JPH0963201A - Disk type recording medium and reproducing apparatus - Google Patents

Abstract

Kind Code: A1 Abstract: The reproduced and PR-equalized user data is accurately decoded even when the conditions for recording and reproducing the user data on the magneto-optical disk are different. SOLUTION: This data is reproduced on a magneto-optical disk 22 having a user data area DT prior to reproduction of this data,
A training pattern for calculating the ideal amplitude value for determining the logic of the data is discretely recorded in this data, and the reproducing device determines the ideal amplitude value from the reproduced pattern, and the user input immediately after that is determined. The data is decoded based on this determined latest amplitude value. As a result,
Since the user data can be decoded by the ideal amplitude value corresponding to the data to be processed, accurate decoding can be performed even if the DC component of the reproduced signal varies, and therefore bit errors in the reproduced data (decoded data) can be prevented. it can.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disc type recording medium such as a magneto-optical disc and a reproducing apparatus for reproducing user data recorded on the disc type recording medium.

[0002]

2. Description of the Related Art Disc-type recording media such as magneto-optical discs of this type allow users to record their own data as needed and, of course, recorded data can be reproduced. It is becoming popular. FIG. 14 is a diagram schematically showing the principle of data recording on such a magneto-optical disk, where 11 is a laser for recording data and 1 is a laser.
Reference numeral 2 is a magnetic field generator, and 22 is a magneto-optical disk. The magneto-optical disk 22 is always magnetized in the upward direction (the direction of the solid arrow in the figure). Such a magneto-optical disk 22
On the other hand, when recording data, the laser 11 irradiates a pulsed laser beam according to the data. Then, the recording portion 22A in which the data of the magneto-optical disk 22 is recorded
Is heated by the irradiation of laser light and is magnetized by the magnetic field generated by the magnetic field generator 12 in the direction of the dotted arrow (downward) in the figure. In this manner, the magneto-optical disk 22 which is normally magnetized in the upward direction is magnetized in the downward direction, so that the mark corresponding to the user data is recorded in the recording portion 22A.

The digital user data recorded on the magneto-optical disk 22 in this way is reproduced by a reproducing device (not shown). However, if the bits of the data are recorded at a high density, the reproduced waveforms of adjacent bits are mutually different. The data cannot be accurately identified because it is reproduced in a form interfered with. Therefore, the data is reproduced by the reproduction method called the partial response method. That is, the reproduction signal of the data recorded on the magneto-optical disk 22 is set to P
A PR signal (partial response signal) is generated by conversion by a circuit called an R equalization circuit. Then, the PR signal is sampled by the clock signal generated from the PR signal to generate sample data.

The sample data thus generated is decoded by a decoder (that is, a bit-by-bit decoder or a Viterbi decoder described later) based on an ideal value, and is output as reproduced data. In this case, in both the system called a PR system using a bit-by-bit decoder and the system called a PRML system using a Viterbi decoder, the above ideal value of N (N is an integer of 2 or more) values is set in advance. It is set. The bit-by-bit decoder determines the threshold level from these ideal values,
Decoding is performed based on this threshold level to generate reproduced data. In addition, the Viterbi decoder performs Viterbi decoding from these ideal values to generate reproduction data.

[0005]

However, the recording conditions when recording user data on a magneto-optical disk are not always constant, and even if an attempt is made to record data of a predetermined bit length, the surroundings at the time of recording The length of the actually recorded user data fluctuates due to fluctuations in the temperature and the power of the laser that records the data. Further, since the magneto-optical disc on which the user data is recorded is not always reproduced by the same reproducing device, when the data recorded on the magneto-optical disc is reproduced by different reproducing devices, the reproduced data respectively. Varies in length. Such variations in recording the user data and reproducing the user data cause a shift in the waveform of the reproduction signal from the magneto-optical disk, and as a result, the sample data generated by PR equalization of the reproduction signal is generated. Needless to say, it also varies depending on the above-mentioned recording and reproducing conditions.

Conventionally, the ideal value of the sample data required for each decoding process of the bit-by-bit decoder of the PR system and the Viterbi decoder of the PRML system is fixed regardless of the above-mentioned recording and reproducing conditions of the user data. Since it is set to the desired value, there is a gap between the ideal value and the sample data to be processed, so that the bit-by-bit decoder and the Viterbi decoder cannot correctly decode the data, and as a result, the data is reproduced by each decoder. There is a problem in that the reproduced data has increased bit errors. The present invention accurately decodes sample data of reproduced and PR-equalized user data in each decoder even if the conditions for recording and reproducing user data with respect to the magneto-optical disk are different, thereby increasing the bit error of the reproduced data. The purpose is to suppress.

[0007]

In order to solve such a problem, the present invention reproduces the user data on a magneto-optical disk having a first area for recording the user data prior to the reproduction of the user data. A second area for recording a training pattern for calculating an ideal amplitude value for determining the logic of data is provided. As a result, when the sample data of the reproduced user data is decoded based on the ideal amplitude value and is generated as the reproduction data, the sample data is decoded by the ideal amplitude value corresponding to the sample data to be processed. It is possible to prevent an increase in the number of bit errors in the reproduced data (decoded data) due to erroneous decoding and an error in the decoding process due to the value being deviated from the actual sample data.

Further, the second area is provided in the head portion of the first area and in the first area. As a result, the DC component of the user data fluctuates when the user data of the magneto-optical disk is reproduced, and even when the reproducing device side removes the DC component and reproduces the signal, the fluctuation of the DC component of the user data occurs. The ideal amplitude value can be updated according to the above, and therefore, the decoding circuit can always decode the user data with the ideal amplitude value that matches the amplitude distribution of the user data, and the bit error of the reproduced data can be reliably prevented. In addition, the second area is provided at the beginning of the first area, and the value of each bit of the user data recorded in the first area is biased to one of "0" and "1". If so, the second region is provided in the first region. As a result, it is possible to prevent bit errors in the reproduced data as necessary.

Further, a PR equalizer circuit for equalizing a reproduced signal of user data recorded in the first area of the magneto-optical disk based on the partial response system, and a clock signal is generated from an output signal of the PR equalizer circuit. A PLL circuit, an A / D converter that samples the output signal of the PR equalization circuit based on a clock signal to generate sample data, and decodes and reproduces sample data of user data based on a preset ideal amplitude value. In a reproducing device including a decoding circuit for generating data, a control circuit for outputting a control signal during reproduction of a training pattern recorded in the second area of the magneto-optical disk, and A / D conversion during output of the control signal An ideal value determination circuit that determines the ideal amplitude value from the sample data of the training pattern output from the device and provides it to the decoding circuit is provided. Than it is. As a result, the sample data is decoded by the ideal amplitude value corresponding to the sample data of the user data to be processed, so that the bit error of the decoded reproduction data can be reduced. Further, if the training pattern is discretely recorded in the user data, the DC component of the reproduced user data varies, and the reproducing device side removes the DC component to reproduce the signal. Even in this case, since the ideal amplitude value can be updated according to the fluctuation of the DC component, the decoding circuit can always decode the user data with the ideal amplitude value that matches the amplitude distribution of the user data, and thus the bit error that occurs in the reproduced data can be reduced. It can be surely prevented.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the drawings. Generally, in an information reproducing apparatus for reproducing digital data, which is user data recorded in a magneto-optical disk apparatus or the like, data is reproduced by a reproducing method called a partial response (hereinafter referred to as PR) method. There are various types of PR methods, and here, the reproduction principle will be described by taking the PR (1,1) method, which is often used in magneto-optical disk devices, as an example. FIG. 2 shows a system for recording / reproducing data based on the PR (1,1) system (hereinafter,
PR system). Further, FIG. 3 is a waveform diagram showing operation waveforms of respective parts of this PR system.

When the precoder 21 shown in FIG. 2 receives the digital data shown in FIG. 3A output from an external circuit (not shown), it converts it into pattern data shown in FIG. 3B. Such pattern conversion is called NRZI conversion, and is realized by the adder 41 that performs modulo 2 addition and the delay circuit 42 that delays one bit of data shown in FIG. That is, the adder 41 performs modulo 2 addition of the input digital data and the data obtained by delaying the output of the adder 41 by 1 bit, and outputs the NRZI data to the magneto-optical disk 22 for recording.

NRZI by such a precoder 21
The conversion is necessary because the propagation of an error can be prevented by a process by bit-by-bit decoding which will be described later and is performed at the time of reproduction. The data thus recorded on the magneto-optical disk 22 is reproduced as a reproduction signal as shown in FIG. 3C and output to the PR equalization circuit 23. PR
The equalizing circuit 23 converts this reproduced signal into a PR signal shown in FIG. Such conversion by the PR equalization circuit 23, that is, PR equalization is realized by the adder 51 and the delay circuit 52 shown in FIG.

That is, FIG. 3 from the magneto-optical disk 22.
The reproduced signal shown in (c) is supplied to the adder 51, while the delay circuit 52 outputs 1T (T is a time for 1 bit).
Is delayed by the adder 5 as a delayed signal shown in FIG.
Given to one. The adder 51 adds these two signals together to generate the PR signal shown in FIG.
4 and the PLL circuit 25. In this case, the PLL circuit 25 generates the clock signal of FIG. 3 (f) synchronized with the input PR signal from the input PR signal to generate the A / D converter 2
Give to 4. In the A / D converter 24, the PR signal is sampled by this clock signal to generate ternary pattern data of i2, i1, i0 as shown in FIG. 3 (g), which is given to the bit-by-bit decoder 26 as sample data. .

Here, the state of the medium (magneto-optical disk),
Jitter is generated in the PR signal of FIG. 3E due to the noise caused by the recording state of the medium and the reproducing circuit that reproduces data from the medium, so that the amplitude distribution of the actual sample data is i2, 3 centered on the three values of i1 and i0
It has a mountainous distribution. Here, the central values i2, i1,
i0 is called an ideal amplitude value. Bit-by-bit decoder 2
In 6, the input sample data is converted into the reproduction data shown in FIG.

FIG. 6 is a block diagram showing the configuration of the bit-by-bit decoder 26, which comprises a ternarization circuit 61 and a binarization circuit 62. When the ideal amplitude values input from an external device (not shown) are i2, i1, and i0, the ternarization circuit 61 sets threshold levels between i2 and i1 and between i1 and i0, respectively. Which logic the above sample data has is judged based on the threshold level.

Here, in the ternarization circuit 61, when the threshold level is set from the input ideal amplitude value, i2 and i1
It is conceivable to set the intermediate value between and and the intermediate value between i1 and i0 as the threshold level. On the other hand, the binarization circuit 62 which receives the output of the ternarization circuit 61 inputs "0" as the reproduction data when the ternarization circuit 61 logically determines the maximum value or the minimum value, and the intermediate value and the logical value. When it is determined, "1" is output as the reproduction data. As a result, in the PR system shown in FIG. 2, the digital data of FIG. 3A recorded on the magneto-optical disk 22 is recorded in the PR system shown in FIG. The data and the reproduction data of FIG. 3 (h) are the same data.

By the way, due to noise caused by the state of the medium, the recording state of the medium, the reproducing circuit, etc., P in FIG.
When the jitter generated in the R signal is large and the data at an arbitrary sample point is changed to exceed the threshold level, the PR system shown in FIG. 2 is directly connected to the bit error of the reproduced data. Therefore, in order to reduce the occurrence of such bit errors, a PRML system is known in which the Viterbi decoder 76 shown in FIG. 7 is used instead of the bit-by-bit decoder 26. That is, in the PRML system shown in FIG. 7, the magneto-optical disk 22, the PR equalization circuit 23, the A / D converter 24, and the PLL circuit 25 are used as in the PR system shown in FIG.
The precoder 21 that performs the NRZI conversion required in the R system uses the Viterbi decoder 76 for PRML in FIG.
No longer needed in the system.

FIG. 8 is a diagram showing operation waveforms of each part of the PRML system. That is, the NR already shown in FIG.
The ZI-converted digital data is the magneto-optical disk 2
When recorded in 2, the signal reproduced from the magneto-optical disk 22 becomes a signal as shown in FIG. 8B and is output to the PR equalization circuit 23. In the PR equalization circuit 23, P of FIG.
Conversion is performed in the same manner as in the R system, and the converted PR signal shown in FIG. 8C is given to the A / D converter 24 and the PLL circuit 25. The PLL circuit 25 generates the clock signal shown in FIG. 8D which is synchronized with the input PR signal from the input PR signal and gives it to the A / D converter 24. In the A / D converter 24, the PR signal is sampled by this clock signal and given to the Viterbi decoder 76 as sample data as shown in FIG.

Here, due to noise caused by the state of the medium, the recording state of the medium and the reproducing circuit, signal jitter is generated in the PR signal of FIG. 8C, so that the sample data of FIG. Similar to the PR system in FIG. 2, the actual amplitude distribution of is a three-peaked distribution centered on the ideal amplitude values i2, i1, and i0. The Viterbi decoder 76 reproduces digital data similar to the data shown in FIG. 8A from such sample data.

FIG. 9 is a block diagram showing the configuration of the Viterbi decoder 76. Two state registers 91 and 92 and two state registers 91 and 92 are provided.
It is composed of N (N is an arbitrary natural number) bit data series register 93, 94, a state calculation circuit 95, and a determination circuit 96. Here, “0” is set in the lower bit of the data series register 93, and the data series register 94
"1" is set in the lower bit of the. And
Each time the A / D converter 24 supplies sample data (referred to as sample data y), the Viterbi decoder 76 performs the following processing.

That is, first, in the state calculation circuit 95, the values M0 and M1 of the two state registers 91 and 92, the sample data y output from the A / D converter 24, and the sample data supplied from an external device (not shown) Expected values d0, d
1, d2 (where d2 is the maximum value, d1 is the intermediate value, and d0
Corresponds to the minimum value, and when the ideal amplitude values of the sample data are i2, i1, and i0, respectively, d2 = i2, d1 = i
1, d0 = i0), the following equations (1) to
By (4), four values M00, M01, M10, M1
Calculate 1. M00 = M0-2 · y · d0 + d0 ² (1) M01 = M1-2 · y · d1 + d1 ² (2) M10 = M0-2 · y · d1 + d1 ² (3) M11 = M1-2 · y · d2 + d2 ² ( 4)

Next, the decision circuit 96 first determines the value M00 and the value M.
Compare the size with 01. When M00 <M01, M00 is stored in the state register 91, and the value of (N-1) to 0 bits of the data series register 93 is shifted by 1 bit in the upper direction. When M00 ≧ M01, M
01 is stored in the status register 91, and the (N-1) to 0 bit values of the data series register 94 are copied to the N to 1 bits of the data series register 93, respectively.

Subsequently, the decision circuit 96 outputs the value M10 and the value M1.
Compare the size with 1. If M10> M11, M11 is stored in the status register 92, and the value of (N-1) to 0 bits of the data series register 94 is set to 1 in the upper direction.
Bit shift. When M10 ≦ M11, M1
0 is stored in the status register 92, and (N-1) to 0 bits of the data series register 93 are copied to N to 1 bits of the data series register 94, respectively. afterwards,
The determination circuit 96 outputs the most significant bit of the data series register 93 to the outside as reproduction data.

As described above, when the sample data is input, the Viterbi decoder 76 determines whether the data is "0" or "1" according to a certain threshold level, like the bit-by-bit decoder 26. It does not perform, but makes a soft decision to determine a more probable data sequence while looking at the states before and after the data. Therefore,
The two series registers 93 and 94 converge to either “0” or “1” each time the shift operation is performed, and when the data reaches the position of the most significant bit and is output as the reproduction data, the data is not stored. It is set to either "0" or "1".

Here, the number of bits N of the data series register
If it is small, there is a risk that it will not be fixed to either "0" or "1", and the larger the number of bits N, the smaller the undetermined probability, and therefore the smaller the bit error of the reproduced data. You can Therefore, if the number of bits N of the two series registers 93 and 94 is increased, the data is fixed. Therefore, the most significant bit of either register may be output as the reproduction data. The upper bits are output as reproduction data.

In the Viterbi decoder 76, the state calculation circuit 95 calculates the occurrence probability of each data series, and the determination circuit 96 calculates the occurrence probability of the two data series as the processing result of the state calculation circuit 95. Although the surviving data series is determined such that the higher one is regarded as the surviving data series, the data series registers 93 and 94 store the two surviving data series, and the status register 9
1, 92 have a role of storing the occurrence probability of the survival data series.

As described above, the operation of the PR system and the PRML system has been described. In any system, since reproduction data is generated from sample data of the PR signal, ternary data for decoding ternary data is decoded. The ideal amplitude value of is required. The bit-by-bit decoder 26 of the PR system sets a threshold level from these ideal values and binarizes the sample data according to this threshold level. Further, the Viterbi decoder 76 of the PRML system uses these ideal amplitude values as expected values to decode the sample data.

By the way, the recording condition when recording digital data on the magneto-optical disk 22 is not always constant, and even if an attempt is made to record a mark of a predetermined length corresponding to the digital data, the ambient temperature at the time of recording The length of the mark actually recorded on the magneto-optical disk fluctuates due to fluctuations in the power of the laser for recording data and the like. Further, since the magneto-optical disc is not always reproduced by the same reproducing device, when the marks recorded on the magneto-optical disc 22 are reproduced by different reproducing devices, the length of the reproduced mark is different for each reproducing device. Variation occurs from device to device.

Therefore, if the ideal amplitude value required for each decoding process is fixed in the above-mentioned bit-by-bit decoder 26 or Viterbi decoder 76, this fixed ideal amplitude value is the amplitude distribution of the sample data of the user data to be reproduced. Therefore, there is a problem in that reproduced data decoded based on the ideal amplitude value often has bit errors. Therefore, as shown in FIG.
The training pattern is recorded in the training pattern area TP adjacent to the head portion of the user data area DT in which the user data is recorded in the magneto-optical disk 22. Then, this training pattern is set to a pattern as shown in FIG. 11A, and PR equalized and sampled training pattern sample data is used as shown in FIG.
A pattern including all three levels is obtained as shown in 1 (b).

The sample data of such a training pattern is based on a gate signal from an ideal value arithmetic control circuit, which will be described later, while the pattern is being reproduced.
The pattern read circuit 7 shown in FIG. Then, the ideal value determination circuit 8 shown in FIG. 12 determines the ideal amplitude values i2, i1, i0 corresponding to the three levels of maximum, intermediate, and minimum from the obtained training pattern sample data.

As a method of determining this ideal amplitude value, the sample data taken in by the pattern read circuit 7 is sequentially sorted from the beginning into two logical levels with two threshold levels, and the first sample belonging to each group is selected. It is conceivable to use the data as the ideal amplitude value of each logic level. Using this method, FIG.
I2 ₀ , i1 ₀ , and i0 ₀ in the data shown in are selected as the ideal amplitude values i2, i1, and i0.

The ideal amplitude values i2, i1, i0 determined by the ideal value determination circuit 8 in this way are given to the bit-by-bit decoder 26 or the Viterbi decoder 76 and used for decoding the subsequent user data. In this case, the bit-by-bit decoder 26 determines the threshold level from the given ideal amplitude value as described above. Then, the sample data of the user data area DT that is subsequently reproduced is decoded based on this threshold value and output as reproduction data. The Viterbi decoder 76 uses this ideal amplitude value as the above-mentioned expected value, and similarly decodes the sample data of the subsequent user data area DT based on this expected value.

In general, a magneto-optical disk has a recording / reproducing characteristic that allows the direct current component of a signal to pass therethrough, but when a signal containing a direct current component is reproduced by a reproducing device, it is not shown in the reproducing device. The amplifier circuit needs to have a wide frequency band. However, when such a wide band amplifier circuit is used, signal quality generally tends to deteriorate. Therefore, the band of the amplifier circuit is limited to remove the DC component and reproduce the signal. If the user data recorded on the magneto-optical disk 22 contains a DC component and this DC component fluctuates, the magneto-optical disk 2
The envelope of the reproduction signal of the user data reproduced from 2 fluctuates as the DC component changes. As a result,
The amplitude distribution of the sample data of the user data obtained from the reproduced signal also changes as the DC component changes.

Therefore, when the sample data of the user data area DT is to be decoded by the ideal amplitude value determined based on the pattern of the training pattern area TP, the ideal amplitude value obtained from the leading training pattern is The amplitude distribution of the user data in the second half of the data area DT may deviate, and as a result, the data in the second half of the user data area DT may not be decoded correctly. Therefore, as a data area of the magneto-optical disk 22, a training pattern area is provided in the head of the user data area DT and in the user data area DT, respectively.

FIG. 1 is a diagram showing the structure of such a magneto-optical disk 22. In FIG. 1A, T indicates a track on which user data and a training pattern are recorded. The track T is composed of a plurality of sectors S _{1 to} S _N, as shown in FIG.
Each sector has an area I in which an identification code is recorded at the beginning.
D is provided, and subsequently, a training pattern area and a user data area are provided alternately in the order in which the data is reproduced. The training pattern area TP
.., TP2, ... Are the same patterns and are the patterns shown in FIG. Also,
When the respective data in the user data areas DT1, DT2, ... Are summed up, they become the same as the data in the user data area DT shown in FIG. That is, in the present embodiment, when the user data is recorded on the magneto-optical disk 22, the training pattern is scattered and recorded discretely at various points in the user data together with the user data.

FIG. 13 is a block diagram of a reproducing apparatus for reproducing data from the track T of the magneto-optical disk 22 constructed as described above. This reproducing apparatus is configured by adding the above-described ideal value arithmetic control circuit 5, the pattern read circuit 7 and the ideal value determining circuit 8 shown in FIG. 12 to the PR system of FIG. Next, the operation of this reproducing apparatus will be described. The ideal value calculation control circuit 5 enables the gate signal to the pattern read circuit 7 at the timing when the data of the training pattern area TP is reproduced from the magneto-optical disk 22.

That is, while the gate signal is enabled, the data in the training pattern area TP of the magneto-optical disk 22 is reproduced, this reproduced signal is equalized by the PR equalizing circuit 23, and then A / D. The converter 24 generates the sample data. The pattern read circuit 7 takes in the sample data of the training pattern generated while the gate signal is enabled and sends it to the ideal value determination circuit 8. The ideal value determination circuit 8 determines the ideal amplitude values i2, i1, i0 from this sample data. Note that the ideal value determination method in the ideal value determination circuit 8 is the method described in FIG. 11 (b) above, but in addition to this, an average circuit is used to average each sample data for each three-value level. There is also a method.

In this way, when reading the data of an arbitrary sector S in the track T of the magneto-optical disk 22,
The ideal amplitude calculation control circuit 5, the pattern read circuit 7, and the ideal value determination circuit 8 are used to determine the ideal amplitude value in all the training pattern regions in the sector S. The determined ideal amplitude values i2, i1, i0 are sent to the bit-by-bit decoder 26 and used for decoding the sample data of the user data area reproduced immediately after each training pattern area. That is, in the bit-by-bit decoder 26, the threshold level is set based on the determined ideal amplitude value, and the sample data of the user data area that is input immediately after is decoded according to the threshold level to obtain the reproduction data. To generate.

Although the example of the PR system using the bit-by-bit decoder 26 as the decoder has been described in the example of FIG. 13, it is obvious that the same can be applied to the PRML system using the Viterbi decoder 76. That is, when the data of a certain sector S in the magneto-optical disk 22 is read, similarly, the ideal value calculation control circuit 5, the pattern read circuit 7 and the ideal value determination circuit 8 are used to read the sector S.
Determine the ideal amplitude value for all training pattern regions in. Then, the determined ideal amplitude values i2, i1, i
0 is sent to the Viterbi decoder 76. Viterbi decoder 76
Then, using the determined ideal amplitude value as the expected value, the sample data of the user data area input immediately after is
It is decoded based on the expected value and generated as reproduction data.

As described above, the training pattern is recorded not only in the head of the user data area of the magneto-optical disk but also in the user data area, and the training pattern is reproduced and reproduced prior to the reproduction of the user data area. The ideal amplitude value is determined from the pattern, and the user data input immediately after is determined based on the determined latest ideal amplitude value. As a result, when the user data recorded on the magneto-optical disk 22 is reproduced, the DC component of the user data fluctuates, and even if the reproducing device removes the DC component to reproduce the signal, the DC of the user data is reproduced. The ideal amplitude value can be updated according to the variation of the component. Therefore, the bit-by-bit decoder 26 or the Viterbi decoder 76 can always decode the user data with an ideal amplitude value that matches the amplitude distribution of the user data, and it is possible to prevent a bit error from occurring in the reproduced data.

When recording digital user data in each sector S of the magneto-optical disk 22, the number of bit values "1" and the number of bit values "0" of this digital data are compared, and the bit value " Only when the number of "0" s is much larger than the number of bit values "1", or when the number of bit values "0" is much smaller than the number of bit values "1", the training pattern is set to the user data. Recording may be performed discretely in the area. That is, if the above-mentioned imbalance occurs between the number of bit values "1" and the number of bit values "0" of recorded digital data, the reproduced signal has a large DC portion, and this DC component is If it fluctuates, a bit error frequently occurs in the data reproduced by each decoder. Therefore, if the training pattern is discretely recorded only in such a case, the bit error can be suppressed as necessary.

[0042]

As described above, according to the present invention, the ideal amplitude value which is reproduced prior to the reproduction of the user data in the first area of the magneto-optical disk and which determines the logic of this user data is calculated. Second training pattern for
Since it is recorded in the area of, when the sample data of the reproduced user data is decoded based on the ideal amplitude value and is generated as reproduction data, the sample data is decoded by the ideal amplitude value corresponding to the sample data to be processed. Therefore, it is possible to suppress an increase in the bit error of the reproduced data due to erroneous decoding and an error in the decoding process due to the deviation of the ideal amplitude value from the actual sample data. Further, since the second area is provided in the head portion of the first area and in the first area, the DC component of the user data fluctuates when reproducing the user data of the magneto-optical disk, and the reproducing apparatus is also used. Even when the DC component is removed on the side to reproduce the signal, the ideal amplitude value can be updated according to the fluctuation of the DC component of the user data. Therefore, in the decoding circuit, the ideal amplitude value that always matches the amplitude distribution of the user data. User data can be decoded with the amplitude value, and bit errors in the reproduced data are reliably prevented. Also, the second area is provided at the beginning of the first area,
When the value of each bit of the user data recorded in the first area is biased to either one of "0" and "1", the second area is discretely distributed in the first area. The bit error of the reproduced data can be accurately prevented as necessary. Further, the training pattern is reproduced prior to the reproduction of the user data of the magneto-optical disk, the ideal amplitude value is determined from the sample data of the reproduced pattern, and the sample data of the user data input immediately after is determined by the determined latest data. Since the decoding is performed based on the ideal amplitude value, the sample data is decoded by the ideal amplitude value corresponding to the sample data of the user data to be processed, so that the bit error of the decoded reproduction data can be reduced. Further, if the training pattern is discretely recorded in the user data, the DC component of the reproduced user data varies, and the reproducing device side removes the DC component to reproduce the signal. Even in this case, since the ideal amplitude value can be updated according to the fluctuation of the DC component, the decoding circuit can always decode the user data with the ideal amplitude value that matches the amplitude distribution of the user data, and thus the bit error that occurs in the reproduced data can be reduced. It can be surely prevented.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a disk type recording medium (magneto-optical disk) according to the present invention.

FIG. 2 is a diagram illustrating a P for processing a reproduction signal according to a PR system in a reproducing device for reproducing data on the magneto-optical disk.
It is a block diagram which shows the structure of R system.

FIG. 3 is a diagram showing operation waveforms of respective parts of the PR system.

FIG. 4 is a diagram showing a configuration of a precoder used in a PR system for performing NRZI conversion.

FIG. 5 is a diagram showing a configuration of a PR equalization circuit used in a PR system.

FIG. 6 is a diagram showing a configuration of a bit-by-bit decoder used in a PR system.

FIG. 7 is a block diagram showing a configuration of a PRML system in the playback device.

FIG. 8 is a diagram showing operation waveforms of respective parts of the PRML system.

FIG. 9 is a diagram showing a configuration of a Viterbi decoder used in the PRML system.

FIG. 10 is a diagram showing a data recording state of a magneto-optical disk.

FIG. 11 is a diagram showing a format of a training pattern recorded on a magneto-optical disk.

FIG. 12 is a block diagram showing a main part of the playback device.

FIG. 13 is a block diagram when the present invention is applied to a PR system.

FIG. 14 is a diagram schematically showing the principle of recording data on a magneto-optical disk.

[Explanation of symbols]

5 ... Ideal value calculation control circuit, 7 ... Pattern read circuit, 8
... ideal value determination circuit, 21 ... precoder, 22 ... magneto-optical disk, 23 ... PR equalization circuit, 24 ... A / D converter, 2
5 ... PLL circuit, 26 ... bit by bit decoder, 4
1, 51 ... Adder, 42, 52 ... Delay circuit, 61 ... Ternary circuit, 62 ... Binarizing circuit, 76 ... Viterbi decoder, 9
1, 92 ... Status register, 93, 94 ... Data series register, 95 ... Status operation circuit, 96 ... Judgment circuit, T ... Track, S _{1 to SN} ... Sector, ID ... Identification code recording area, TP, TP1, TP2 ... training pattern area, DT, DT1, DT2 ... user data recording area.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G11B 19/04 501 G11B 19/04 501H 20/14 341 9463-5D 20/14 341B

Claims (4)

[Claims] 1. A disc-type recording medium in which user data is recorded in a first area, for calculating an ideal amplitude value which is reproduced prior to reproduction of the user data and which determines the logic of the user data. A disk-type recording medium having a second area for recording a training pattern. 2. The disc-type recording medium according to claim 1, wherein the second area is provided in a head portion of the first area and in the first area. 3. The disc-type recording medium according to claim 1, wherein the second area is provided at a head portion of the first area, and the value of each bit of the user data recorded in the first area is A disc-type recording medium, wherein the second area is provided in the first area when the value is biased to one of "0" and "1". 4. A PR equalizer circuit for equalizing a reproduction signal of user data recorded in a first area of a disc type recording medium based on a partial response system, and a clock signal from an output signal of the PR equalizer circuit. A PLL circuit to generate,
An A / D converter that samples the output signal of the PR equalization circuit based on the clock signal to generate sample data, and decodes the sample data of the user data based on a preset ideal amplitude value to reproduce data. In the reproducing device including a decoding circuit for generating the data, a training pattern for reproducing an ideal amplitude value for determining the logic of the user data is recorded on the disk-type recording medium prior to the reproduction of the user data. And a control circuit that outputs a control signal during reproduction of the training pattern, and sample data of the training pattern output from the A / D converter during output of the control signal. A reproducing apparatus provided with an ideal value determining circuit for determining the ideal amplitude value and giving it to the decoding circuit. .