is used for this determination.

If there is a plurality of audio channels, multiplexing is performed within one block so that the number of SWS data for each channel is equivalent to an integer frame (field). If this is expressed as a formula, N _D = N _C・N _F・₁ / _F. N _D is the number of SWS sampling data in one block, N _C is the number of audio channels, N _F is the corresponding frame (number of fields) of SWS data in one block, ₁ is the sampling frequency, and _F is the frame (field) frequency. be.

FIG. 9 is a diagram showing still another relationship between the audio information insertion mode and the reproduced image, in which the audio information insertion blocks are further grouped and a group number is assigned to each group, and a series of significant numbers different for each group are set. This is audio information. for example,
As shown in the figure, the SWS data group inserted into all blocks of the first and second frames (fields) is set to group number 0, and Japanese audio is inserted into this group. All inserted SWS data groups of the third and fourth frames are set as group number 1, and English audio is inserted into this group. Similarly, the SWS data groups of the fifth and sixth frames are set to number 2 and are made into French audio. It is assumed that only image information is inserted into subsequent frames. These group numbers are included in one piece of control information, and this group number is designated by a command signal from an external switch, computer, etc., and selective reading is performed.

For example, the figure shows a state in which the French audio of group number 3 is specified and read out, and the first to sixth frames are sent in normal playback, but are written to the memory only after the data of group number 3 arrives. In rare cases, this is expanded on the time axis and read out from the seventh frame. Of course, data writing may be performed for all groups, and selective operation may be performed only during reading, but this is not preferable if the memory capacity is limited.

In addition to selecting this language, this example can also be used to ask a question and select and output different replies, comments, etc. depending on the user's response, making it ideal for educational purposes. This is particularly useful for interactive video systems combined with computers.
This becomes effective when the SWS system is installed.

FIG. 10 is a diagram illustrating the relationship between the insertion mode of audio information and control information and the playback audio output mode,
This is an application example where the group number of the SWS data shown in FIG. 9 is included in the control information. In figures A and B, S ₀ to _{S i} are SWS data group numbers,
C ₁ is a control signal indicating that selective writing to memory is possible, C ₂ is a control signal to perform continuous audio output operation, and C ₃ is a control signal to start reading from memory. The signals are shown respectively.

In the example shown in FIG. 7A, audio information of group number S ₁ of the SWS data is selectively written into the memory and operated as continuous audio output as shown in FIG. 7A. Initially, VDP performs normal playback operation, and then performs still playback for each group S _i that appears from the second time onward. By switching to normal playback when the readout content is switched, continuous audio output is possible while advancing frame by frame. There is. If the length of i+1 frames is to be accommodated even if one readout time elapses, it is sufficient to send further images and output continuous sound for the moving image in group S _i instead of performing still playback.

FIG. 10B shows the same operation as in FIG. 9, in which the audio information of group number S ₁ is selectively written into the memory, reading is started after writing is completed, and after this reading is completed, the next number S ₁ data is being written to memory. Note that the mark (↑) in FIG. B means that player control is not necessary when outputting audio to a moving image.

In both the examples in Figures A and B, the reading starts at the end of S _i no matter which one of S ₀ to S _i is selected, but reading may start immediately after writing is completed; in this case, In this case, control signal _C3 is not required. still,
The insertion block position of the SWS data in the video format signal is an arbitrary position specified by the control signal.

Selection of writing and reading of SWS data is possible using the control information as described above, but it is also possible using external control. In other words, it is possible to provide a flexible and expandable system that can be controlled by a computer. For example, if the group number is used to differentiate the time axis expansion memory area in which SWS data is written,
It is possible to change the group number of input SWS data according to predetermined conditions and write it in a different area so that previously written SWS data is not erased. By using control signals as described above, various advanced functions become possible.

The above-mentioned system is a versatile and multi-functional type, but as a result, it has a complicated configuration and is expensive, so in order to put it into practical use, it is necessary to have a somewhat limited function, a simplified configuration, and a lower cost. There may be cases where For example, if the number of audio channels and the sampling frequency are fixed, no control signals are needed for that purpose, and if the reading timing of SWS data from the time axis expansion memory is set to start immediately after writing, A control signal for starting reading is also not required. Furthermore, if the player's actions are also limited, it will be simplified.
For example, with an optical video disc player, you can use the Phillips code to play stills.
If it is determined that normal playback begins when the reading of SWS data from the memory ends, there is no need to add these control signals.

As an example, let's use the human voice as audio information.
Expressed as ADPCM digital signal, per second
If you want to send an image with 40K bits, that is, a video,
A transmission rate of about 666 bits per field is sufficient. This is recorded on two horizontal scanning lines that do not appear on the monitor television screen at or near vertical blanking. That is, as shown in FIG. 11, 333-bit data may be superimposed on 1H.
In Figure A, a is a test signal, b is audio information, and Figure B is an enlarged view thereof. If the audio channel output is 2 channels, it is sufficient to record in 4H. If music is also transmitted as audio information, the transmission rate must be increased, so the number of Hs on which data is superimposed is increased. In addition, when adding redundant bits as an error correction code, H
Either increase the number of bits in 1H or increase the number of bits in 1H. In this way, audio information can be added to a moving image without substantially reducing images.

An example in which the capacity of the time axis expansion memory can be significantly reduced in the case of continuous audio output will be described with reference to FIG. It is assumed that the amount of audio information inserted into one frame (field) is constant and corresponds to the image information of multiple frames (fields), and the audio information insertion position within the frame is constant. Currently, the writing speed to memory is faster than the reading speed, so if reading is started immediately after writing starts as shown in the figure, the information that has been written will always be read. Each audio information indicated by a, b, c, d... is equivalent to 3 frames, so
The start time of each reading of information a, b, c, d, etc. is always constant within each frame. Therefore, by setting the end position of each readout immediately before writing the last data of the next audio information, and reading at the same timing at the switching point of audio information, continuous audio output can be achieved, and at the same time, the memory capacity can be reduced to a, b, The amount of information read out until the end of writing of each audio information is subtracted from the amount of audio information of one audio information c, d, . . . , resulting in an extremely small capacity. In reality, it is necessary to provide a margin for jitter, etc. in the reproduced signal, and in order to facilitate control, the capacity may be set to be the same as each piece of audio information in one frame. The number of frames to which each piece of audio information corresponds is arbitrary as long as it is an integer value, and one frame constitutes a moving image.

Next, the error correction method for control information will be described.
The control information is divided into blocks of a certain length as digital data. Redundant bits of the BCH code are added to each block so that bit errors can be corrected. Then, FM (Frequency Modulation) is applied to each block to which redundant bits are added.
Or it is modulated by PM (Phase Modulation) method. A PM modulation system is shown in FIG. 13A, in which the signal is modulated with such regularity that it rises at the center of the bit cell when the digital data is "1" and falls when it is "0". So-called burst errors (continuous errors caused by dropouts, etc.) that cannot be corrected only by the redundant bits of the BCH code added to each block are detected in Figure 13B.
This burst error may occur as shown in the figure below, but this burst error is detected when the regularity of inversion in the PM method is broken, and is then used as the error pointer for that block.

On the other hand, parity C obtained using two consecutive blocks A and B (see FIG. 14) among the blocks to which redundant bits have been added is added as a redundant bit. This parity C is obtained as C _i =A _i +B _i (i=0, 1, . . . , n), where + indicates modulo-2 addition. The burst error is corrected using this parity C and an error pointer indicating the occurrence of the previous burst error. An error pointer is generated when the number of bits in one block becomes larger than the number of bits that can be corrected by the BCH code by counting cases where there is no inversion of rising or falling edges at the center of each bit cell.

Figure 15 shows the PM shown in Figures 13 and 14.
2 is a block diagram for modulation and error correction encoding, where input data A and B are input to a selector 20 and a parity generation circuit 21;
Then, a parity code of C=A+B is created. The selector 20 outputs the input data A and B as they are, and subsequently outputs the parity C. The output of the selector 20 is input to the next stage selector 22 and BCH code redundant bit generation circuit 23, and the latter circuit 23 generates redundant bits for each of A, B, and C. The selector 22 outputs data A followed by redundant bits A, data B followed by redundant bits B, and parity C followed by redundant bits C. This output should look like the one shown in Figure 14.
It is modulated by the PM modulator 24. In addition to using a parity code, a b-adjacent code or the like may be used to correct burst errors.

FIG. 16 is a block diagram for decoding and error correction of error correction encoded and PM modulated data. The input data is demodulated by a PM demodulator 25, and at the same time, a burst error detector 26 checks each block A, B, and C for burst errors. Bit errors in the demodulated data are corrected by a corrector 27 for each block A, B, and C using the BCH code. The corrected data is input to the delay circuit 28 and the syndrome generation circuit 29. The syndrome generation circuit 29 generates syndrome S=A+B+C. Delay circuit 28
delays the data so that the first data A is input to the selector 30 when the generation of the syndrome S is completed.

The adder circuit 31 adds the syndrome S and each data modulo 2, and if block B includes a burst error and becomes B'=B+e, then S+S'=B+e+A+B+e+C =B (∵B=A+C), so the addition The output is burst error corrected data.

If either data A or B has a burst error and parity C is correct, selector 30 selects the output of addition circuit 31 when the incorrect data is input, and always outputs correct data. That's what I do. The selector 30 is controlled by the output of the burst error detector 26. If a burst error occurs in two or more of blocks A, B, and C, it cannot be corrected, so an uncorrectable flag is output and appropriate processing is performed in the next stage circuit. Since parity C is a correction code and does not need to be output, error correction of parity C itself is not performed.

The burst error detector 26 is composed of gates, counters, latches, etc., and the BCH corrector 27
The syndrome generation circuit can be constructed from gates, shift registers, exclusive OR circuits, ROM (read only memory), etc., and the delay circuit can be constructed from shift registers, latches, etc.

Next, an error correction method for audio information will be described. As an example, in Figure 1, the division of the screen is m=
1, n=14. W _i
(i=0 to 1049), and each sample W _i consists of 4 bits. and the subscript i of W
indicates the order of sampling, and is considered as a three-dimensional array configuration as shown in FIG. In the figure, the numbers attached to the solid line arrows indicate the sampling order, and the numbers attached to the dotted line arrows indicate the order of insertion into the video format signal.

That is, after sampling 15 samples from W ₀ at the upper left corner of the three-dimensional array configuration to W ₁₄ at the bottom, redundant parity P _0,0 and Q _0,0 of b adjacent codes are added for error correction. . Then W ₁₅ to W ₂₉
After up to 15 samples, P _0,15 and Q _0,15 are added in the same way. P _0,90 and Q _0,90 are added after the 15 samples from W ₉₀ to W ₁₀₄ using a similar procedure.
After that, from W ₁₀₅ downwards (not shown)
P _0,105 and Q _0,105 are added after the samples up to W ₁₁₉ . In this way, P _0,1035 and Q _0,1035 are added after the samples from W ₁₀₃₅ to W ₁₀₄₉ . These added P _0,j , Q _0,j (j=0, 15, 30, 45,...,
1035) will be used as parity in the vertical direction (in the direction of the solid arrow) of the three-dimensional array configuration.

Next, P _1,k and Q _1,k are created as horizontal parity (in the direction of the solid arrow) of the three-dimensional array configuration, but for example, P _1,0 and Q _1,0 are W ₀ , W ₁₀₅ ,
It is the parity for 10 samples of W ₂₁₀ , ..., W ₉₄₅ and is added after the sequence of these 10 samples. Similarly, the parities of P _1,90 and Q _1,90 are added after the sequence of 10 samples W ₉₀ to W ₁₀₃₅ .

One block of audio information including the error correction code thus obtained is interleaved, rearranged, and inserted over 17H within one field of the video format signal. FIG. 18 is a diagram showing the arrangement of samples on the first H into which interleaved data is inserted. A data sync is superimposed on the leading part following the color burst signal, and immediately after that, 12 sets of 7 samples each are sequentially arranged. In this case, the arrangement method (interleaving) is performed according to the numbers indicated by the dotted line arrows in FIG.

By doing this, each sample W ₀ , W ₁₀₅ , W ₂₁₀ , ..., W ₉₄₅ on the same −H has parity P _1,0 , Q _1,0
This is an error correction code. In general, parity P _1,k ,
Q _1,k (k=0,1,…,14,P _0,0 ,Q _0,0 ,15,…,
29, P _0,15 , Q _0,15 , 30,..., 104, P _0,90 , Q _0,90 ) are error correction codes for 10 samples distributed on the same H. .

During reproduction, in addition to correcting errors in the 10 sample values distributed on the same H using P _1,i and Q _1,i obtained in this way, errors in the 10 sample values distributed among the H are corrected. For example, error correction for 15 samples W ₀ to W ₁₄ is performed using P _0,0 and Q _0,0 .

The error correction code is assumed to be a b-adjacent code, which corrects adjacent errors of b=4 bits. That is, 1
Assume that an error in one sample consisting of 4 bits included in one parity check is to be corrected. Parity P _0,j and Q _0,j (j=0, 15,..., 1035) are generated as follows.

P _0,0 =W ₀ +W ₁ +…+W ₁₄ P _0,15 =W ₁₅ +W ₁₆ +…+W ₂₉ 〓 P _0,1035 =W ₁₀₃₅ +W ₁₀₃₆ +…+W ₁₀₄₉ Q _0,0 =T ¹⁴・W ₀ +T ¹³・W ₁ +…+T・W ₁₃ +W ₁₄ Q _0,15 =T ¹⁴・W ₁₅ +T ¹³・W ₁₆ +…+T・W ₂₈ +W ₂₉ 〓 Q _0,1035 =T ¹⁴・W ₁₀₃₅ +T ¹³・W ₁₀₃₆ +...+T ・W ₁₀₄₈ +W ₁₀₄₉ However, P, Q, and W are 4-bit column vectors, and T is expressed by the following formula. + means modulo 2 addition, and the same applies hereafter.

T=0001 1001 0100 0010 Now, if there is one error among W ₀ to _{W 14} , it will be corrected by P _0,0 and Q _0,0 .
For example, suppose that W _l mistakenly becomes W′ _l = el (el
are error patterns) and the syndromes S _0,P , S _0,Q
is the following formula.

S _0,P =P _0,0 +W ₀ +W ₁ +…+W′ _l +…+W ₁₄ =el S _0,Q ≡Q _0,0 +T ¹⁴・W ₀ +T ¹³・W ₁ +…+T ^14-l・
W′ _l +…+W ₁₄ =T ^14-l・el T ^l+1・S _0,Q =T ¹⁵・el=elS _0,P (∵T ¹⁵ =〓), so S _0,Q =S Multiply S _0,Q by T until it becomes 0 _,P , and find the erroneous sample position l from the number of times l+1 of this multiplication by T. When l is determined, correct W′ _l as follows.

W′ _l +S _0,P = W _l +el+el=W _l Syndrome S _0,P , S _0,Q is parity check matrix H
Expressed using , it becomes as follows.

〓＝S _O,P S _O,Q ＝〓・〓＝〓〓…〓〓〓0 T ¹⁴ T ¹³ …T〓0〓W ₀ W ₁ 〓 W ₁₄ P _0.0 Q _0,0However , 〓 is 4× Let it be the identity matrix of 4.

Similarly for parity P _1,k and Q _1,k , P _1,0 =W ₀ +W ₁₀₅ +…+W ₉₄₅ P _1,1 =W ₁ +W ₁₀₆ +…+W ₉₄₆ 〓 P _1,14 =W ₁₄ +W ₁₁₉ +…+W ₉₅₉ P _1,P,00 =P _0,0 +P _0,105 +…+P _0,945 P _1,Q0,0 =Q _0,0 +Q _0,105 +…+Q _0,945 〓 Q _1,0 =T ⁹・W ₀ +T ⁸・W ₁₀₅ +…+T・W ₈₄₀ +W ₉₄₅ Q _1,1 =T ⁹・W ₁ +T ⁸・W ₁₀₆ +…+T・W ₈₄₁ +W ₉₄₆ 〓 Q _1,14 =T ⁹・W ₁₄ +T ⁸・W ₁₁₉ ＋…＋T・W ₈₅₄ ＋W ₉₅₉ Q _1,900 ＝T ⁹・P _0,0 ＋T ⁸・P _0,105 ＋… ＋T・P _0,840 ＋P _0,945 Q _1,Q00 ＝T ⁹・Q _0,0 ＋T ⁸・Q _0,105 +… +T・Q _0,840 +Q _0,945 〓 If the syndrome is expressed as a parity check matrix, 〓=S _1,P S _1,Q =〓・〓=〓〓…〓〓〓0 T ⁹ T ⁸ …T〓0〓W ₀ W ₁₀₅ 〓 W ₉₄₅ P _1,0 Q _1,0 .

Now, if W _n is mistakenly set as W′ _n ==W _n +e _n , then S _1,P = e _n , S _1,Q = T ^9-m・e _n T ^m+6・S _1,Q =T ¹⁵ ·e _n =e _n =S _1,P Find m from W′ _n +S _1,P =W _n +e _n +e _n =W _n to perform correction. As a correction format, first _{W i and} P _{0 ,j} ,
Correct Q _0,j and then correct W _i again using P _0,j and Q _0,j . Other correction formats include P _1,k ,
Detect the error from Q _1,k and find the error position,
It is also possible to correct errors in two samples by P _0,j and Q _0,j .

FIG. 19 is a block diagram for interleaving and error correction encoding of the sampling data shown in FIG. 17. Redundant bit parity P _0j for correction between different H when data is input
and Q _0j and redundant bit parities P _1,k and Q _1,k for error correction within the same H are generated by each generation circuit 40 to
43, respectively. P _0,j and Q _0,j are generated every 15 samples of data, while P _1,k and
Since Q _1,k is generated for data in the same H after interleaving, it is necessary to temporarily store the values of P _1,k and Q _1,k during calculation, and the memory 44 and 45 are provided. In other words,
P _1,0 , Q _1,0 are generated for W ₀ , W ₁₀₅ , ..., W ₉₄₅ , but in the middle of that, input samples, e.g.
It is necessary to proceed with the calculations for W ₁ , W ₁₀₆ , ... in the same way, so when the data is input, the corresponding intermediate values of P _1,k , Q _1,k are stored in the memory 44, 45. After temporarily storing the data in the memory 44 and 45, the data is read out to proceed with the calculation, and then written into the memories 44 and 45 again. This is repeated until the generation of P _1,k and Q _1,k is completed.

Input data and redundant bit parity P _0,j , Q _0,j ,
P _1,k and Q _1,k are each input to the selector 46, selected according to the timing at which each is generated, and written to the interleave memory 47. Interleaving is performed by changing the order of writing and reading samples in this memory 47. This memory can also be used when performing time axis compression. The memory control circuit 48 reads/
The write control signal and timing clock generate various timing signals necessary for the memory and also perform address control.

Each generation circuit 40 to 43 is composed of a shift register and an exclusive OR circuit, and the memory is composed of a register or a RAM.

FIG. 20 is a block diagram for deinterleaving and error correction of data obtained by the circuit of FIG. 19. Reproduction input data is sent to delay circuit 4
9, and are input to circuits 50 and 51 that generate syndromes of S _1,P and S _1,Q . S _1,P and S _1,Q are the same H
This is a syndrome for error correction in Since the data within the same H is interleaved, it is necessary to temporarily store the values in the middle of the calculation, similar to the generation of P _1,k and Q _1,k in FIG.
2,53 is used. Syndrome S _1,P ,S _1,Q
The error position detection circuit 54 determines the position of the sample with more errors. Furthermore, since the syndrome S _1,P is a parity sum, the erroneous sample can be corrected by performing modulo 2 addition in the adder circuit 55 on the erroneous sample. The timing clock 1 is synchronized with data samples and counted in a circuit 54 for determining the error position, and sends an addition command to the addition circuit 55 in synchronization with the timing at which an erroneous sample is input to the addition circuit 55. be.

The delay circuit 49 receives the first data from the adder circuit 55.
Delay the data so that the generation of that syndrome is completed before it is input into the .

The data on which errors have been corrected within the same H are sent to a deinterleave memory 56, where the order of writing and reading is changed and deinterleaving is performed. Further, time axis expansion may be performed using this memory. A memory control circuit 57 controls the operation of the memory 56. The day interleave memory 56 first stores the syndrome S _0,P ,
A readout is performed to generate S _0,Q , and then a readout is performed to correct and output each sample. For example, when reading W ₀ , W ₁ , ..., W ₁₄ , first read W ₀ , W ₁ , ... W ₁₄ , P _0,0 , Q _0,0 , and the syndrome generation circuits 58 and 59 convert S _{0 ,P} ,S _0,Q are generated. At this time, no input is made to the adder circuit 55. Next, W ₀ , W ₁ , ..., W ₁₄ are read again. At this time,
If there is an error in W ₀ , W ₁ ,...W ₁₄ , the syndrome S _0,P is corrected in the same way as error correction in H.
, S _0,Q , and the timing clock 3, an addition command is output from the error position detection circuit 60. In the adder circuit 61, erroneous samples and syndromes
Correction is performed by adding modulo 2 to S _0,P .
These circuits 58 to 61 perform error correction between different H's.

The error correction circuits before and after the de-interleave memory 56, that is, the error correction circuits within the same H and the error correction circuits between different H, essentially perform the same operation, so if there is margin in the operation timing, the circuits can be It is possible to perform time-division operation using a common method. The illustrated circuit example is one example, and other configurations are also possible.

For example, error correction within 1H is connected to the output of the deinterleave memory 56 in parallel with other error correction circuits. In this way, the syndrome S _1,P ,
Since the generation and addition of S _1,Q are also performed by reading out the deinterleave memory, the memories 52 and 53 and the delay circuit 49 are not required.

As described above, according to the present invention, audio information in the form of a block is inserted at an arbitrary location on a screen generated by a video format signal, and control information including address information indicating the address of the inserted block of audio information is inserted. By inserting the same, it becomes possible to record the procedure for playback operation, that is, the software, on the recording medium itself, and this software is decoded by a decoder (Figure 5) installed separately in addition to the VDP and transferred to the VDP. Because it instructs
This has the advantage that various playback operation controls can be controlled without adding a special circuit to the VDP side.

Furthermore, since error correction can be performed independently within the insertion block of audio information and control information, the reliability of reproduction information and reproduction operation is significantly improved.

[Brief explanation of drawings]

Fig. 1 is a diagram explaining the basic principle of the present invention, Fig. 2 is a diagram showing a playback screen pattern of an embodiment of the invention, and Fig. 3 is a diagram showing a video format signal waveform of an embodiment of the invention. 4 is a block diagram of an encoder for implementing the recording method of the present invention, FIG. 5 is a block diagram of a decoder for implementing the reproducing method of the present invention, and FIG. 6 is an embodiment of the present invention. 7 to 9 are diagrams illustrating the reproduction operation according to each embodiment of the present invention, and FIG. 10 is a diagram illustrating the relationship between SWS data, control information, and reproduction operation. FIG. 11 is a signal waveform diagram showing another embodiment of the present invention, FIG. 12 is a diagram showing the reproduction operation of another embodiment of the present invention, and FIG. 13 explains the modulation method of control information. Figure 14 is a bit pattern diagram explaining error correction encoding of control information, Figure 15 is a block diagram for modulation and error correction encoding of control information, and Figure 16 is a diagram of encoded and modulated control. A block diagram for demodulating and error correction of information, FIGS. 17 and 18 are diagrams explaining interleaving aspects for error correction of audio information, and FIG. 19 is a diagram for explaining error correction encoding and interleaving of audio information. FIG. 20 is a block diagram for error correction and deinterleaving of audio information that has been subjected to error correction encoding and interleaving. Explanation of codes of main parts, 1, 18...A/D converter, 2, 4...Error correction encoding circuit, 3...Time axis compression memory, 8, 9...Error correction circuit, 10...Time axis expansion memory, 11 ...D/A converter, 17...player controller.

Claims (1)

[Scope of Claims] 1. A two-dimensional screen obtained by a video format signal is divided into a plurality of blocks, audio information is inserted into a predetermined block, image information and control information are inserted into the remaining blocks, and the divided blocks are recorded on a recording medium. The video format is characterized in that the control information includes at least address information indicating the address of the audio information insertion block, so that the audio information corresponding to the image information can be inserted at any location. Matsuto signal recording method. 2. A patent claim characterized in that both the audio information and the control information are recorded as digital signals, and an error correction code is added so that errors can be corrected within the inserted blocks of the audio information and control information, respectively. The recording method described in item 1 of the scope. 3. Each block into which the audio information has been inserted is classified into a plurality of groups, each group is assigned a group address, and this group address is used as one of the control information. Recording method described in Section 1 or Section 2. 4. The image information is moving image information, and the audio information is time-axis compressed information whose length corresponds to the time equivalent to one frame of the moving image information and is inserted and recorded. The recording method described in Section 1, Section 2, or Section 3. 5 Divide the two-dimensional screen obtained by the video format signal into a plurality of blocks, insert time-axis compressed audio information into predetermined blocks, and add at least image information and address information indicating the address of the insertion block of the audio information to the remaining blocks. The included control information is inserted and recorded on a recording medium, and upon reproduction, the time-axis compressed audio information is stored in a memory means, and read out from this memory as audio information on a real time axis. Characteristic recording and playback method. 6. The recording and reproducing method according to claim 5, characterized in that external audio input is stored in the memory means, and this audio input is read out on a real time axis and derived together with the audio information. . 7. Claim 5 or 6, characterized in that the time-axis compressed audio information inserted into one field (frame) is information corresponding to image information of an integer field (frame). Recording and playback method described. 8. Claims 5 and 6 are characterized in that the block portion into which the audio information is inserted is clamped by a predetermined potential signal and output.
The recording and reproducing method described in paragraph 7 or paragraph 7.