JPS63244370A - Sound signal recording system - Google Patents

Abstract

PURPOSE:To favorably perform the amendment or the correction of a burst error by interleaving a code sequence corresponding to a sound signal being error-corrected and encoded, while the sections corresponding to n-fields periods of the code sequence being made into one unit, and afterwards, recording it while being multiplexed in the vertical flyback time of a video signal. CONSTITUTION:After the code sequence, corresponding to the sound signal which is made into a PCM and is corrected and encoded, is interleaved by a deinterleave circuit 4 while the sections corresponding to n-fields (n is 2 or a larger natural number) of the code sequence being made into one unit, the said code sequence is recorded while being multiplexed 2 to the vertical flyback time of the video signal. Thus, even if at the time of a reproduction, the part of the video signal to be clamped for a direct current reproduction, is omitted by a drop-out or the like, and the sound data of one field portion is omitted, it is converted into a random error by the deinterleave, and the error correction of the burst error by an error amendment or an interpolation can be favorably executed.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for multiplexing and recording an audio signal onto a video signal.

MUS compresses the bandwidth of a single high-definition television signal to approximately 8 MH2, making it possible to transmit it by broadcasting satellite.
E (Multiple 5ub-Nyquist
A method called Sampling Encoding (Sampling Encoding) has been proposed. According to this MUSE system, it is also easy to record high-quality television signals onto a recording medium such as an optical video disc.

In this MUSE system, audio signals are time-division multiplexed within the vertical blanking period from the 3rd line to the 46th line and from the 565th line to the 608th line.

If the MUSE signal on which this audio signal is multiplexed is recorded as it is on a recording medium and the audio signal is reproduced from this recording medium, the clamp for DC reproduction of the reproduced MUSE signal due to the dropout peculiar to the recording medium. When the portion where the clamping is performed is missing, the level of the missing clamped portion is clamped to a predetermined level, and the instantaneous level of the reproduced signal becomes abnormal over the period until the next clamped portion. In this case, it becomes impossible to extract the audio data, and the audio signal is lost for one field period, that is, about 16.7 ms, causing problems such as the generation of abnormal sounds.

&Dangomi 1 Therefore, it is an object of the present invention to provide an audio signal recording method that can satisfactorily correct or correct burst errors.

The audio signal recording method according to the present invention performs interleaving in units of intervals corresponding to n (n is a natural number of 2 or more) field periods of a code sequence corresponding to an audio signal that has been converted into PCM and subjected to error correction encoding. It is characterized in that this code sequence is later multiplexed and recorded during the vertical retrace period of the video signal.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

In FIG. 1, a MUSE signal is supplied to a pre-emphasis circuit 1, and after the video signal is emphasized, it is supplied to a time division multiplexing circuit 2.

On the other hand, depending on the predetermined clock frequency and quantization number, the A/D
The converted digital audio signal is sent to the CIRC encoder 3.
is supplied to. The C[RC encoder 3 is formed based on the CIRC method, which is an error correction encoding method used for digital audio discs. In the CIRC encoder 3, 24 symbols of data are treated as one block, and interleaved and parity is added to form one block consisting of 32 symbols (one symbol is 8 bits). Data encoded by this CIRC encoder 3 is supplied to an interleaving circuit 4.

In the interleaving circuit 4, the output data of the CIRC encoder 3 is supplied to a changeover switch 5. The changeover switch 5 is configured to selectively supply input data to one of the six RAMs 68 to 6f, for example, based on the control data of the three bits 1 to 6f. Control data for this changeover switch 5 is output from a changeover control circuit 7. The switching control circuit 7 is configured to generate 3-bit control data that changes each time one block of data is output from the CIRC encoder using, for example, a clock pulse from a clock generation circuit (not shown).

Each of the RAMs 68 to 6f acts as a FIFO (first in, first out) memory by the memory control circuit 8, and
Each time a block of data is output from the changeover switch 5, the write mode is entered sequentially, and these RAMs 6a to 6f are
When 735 blocks of data are written to the RAM
The reading of 122 blocks or 123 blocks of data written in each of 68 to 6f is controlled to be performed sequentially.

That is, when the output data of the CIRC encoder as shown in FIG.
As shown in each of F), each block forming this output data is sequentially supplied and written. RAM6a~6f
When 735 blocks of data are written to RAM 6a to 6ft, the written data is read out and interleaved with 6 fields as one unit, as shown in FIG. The data read from these RAMs 68 to 6f is supplied to the changeover switch 9.

The changeover switch 9 is configured to selectively output data read out from one of the RAMs 6a to 6'' using, for example, 3-bit control data. is output from the circuit 10.The switching control circuit 10 outputs 1 from the RAMs 6a to 6f.
It is configured to output 3-bit data that changes every time 22 blocks or 123 blocks of data are read. This switching control circuit 10 sequentially outputs read data from the RAMs 6a to 6f from the changeover switch 9, and becomes the output data of the interleaving circuit 4.

The output data of the interleaving circuit 4 is supplied to a 41ifI encoding circuit 11 and subjected to quaternary encoding. That is, the output data of the interleaving circuit 4 is divided into two bits each and converted into a four-value symbol series.

A Gray code is used as each code in this four-value code series, so that even if an error occurs in an adjacent level during level identification, the number of bit errors will be 1 bit. That is, as will be described later, the 2-bit codes <00, 01.11.10 are sequentially associated with each of four different levels.

The four-value coded data is supplied to a time axis compression circuit 12 and subjected to time axis compression. Data is intermittently outputted at a sampling clock frequency of 16.2M1-1z during 34 lines during the vertical branching period of the MUSE signal, ie, 191 of 68 lines in one frame. Note that the time axis compressor 12 is not necessarily the 4fi encoding circuit 4.
It is not necessary to place it after the , but it can be placed before it.

The output of the time axis compression circuit 12 is supplied to a 4-level circuit 13. The 4-level circuit 13 has a 4-value symbol 0O101.
, 11, and 10 correspond to level 0, level 1, level 2, and level 3, respectively. In other words, as shown in Figure 4, one sample value of a video signal is represented by an 8-bit level (256 levels), whereas an audio signal uses only 70% of the range of the video signal. being done. For example, the audio signal is designed to correspond to any one of 38 levels (level O), 98 levels (level 1), 158 levels (level 2), and 218 levels (level 3) of the video signal.

When the level range of the audio signal is increased (approximately 100%), overshoot and undershoot occur due to the influence of the 79127792 circuit 16, and the frequency band of the recording signal is occupied more than necessary, which tends to cause distortion. Also,
On the other hand, if it is made smaller, the data S/N deteriorates, and the influence of noise on the four-level discrimination increases. Therefore, it is preferable to use a range of around 70% as in the embodiment.

The time division multiplexing circuit 2 time division multiplexes the 256-level video signal and the 4-level audio signal output from the pre-emphasis circuit 1-. That is, the audio signal is transmitted to each of the 3rd to 36th lines in the vertical retrace period of the first field of the video signal and the 565th to 598th lines of the second field, a total of 68 lines, as shown in FIG. It is inserted into the formed audio data section. The frequency of the sampling clock of the MUSE signal is 16.2MH2, and the number of clocks per line is 480. In each line into which an audio signal is inserted, an HD double signal exists in the first 12 clock periods. An interval of 464 clocks that is separated by a guard interval of 2 clocks from this HD signal interval is an audio data interval, and an interval of 2099 minutes following the audio data interval is a guard interval. The guard section of 2 O's placed before and after the 464 clock is )-
Preparations are being made to reduce the impact on ID signals. In addition, the number of clocks of 464 clocks means that the number of bytes in one line is 116 bytes, which is 1 byte (8 bits).
Consideration has been given to ensuring that the data of each line is not configured across different lines.

The output of the time division multiplexing circuit 2 is supplied to a transmission filter 14 composed of, for example, an FIR digital filter, and its frequency band is limited to a predetermined width. In order to minimize code amount interference, the overall characteristic formed by the transmission filter 14 and the reception filter 24, which will be described later, is made to be a cosine roll-off characteristic. Increasing the roll-off rate α here widens the band and lowers the S/N.

On the other hand, if it is made smaller, the configuration becomes complicated in order to accurately realize the characteristic (@sharp characteristic). Therefore, it is preferable to set the roll-off rate α to about 0.3, for example.

The output of the transmission filter 14 is supplied to a D/A converter 15 and converted into an analog signal. The output of the D/A converter 15 is emphasized by a pre-emphasis circuit 16 and then supplied to an FM modulation circuit 17 .

In the FM modulation circuit 17, a carrier wave of a predetermined frequency is frequency-modulated by the supplied signal. The output of this FM modulation circuit 17 is passed through a limiter 18 to an E10 modulator (
(not shown), etc., and recorded on a video disc.

Since audio data is inserted as a four-value signal by the recording device as described above, the audio data section of each line contains one
This means that 2 bits of data can be inserted per block, and 116 bytes of audio data can be multiplexed.

Here, in CIRC encoding, one block is formed of 32 bytes, so three blocks of data and 20 bytes of data are inserted into the audio data section per line, and the vertical retrace line An audio data block is recorded in each line in the period as shown in FIG. 6, and 245 blocks of audio data can be recorded in one entire frame. In this case, the data transmission capacity is 1,8816Mbps, 44.1kH2t
'It becomes possible to multiplex two channels of sampled and 16-bit linear quantized audio data.

Also, consecutive 6 fields 6f, 6f+1.6f+2
, . . . 6f+5, the arrangement of the audio signals multiplexed during each vertical retrace period is changed by the interleaving circuit 4 as shown in FIG. In this way, after encoding, interleaving is performed using 6 fields as a unit, so during playback, the portion where clamping is performed for DC playback of the video signal obtained after FM demodulation is free from burst errors due to dropouts etc. Even if 1 field worth of audio data is lost due to a drop, the 68 μs length will be reduced to 400 μs due to deinterleaving.
The error is converted into a random error that periodically occurs every 0.1 seconds, so that error correction of burst errors or error correction by interpolation can be performed satisfactorily.

In the above embodiment, interleaving is performed in units of six fields, but the unit of interleaving may be any value as long as it is two or more fields and an integral multiple of one field. However, it is necessary to change the number of RAMs in the interleave circuit 4 and the number of terminals of the changeover switch 5.9 depending on the number of unit fields. Further, interleaving creates a time difference between audio and video, so when the number of unit fields is increased, it is desirable to provide means for delaying the video signal in order to eliminate the time difference between audio and video.

On the other hand, when the video signal is not delayed, etc., if the maximum allowable time difference is Dm, one field period is tF, and the number of unit fields of interleaving 1 is n, then Dm
It can be expressed as ≧n-tF.

FIG. 5 is a block diagram showing a part of a reproducing apparatus for reproducing information recorded on a video disc by the recording apparatus as described above. In the figure, an RF multiplied signal read from a video disc is supplied to an FM demodulation circuit 21 and frequency demodulated. The demodulated output is sent to the de-emphasis circuit 22.
The level of the component emphasized during recording is returned to its original level. The output of this de-emphasis circuit 22 is supplied to a reception filter 24 to limit the band.
The D converter 25 converts the signal into a digital signal.

The output of this A/D converter 25 is transmitted to the time division separation circuit 2.
6 and is time-division separated into a video signal and an audio signal. The video signal is supplied to a de-emphasis circuit 23, and the level of the component emphasized during recording is returned to the original level.

The output of this de-emphasis circuit 23 is supplied to a MUSE decoder (not shown) or the like.

On the other hand, the audio signal is supplied to the four-level identification circuit 27, which of the four levels described above is identified, and converted into a four-value code. The signal converted into quaternary encoding is subjected to time axis expansion by a time axis expansion circuit 28, and then supplied to a quaternary decoding circuit 29, where the quaternary encoding is converted into a normal binary encoding. The output of this four-value decoding circuit 29 is supplied to a deinterleaving circuit 30. Similar to the interleaving circuit 4, the deinterleaving circuit 30 includes changeover switches 31 and 32, RAMs 33a to 33f, switching control circuits 34 and 35, and a memory control circuit 36.
Switching of each of the changeover switches 31 and 32 is performed in the same manner as each of changeover switches 9 and 5 in the interleaving circuit 4, and writing and reading of RAMs 33a to 33f are performed to the RAMs 68 to 6f in the interleaving circuit 4.
This is done in the same way as reading and writing. The deinterleaving circuit 30 restores the arrangement changed during recording. The output of the deinterleaving circuit 30 is supplied to a CIRC decoder 37, where it undergoes processing such as error detection and correction, and then supplied to an audio decoder (not shown) or the like.

Pre-emphasis circuit 16 and de-emphasis circuit 22
The characteristics are set so that the emphasis of the audio signal is optimal. Since the frequency spectrum of the PCM audio signal has relatively many high frequency components, the amount of emphasis is set relatively shallow. Since the pre-emphasis circuit 16 and the de-emphasis circuit 22 are constructed from analog circuits, their characteristics can be set relatively freely, and their constructions are not very complicated. When setting the emphasis characteristics as shown in FIG. 9, for example, m=1.7 and fo=IMH2.

On the other hand, the characteristics of the pre-emphasis circuit 1 and the de-emphasis circuit 23 are set so that the emphasis of the video signal is optimized. However, since the video signal is also emphasized by the pre-emphasis circuit 16 and the de-emphasis circuit 22, the overall characteristics of both are set to be optimal. Since the frequency spectrum of the video signal has relatively few high frequency components, the amount of emphasis is set relatively deep. Pre-emphasis circuit 16
Since shallow emphasis is performed by the pre-emphasis circuit 1 and the de-emphasis circuit 22, the emphasis of the pre-emphasis circuit 1 and the de-emphasis circuit 23 can be shallower than when the characteristics are set alone. Therefore, although these are constructed from digital circuits, their scale can be made smaller than when the characteristics are set alone.

In the above embodiment, the pre-emphasis circuit 1 and the de-emphasis circuit 23 are inserted in the video signal path, but the same effect can be obtained even if they are inserted in the audio signal path. It is also possible to provide a dedicated emphasis circuit for each path of the video signal and the audio signal.

Figure 10 shows the relationship (theoretical value) between the S/N ratio (horizontal axis) and symbol error rate (vertical axis) when audio signals are transmitted in binary, 4-value, 8-value, and 16-value format. There is. Here, the S/N ratio is the ratio between the peak-to-peak value of the data level and the effective value of noise, and the noise is Gaussian noise. As is clear from the figure, in order to realize the symbol error rate of i x i o', the four-value case 27c[3,8
A signal-to-noise ratio of 35 dB is required. In the above embodiment, it is possible to secure an S/N ratio of about 35 for the reproduced baseband signal from the video disc, but the symbol error rate also deteriorates due to factors such as jitter and code amount interference. It is also necessary to consider deterioration due to dropout, which is unique to disks. Therefore, 8
It is difficult to record or reproduce data exceeding this value.

Since it is advantageous for signal processing to set the level to 2n values, it is preferable to set the level to 4 values as in the embodiment.

As described in more detail with Bon JJ, the audio signal recording method according to the present invention is based on P.
After interleaving is performed in units of sections corresponding to n (n is a natural number of 2 or more) field periods of a code sequence corresponding to an audio signal that has been converted into a commercial and error correction encoded, this code sequence is used as a video signal. Since it is multiplexed and recorded during the vertical retrace period, even if the portion that is clamped during playback of the video signal is dropped due to dropout, etc., and one field's worth of audio data is missing, there will be no error. By interleaving, the error is converted into a random error, and it is possible to perform error correction of the burst error or error correction by interpolation.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIGS. 2 and 3 are diagrams showing the operation of the apparatus shown in FIG. 1, and FIG. 4 is an explanatory diagram of the recording signal level. 5 to 7 are diagrams showing multiplexing formats of audio signals recorded by the device in FIG. 1, and FIG. 8 is a block diagram showing a device for reproducing the signals recorded by the device in FIG. 1. FIG. 9 is a characteristic diagram of emphasis, and FIG. 10 is a diagram showing theoretical values of error rate. Qin 4 Tuo 5 Congwei 7wJ

Claims (1)

[Claims] This is an audio signal recording method in which a PCM audio signal is multiplexed with a video signal and recorded on a recording medium. (Natural number) An audio signal recording method characterized in that after performing interleaving in units of intervals corresponding to field periods, the code sequence is multiplexed and recorded in the vertical retrace period of the video signal.