JP2003051163A - Method for evaluating signal quality and reproducing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform simply and accurately evaluation of signal quality performed by obtaining the standard deviation of frequency distribution for the path metric difference of two lines in a correct solution state of trellis diagram in a PRML decoding process. SOLUTION: In PRML decoding process of a reproduced signal from a magneto-optical disk 1, path metric difference is obtained with a path metric calculating circuit 6, a viterbi decoding circuit 7 and a threshold value register 8. The relative frequencies of one side part in which the frequency distribution of the path metric difference is sectioned by using the prescribed threshold value is obtained with a comparator 9, counters 10, 11 and a dividing device 12. The quality of a reproduced signal is evaluated with a controller 13 on the basis of this relative frequencies.

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to PRML (Partia
The present invention relates to a signal quality evaluation method and a reproducing apparatus capable of highly accurately evaluating the reproduced signal quality in an optical reproducing apparatus of the Response Maximum Likelihood system with a simple circuit configuration.

[0002]

2. Description of the Related Art Conventionally, jitter has often been used as an evaluation value of reproduction signal quality of an optical disk.
The PRML method is being adopted as a data detection method for realizing higher density recording. In such a situation, the jitter indicating the variation in the time axis direction is not appropriate as the evaluation value. Although the bit error rate of the data detection result by PRML is also used as an evaluation value, the required number of measurement sample bits is large,
There are many disadvantages such as being easily affected by defects due to scratches on the disc.

Against this background, SAM (Sequen
A method called ced Amplitude Margin) for reproducing signal quality evaluation has been proposed (T. Perkins, A Window Margin L
ikeProcedure for Evaluating PRML Channel Performan
ce; IEEE Transactions on Magnetics, Vol.31, No2, 19
95, p1109-1114).

First, the concept of SAM will be described with reference to FIGS. Here, (1,7) RLL (Run Length
Limited) PR the reproduced signal of the bit string recorded with the code
The case of performing PRML detection based on the (1, 2, 1) characteristic will be described as an example.

An ideal reproduced signal waveform of a 1T mark without distortion and noise according to the PR (1,2,1) characteristic has a sample level ratio of 1: 2: 1 for each channel clock as shown in FIG. Become. The reproduction signal waveform of the mark of 2T or more is obtained by superposing the reproduction signal waveform of the 1T mark, and for example, 1: 3: for a 2T mark.
For 3: 1 and 3T marks, 1: 3: 4: 3: 1 and 4T
If it is a mark, it will be 1: 3: 4: 4: 3: 1.

In this way, an ideal reproduction signal waveform is assumed for an arbitrary bit string, and the ideal sample level takes five levels of 0, 1, 2, 3, and 4. Here, for convenience, the sample level is normalized so that the maximum amplitude is ± 1. At this time, the ideal sample level is five levels of -1, -0.5, 0, +0.5, and +1.

Viterbi decoding is used as a method for specifically implementing PRML decoding. In Viterbi decoding,
Consider a trellis diagram as shown in FIG. In FIG. 14, S (00), S (01), S (10), and S (11) represent states. For example, state S (00) indicates that the previous bit was 0 and the current bit was 0. A line that connects states is called a branch and represents a state transition. For example, S (00)
A bit string of "001" can be represented by a branch of S (01).

In FIG. 14, the letters a to f are assigned as the identifiers of the branches, and the ideal waveform level expected in each state transition is added next to the letters. For example, a
Represents a bit string of “000”, so −1 and b are “10”.
Since it represents a bit string of "0", -0.5 is the ideal level. Here, S (01) → S (10) and S (10) →
The branch S (01) does not exist is (1,7)
In the RLL code, due to the run length limitation of d = 1, “01
This reflects that a bit string of "0" and "101" cannot exist.

In the trellis diagram, considering all combinations of branches (which are called paths) generated from any state through any state corresponds to considering all possible bit strings. Therefore, comparing the ideal waveforms expected for all paths with the reproduced waveforms actually reproduced from the optical recording medium, if the path with the ideal waveform with the closest waveform, that is, the Euclidean distance is the smallest, is searched, The most likely path can be determined as the correct path.

The procedure of Viterbi decoding using the trellis diagram will be concretely described with reference to FIG. At any time, there are two paths in states S (00) and S (11), S
One path merges with (01) and S (10). State where two paths merge S (00) and S (1
Regarding 1), if one having a smaller Euclidean distance between the ideal waveform of each path and the reproduction signal waveform is left as a surviving path, one path reaching each of the four states at any time, a total of 4 paths is obtained. This means that the book path remains.

The square of the Euclidean distance between the ideal waveform of the path and the reproduced signal waveform is called the path metric, and the branch metric obtained as the square of the difference between the ideal sample level of the branch and the sample level of the reproduced waveform constitutes the path. Calculated by accumulating over all branches.

The sample level of the reproduced signal waveform at time t is X [t], and the branch metrics of the branches a, b, c, d, e, f at time t are Ba respectively.
[T], Bb [t], Bc [t], Bd [t], Be
[T], Bf [t], and each state S (0
0), S (01), S (10), and S (11), the path metrics of the surviving paths are M (00) [t],
M (01) [t], M (10) [t], M (11)
If described as [t], the branch metric is calculated according to the equations (3) to (6), and the path metric is calculated according to the equations (7) to (10). M (00)
The process of selecting the smaller path metric in [t] and M (11) [t] corresponds to the determination of the surviving path. Ba [t] = (X [t] +1) ² (3) Expression Bb [t] = Bc [t] = (X [t] +0.5) ² (4) Expression Bd [t] = Be [ t] = (X [t] −0.5) ² (5) Expression Bf [t] = (X [t] −1) ² (6) Expression M (00) [t] = Min {M ( 00) [t-1] + Ba [t], M (10) [t-1] + Bb [t]} (Min {m, n} = m (if m ≦ n); n (if m> n)) (7) Formula M (01) [t] = M (00) [t-1] + Bc [t] ... (8) Formula M (10) [t] = M (11) [t-1] + Bd [ t] (9) Formula M (11) [t] = Min {M (01) [t-1] + Be [t], M (11) [t-1] + Bf [t]} (Min {m, n} = m (if m ≦ n); n (if m> n)) Equation (10) When the procedure for determining the surviving path is repeated every time the sample value of the reproduced signal waveform is input, As the paths with large metrics are selected, the paths gradually converge to one. By setting this as the correct path, the original data bit string is correctly reproduced.

Here, considering the conditions under which the Viterbi decoding is correctly performed, the path that finally converges into one path becomes the correct path. Therefore, in the process of determining the surviving path at each time, the correct path is obtained. Must be smaller than the path metric of the other path that is the wrong path. This condition is expressed by equations (11) to (14) according to the correct answer bit string.

(When the correct bit string is “... 000”) ΔM = (M (01) [t−1] + Bb [t]) − (M (00) [t−1] + Ba [t])> 0 (11) Equation (when the correct answer bit string is “... 100”) ΔM = (M (00) [t−1] + Ba [t]) − (M (01) [t−1] + Bb [t ])> 0 (12) (when the correct bit string is "... 011") ΔM = M (11) [t-1] + Bf [t])-(M (01) [t-1] + Be [t])> 0 (13) (when the correct bit string is “... 111”) ΔM = M (01) [t−1] + Be [t]) − (M (11) [t−1] ] + Bf [t])> 0 Equation (14) (when the correct answer bit string is "... 001" or "... 110") Since the survivor path is always correctly determined, Δ
M> 0 holds.

In equations (11) to (14), ΔM
Is the difference between the path metrics of the two paths that bet on each other for the survival, and this difference is called SAM. SAM> 0 is required to prevent an error, and SAM
The larger the value, the less likely it is that an error will occur.

In order to evaluate the reliability of the system using the SAM value, it is necessary to macroscopically evaluate the distribution state of the entire SAM value calculated at each time. Japanese Unexamined Patent Application Publication No. 10-21651 proposes a method of inspecting the reliability of a reproducing device using the standard deviation of the frequency distribution of SAM values as an evaluation value.

FIG. 15A is a frequency distribution graph of the SAM value obtained from the reproduced signal of the (1,7) RLL code pattern actually recorded on the magneto-optical disk. As can be seen from this result, the SAM distribution has two peaks. This is because the Euclidean distance between the correct path and the incorrect path differs depending on the bit pattern when obtaining the SAM value for all reproduced signals.

Therefore, as shown in FIG.
The SAM distribution in an ideal reproduced signal with no noise obtained from the (1,7) RLL code sequence is 1.5, 2.5, 3.
It takes discrete ideal values such as 5, 4.5, 5, 6, 6, 7, 8, and 9. The frequency of the ideal value is different because the number of types of bit patterns that are the ideal values is different,
7) This is because the appearance frequency of each bit pattern is different in the RLL code string. Since various noises are present in the actual reproduced signal, these ideal values have variations, and as a result, a distribution shape in which a plurality of distributions are overlapped as shown in FIG.

Since the SAM distribution has such characteristics and is a distribution which is largely different from the normal distribution, even if the standard deviation is simply obtained from this distribution, the correlation with the bit error rate is small.

[0020]

Therefore, Japanese Unexamined Patent Publication (Kokai) No. Hei 10
According to Japanese Patent Laid-Open No. 21651, only a bit pattern having a high SAM value <0 due to noise and having a SAM ideal value of 1.5 is selected to generate a SAM distribution, and a standard deviation thereof is calculated. This is PRML
The circuit becomes complicated because it means that a sequence of monitoring a pattern of a plurality of data bits of a decoding result and obtaining a SAM value only when the pattern is determined to be a specific pattern is required. There are drawbacks. Further, in order to obtain the standard deviation, it is necessary to calculate all the squared errors between the individual SAM values and the SAM average value, which causes a problem that the load on the circuit is heavy.

[0021]

In order to solve the above-mentioned problems, a signal quality evaluation method of the present invention comprises a step of reproducing a recording medium and a PRML of a reproduced signal from the recording medium.
In the decoding process, a step of obtaining a path metric difference between two paths that are input to the correct state of the trellis diagram, and a step of obtaining a relative frequency of one side portion obtained by dividing the frequency distribution of the path metric difference by a predetermined threshold value. , Evaluating the quality of the reproduced signal based on the relative frequency.

According to the above invention, it is possible to provide a signal quality evaluation method capable of detecting and evaluating the quality of a reproduced signal simply and highly accurately.

That is, the signal quality evaluation method of the present invention is
In the PRML decoding process of decoding the reproduced signal from the recording medium, the path metric difference between the two paths input to the correct state of the trellis diagram is obtained, and the frequency distribution of the path metric difference is divided by a predetermined threshold on one side. The quality of the reproduced signal is evaluated based on the relative frequency of the part.

Therefore, unlike the conventional case, it is not necessary to select only a bit pattern having a predetermined ideal value to calculate the SAM value. That is, since the signal quality evaluation method does not include a step of monitoring a pattern of a plurality of data bits and determining whether or not the pattern is a specific pattern, it should be performed using a device with a simple circuit. You can

The frequency distribution of the path metric difference has a small spread when the signal quality is good (noise is small), and the signal quality is poor (noise is large).
In some cases, the spread becomes large. That is, the relative frequency of the one side portion obtained by dividing the frequency distribution of the path metric difference by the predetermined threshold value, that is, the relative frequency of the portion below the predetermined threshold value corresponds to the spread of the frequency distribution. Therefore, the relative frequency reflects the magnitude of noise and corresponds to the signal quality. Therefore, by evaluating the quality of the reproduction signal based on the relative frequency, the quality of the reproduction signal can be detected with high accuracy.

In order to solve the above-mentioned problems, the reproducing apparatus of the present invention inputs the correct state of the trellis diagram in the PRML decoding process of the reproducing means for reproducing the recording medium and the reproduced signal reproduced by the reproducing means. The path metric difference detecting means for obtaining the path metric difference between the two paths, the relative frequency detecting means for obtaining the relative frequency of one side of the path metric difference frequency distribution divided by a predetermined threshold, and the relative frequency And a signal quality evaluation means for evaluating the quality of the reproduced signal.

According to the above invention, it is possible to provide a signal reproducing apparatus which can detect the quality of a reproduced signal with high accuracy and which is constituted by a simple circuit.

That is, the path metric difference detecting means is
P of the reproduced signal from the recording medium reproduced by the reproducing means.
In the RML decoding process, the path metric difference between the two paths input to the correct state of the trellis diagram is obtained, and only the bit pattern having a predetermined ideal value is not selected as in the conventional case. In other words, the path metric difference detection means is for obtaining the path metric difference between the two paths, monitors a pattern of a plurality of data bits, and judges whether or not the pattern is a specific pattern. Therefore, it can be configured by a simple circuit.

Further, the relative frequency detecting means obtains the relative frequency of one side of the path metric difference frequency distribution divided by a predetermined threshold value. Then, the signal quality evaluation means evaluates the quality of the reproduced signal based on the relative frequency obtained by the relative frequency detection means.

Here, as described above, the frequency distribution of the path metric difference changes according to the signal quality.
For this reason, the relative frequency of the one side portion obtained by dividing the frequency distribution of the path metric difference by a predetermined threshold reflects the magnitude of noise and corresponds to the signal quality. Therefore, the quality of the reproduced signal can be detected with high accuracy by evaluating the quality of the reproduced signal based on the relative frequency.

The recording medium of the reproducing apparatus is an optical recording medium, and in addition to the above configuration, a reproducing power changing means for changing the reproducing power of the light beam and a reproducing signal reproduced by the reproducing means at each reproducing power. Optimal reproduction power determination means for determining the optimum reproduction power based on the signal quality evaluated by the signal quality evaluation means may be provided.

According to the above-mentioned invention, the optimum reproduction power can be accurately obtained by a simple circuit configuration, and a test read time which is much larger than that in the case where the bit error rate is directly evaluated and the test read is performed. Shortening can be realized.

That is, the reproducing apparatus further includes reproducing power changing means for changing the reproducing power of the light beam and optimum reproducing power determining means for determining the optimum reproducing power. Then, for each reproduction power changed by the reproduction power changing means, the reproduction signal evaluation means obtains the signal quality of the reproduction signal reproduced by the reproduction means, and the optimum reproduction power determination means determines the optimum reproduction power based on the signal quality. Is to determine.

Since the signal quality is an evaluation value that corresponds very well to the bit error rate, the optimum reproduction power can be accurately determined based on the signal quality. Further, in order to obtain the signal quality by the reproduction signal evaluation means, a large number of measurement bits as the bit error rate is not required, so that the test read time is significantly longer than that in the case of directly evaluating the bit error rate. It can be shortened.

Therefore, it is possible to provide a reproducing apparatus which can accurately obtain the optimum reproducing power and can realize a great reduction of the test read time.

Further, it is preferable that the optimum reproducing power determining means of the reproducing apparatus determines the center value of the reproducing power range in which the quality of the reproduced signal is better than a predetermined reference value as the optimum reproducing power.

According to this, even when the optimum reproducing power is changed due to the inclination or temperature change of the disk which is the optical recording medium, the risk of the bit error rate becoming an extremely bad value can be suppressed to a low level. .

The recording medium of the reproducing apparatus is an optical recording medium, and in addition to the above configuration, recording power changing means for changing the recording power of the light beam, and recording means for recording a test pattern at each recording power. And an optimum recording power determining means for determining an optimum recording power based on the signal quality evaluated by the signal quality evaluating means for the reproduced signal of the recorded test pattern reproduced by the reproducing means. It may be

According to the above invention, the optimum recording power can be accurately obtained with a simple circuit configuration.
Compared with the case where the bit error rate is directly evaluated and the test write is performed, the test write time can be significantly shortened.

That is, in the reproducing apparatus, the recording power changing means for changing the recording power of the light beam, the recording means for recording the test pattern at each recording power, and the optimum recording power determining means for determining the optimum recording power. And further have. Then, at each recording power changed by the recording power changing means, with respect to the reproduction signal obtained by reproducing the test pattern recorded in the recording means by the reproduction means, the signal quality is obtained by the reproduction signal evaluation means,
The optimum recording power determining means determines the optimum recording power based on the signal quality.

Since the signal quality is an evaluation value that corresponds very well to the bit error rate, the optimum recording power can be accurately determined based on the signal quality. Further, the measurement of the signal quality by the reproduction signal evaluation means does not require a large number of measurement bits as much as the bit error rate. Therefore, the test write time can be significantly shortened as compared with the case of directly evaluating the bit error rate.

Therefore, it is possible to provide a reproducing apparatus capable of accurately obtaining the optimum recording power and further realizing a great reduction in test write time.

Further, it is preferable that the optimum recording power determining means determines the center value of the recording power range in which the quality of the reproduced signal is better than a predetermined reference value as the optimum recording power.

According to this, even if the optimum recording power is changed due to the inclination of the optical recording medium, the temperature change, etc., the risk of the bit error rate becoming an extremely bad value can be suppressed to a low level. .

In addition to the above structure, the reproducing apparatus is
Servo means for optimizing the servo offset based on the signal quality evaluated by the signal quality evaluation means and performing servo control on the reproduction signal reproduced by the reproduction means may be provided.

As a result, the optimum servo offset can be accurately obtained with a simple circuit configuration, and the processing can be performed significantly more than in the case where the bit offset is directly evaluated and the servo offset is optimized. The time can be shortened.

That is, the reproducing apparatus further includes servo means for optimizing the servo offset, and the reproduction signal evaluating means calculates the signal quality of the reproduction signal reproduced by the reproducing means. Based on this, the optimum servo offset is determined.

Here, since the signal quality is an evaluation value that corresponds very well to the bit error rate, when the servo offset is optimized based on the signal quality instead of the conventional bit error rate. However, very stable servo control can be performed on the reproduced signal.
Furthermore, in the servo offset optimization process based on the signal quality, a large number of measurement bits used in the optimization process based on the bit error rate is not required,
The processing time can be significantly shortened as compared with the case where the servo offset is optimized based on the bit error rate.

Therefore, it is possible to provide a reproducing apparatus in which a servo offset that is accurately optimized can be obtained with a simple circuit configuration and the processing time required for optimizing the servo offset is significantly shortened.

In addition to the above construction, the reproducing apparatus is
A waveform equalizing means may be provided for optimizing the equalization coefficient based on the signal quality evaluated by the signal quality evaluating means and for equalizing the waveform of the reproduced signal reproduced by the reproducing means.

As a result, it is possible to accurately obtain the optimum equalization coefficient with a simple circuit configuration, and it is possible to make a large comparison with the case where the bit error rate is directly evaluated and the equalization coefficient is optimized. Moreover, the processing time can be shortened.

That is, the reproducing apparatus obtains the signal quality of the reproduced signal reproduced by the reproducing means by the reproduced signal evaluation means, determines the optimum equalization coefficient based on the signal quality, and determines the optimum equalization coefficient. It further has waveform equalization means for waveform equalizing the reproduction signal with the equalized coefficient.

Since the signal quality is an evaluation value that corresponds very well to the bit error rate, even when the equalization coefficient is optimized based on the signal quality instead of the bit error rate, It is possible to obtain a reliable optimized equivalence coefficient. Further, since the equalization coefficient optimization processing based on the signal quality does not require a large number of measurement bits used in the optimization processing based on the bit error rate, the equalization coefficient is optimized based on the bit error rate. The processing time can be significantly shortened as compared with the case of performing.

Therefore, it is possible to provide the reproducing apparatus in which the optimum equalization coefficient can be accurately obtained with a simple circuit structure and the processing time required for the optimization of the equalization coefficient is greatly shortened.

In addition to the above construction, the reproducing apparatus is
Tilt servo means for performing tilt correction of the recording medium based on the signal quality evaluated by the signal quality evaluation means for the reproduction signal reproduced by the reproduction means may be provided.

As a result, the tilt correction can be accurately performed with a simple circuit configuration, and the tilt correction time can be significantly shortened as compared with the case where the tilt correction is performed by directly evaluating the bit error rate. be able to.

That is, the reproducing apparatus further includes tilt servo means for obtaining the signal quality of the reproduction signal reproduced by the reproducing means by the reproduction signal evaluating means and performing tilt correction based on the signal quality.

Here, since the signal quality is an evaluation value that corresponds very well to the bit error rate, even if the tilt correction is performed based on the signal quality instead of the bit error rate, the tilt correction is accurately performed. It can be performed. Furthermore, since the tilt correction processing based on the signal quality does not require a large number of measurement bits used in the tilt correction processing based on the bit error rate, the tilt correction processing compared to the case where tilt correction is performed based on the bit error rate is performed. The time can be greatly reduced.

Therefore, it is possible to provide a reproducing apparatus in which the tilt correction can be accurately performed with a simple structure and the processing time required for the tilt correction is significantly shortened.

Further, in the reproducing apparatus, the modulation method of the recording medium has a run length limit code in which d = 1, and the P
The impulse response of the isolated mark assumed by the path metric difference detecting means for performing RML decoding is represented as (a, 2a, a), and when the PRML decoding considers the run length limitation, a predetermined value for obtaining the relative frequency is given. Threshold is 1.6
or it may be a a ² or more 2.4a ² or less.

This makes it possible to simultaneously suppress the error caused by the deviation between the SAM frequency distribution and the normal distribution and the influence of the defect, so that at least one of highly accurate test read and test write can be realized. .

That is, the playback device is RLL (Run
Length Limited) Reproduction signal of bit string recorded with code
The impulse response of the isolated mark in the decoding process of
(A, 2a, a), and the PRML recovery
The numbering process considers run-length restrictions and
The predetermined threshold value for obtaining the number is 1.6a²2.4 above
a ²Within the following range. This allows the SAM frequency
The error caused by the deviation between the normal distribution and the distribution
It is possible to reduce the effect of
At least high-precision test read and test write
One can be realized.

Further, the reproducing apparatus sets the predetermined threshold value to SL and sets the desired bit error rate reference value to BER.
, The mode of the frequency distribution of the path metric difference is μ, the total number of bits for obtaining the path metric difference is N, and the number of patterns in which the ideal value of the path metric difference is the minimum in all bit series is If n,

[0064]

[Equation 3]

From the equation (1), σ is obtained, and this σ is given by the equation (2).

[0066]

[Equation 4]

The value obtained by substituting in may be used as the predetermined reference value.

As a result, it is possible to realize at least one of more reliable test read and test write.

That is, if the noise that causes variations in the SAM values obtained as the path metric difference is close to white noise, the frequency distribution of the individual SAM values can be approximated to a normal distribution. Therefore, for a portion smaller than 1.5 which is the minimum value of the SAM ideal value, 1.
It can be approximated to a normal distribution having a value close to 5 as the mode value μ. At this time, the standard deviation σ representing the variation of the approximated normal distribution and the bit error rate BER
And 1 have a one-to-one correspondence, and this relationship is represented by equation (1).

Further, since σ corresponding to BER is obtained from the equation (1), using this σ, the portion of the reproduced signal having the quality equivalent to the desired BER, which is smaller than an arbitrary threshold value SL in the SAM distribution, can be obtained. The relative frequency of can be calculated by the equation (2).

Therefore, by substituting σ obtained by substituting the BER required for the reproducing apparatus into the equation (1) from the viewpoint of the error correction capability of the reproducing apparatus, the BE
The SAM relative frequency corresponding to R can be obtained. Then, by using the SAM relative frequency as a predetermined reference value for determining the optimum reproduction power or recording power, at least one of highly reliable test read and test write can be realized.

Also, in the reproducing apparatus, when the modulation method of the optical recording medium is the run length limited code in which d = 1, "00111" and "0001" in the entire bit sequence.
The number of patterns "1", "11000" or "11100" may be n.

As a result, the modulation method of the optical recording medium is d =
When the run length limited code is 1, "00111", "00011", "110" in all bit sequences
The parameter conversion between the SAM frequency distribution and the approximate normal distribution can be accurately performed based on the appearance probability of the pattern "00" or "11100". Therefore, the SAM relative frequency for a desired bit error rate can be derived with high accuracy, and at least one of highly reliable test read and test write can be performed.

In addition to the above-mentioned structure, the reproducing apparatus is
It is also possible to provide a frequency detection means for detecting the frequency of the path metric difference equal to or less than the mode value and use a numerical value obtained by doubling the detected frequency as n.

In this case, the portion of the SAM frequency distribution with the mode value μ or less is approximated to a normal distribution having a mode value of about 1.5, and twice the frequency with the mode value μ or less is the normal distribution. It substantially matches the frequency n of. Therefore, even in the case of a special pattern in which the appearance frequencies of the four types of patterns obtained in advance by random patterns are significantly different from the actual appearance frequencies, it is complicated to detect the appearance frequencies of the four types of patterns based on the decoded bit string. With a simple configuration without performing various processing,
It is possible to accurately obtain the coefficient K and perform highly reliable signal quality evaluation.

In addition to the above configuration, the reproducing apparatus may include a mode value detecting means for detecting the mode value of the frequency distribution of the path metric difference.

Since the mode of the frequency distribution of the path metric difference can be detected in real time by the mode detecting means, the reference value corresponding to the mode changing depending on the type of noise is obtained. Is possible,
The reference value can be determined more accurately.

[0078]

DESCRIPTION OF THE PREFERRED EMBODIMENTS [First Embodiment] The following will describe one embodiment of the present invention with reference to FIGS. FIG. 1 is a block diagram when the present invention is applied to a magneto-optical disk reproducing apparatus. As shown in FIG. 1, the magneto-optical disk reproducing apparatus of the present embodiment includes a magneto-optical disk 1, a semiconductor laser 2 and a photo diode. Diode 3, recovered clock extraction circuit 4, A / D converter 5, path metric calculation circuit 6, Viterbi decoding circuit 7, threshold value register 8, comparator 9, counter 10, counter 11, divider 12,
The controller 13 is provided.

The semiconductor laser 2, the photodiode 3, the reproduction clock extraction circuit 4, and the A / D converter 5 function as the reproduction means of the present invention, and the path metric calculation circuit 6 and the Viterbi decoding circuit 7 are the main functions. The function as the path metric difference detecting means of the invention is realized by the above threshold register 8,
The comparator 9, the counter 10, the counter 11, and the divider 12 have the function as the relative frequency detecting means of the present invention, and the controller 13 has the function as the signal quality evaluating means of the present invention.

A reproducing operation by the magneto-optical disk reproducing apparatus having the above structure will be described.

First, when the semiconductor laser 2 irradiates the magneto-optical disk 1 with a light beam, the reflected light is converted into an electric signal by the photodiode 3 and output as a reproduction signal. This reproduction signal is converted into digital data by the A / D converter 5, and then the path metric calculation circuit 6
Entered in.

The A / D conversion here is performed by PLL (Ph
This is performed at the timing of the clock extracted from the reproduction signal by the reproduction clock extraction circuit 4 composed of an ase locked loop). The path metric calculation circuit 6 calculates the path metric as in the conventional example.

That is, according to the equations (3) to (10), the path is formed by the square of the difference (branch metric) between the digital data of the input reproduction signal and the ideal level of each branch of the trellis diagram. Accumulation processing is performed for all branches.

That is, the sample level of the reproduced signal waveform at time t is X [t], branches a, b, c, d,
The branch metrics of e and f at time t are Ba [t], Bb [t], Bc [t], Bd [t] and B, respectively.
e [t], Bf [t], each state S (0
0), S (01), S (10), and S (11), the path metrics of the surviving paths are M (00) [t],
M (01) [t], M (10) [t], M (11)
If described as [t], the branch metric is calculated according to the equations (3) to (6), and the path metric is calculated according to the equations (7) to (10). M (00)
The process of selecting the smaller path metric in [t] and M (11) [t] corresponds to the determination of the surviving path. Ba [t] = (X [t] +1) ² (3) Expression Bb [t] = Bc [t] = (X [t] +0.5) ² (4) Expression Bd [t] = Be [ t] = (X [t] −0.5) ² (5) Expression Bf [t] = (X [t] −1) ² (6) Expression M (00) [t] = Min {M ( 00) [t-1] + Ba [t], M (10) [t-1] + Bb [t]} (Min {m, n} = m (if m ≦ n); n (if m> n)) (7) Formula M (01) [t] = M (00) [t-1] + Bc [t] ... (8) Formula M (10) [t] = M (11) [t-1] + Bd [ t] (9) Formula M (11) [t] = Min {M (01) [t-1] + Be [t], M (11) [t-1] + Bf [t]} (Min {m, n} = m (if m ≦ n); n (if m> n)) Equation (10) When the procedure for determining the surviving path is repeated every time the sample value of the reproduced signal waveform is input, As the paths with large metrics are selected, the paths gradually converge to one. By setting this as the correct path, the original data bit string is correctly reproduced.

The path metric calculated each time the digital data of the reproduced signal is input is input to the Viterbi decoding circuit 7, where the path having the minimum path metric finally remains as the surviving path, and the decoded bit sequence is obtained. Is obtained. The decoded bit sequence is input to the path metric calculation circuit 6, and the correct state can be known by referring to this decoded bit sequence. Therefore, according to the equations (11) to (14),
The SAM value is obtained as the path metric difference ΔM between the two paths input in the correct state.

(When the correct bit string is “... 000”) ΔM = (M (01) [t−1] + Bb [t]) − (M (00) [t−1] + Ba [t])> 0 (11) Expression (when the correct answer bit string is “... 100”) ΔM = (M (00) [t−1] + Ba [t]) − (M (01) [t−1] + Bb [t ])> 0 (12) (when the correct bit string is "... 011") ΔM = (M (11) [t-1] + Bf [t])-(M (01) [t-1] + Be [t])> 0 Equation (13) (when the correct bit string is "... 111") ΔM = (M (01) [t-1] + Be [t])-(M (11) [t −1] + Bf [t])> 0 (14) Expression (when the correct answer bit string is “... 001” or “... 110”) Since the survivor path is always correctly determined, Δ
M> 0 holds.

The processing up to this point is almost the same as the conventional example. However, only the bit pattern whose SAM ideal value is 1.5, which was indispensable in the conventional example, is selected and the SAM is selected.
The configuration for performing the calculation is unnecessary in the present invention.

Now, the SAM value as the path metric difference ΔM output from the path metric calculation circuit 6 is compared with the predetermined threshold value SL stored in the threshold value register 8 by the comparator 9. The comparator 9 has ΔM <S
When L, that is, the SAM value is smaller than a predetermined threshold value, one pulse is output. Since this pulse is input to the counter 10, the output of the counter 10 represents the number of SAM values smaller than the predetermined threshold value.

On the other hand, the clock output from the reproduction clock extraction circuit 4 is also input to the counter 11, and since this 1 clock corresponds to 1 bit of the reproduction signal,
The output of the counter 11 represents the total number of bits of the reproduced signal. Therefore, the counter 1 calculated by the divider 12
The result of dividing the output of 0 by the output of the counter 11 is SA
In the M frequency distribution, the relative frequency of the portion smaller than the predetermined threshold SL (ratio to the total frequency) is shown.
The controller 13 including a CPU or the like can evaluate the quality of the reproduced signal based on the relative frequency.

The reason why the relative frequency corresponds to the signal quality will be described with reference to FIG. FIG. 2 is a graph of a SAM frequency distribution obtained by the reproducing device for a reproduced signal having the same number of bits in (a) good signal quality and (b) poor signal quality, and the horizontal axis represents SAM. value,
The vertical axis represents frequency.

As is clear from the figure, when the signal quality is good, that is, the noise is small, the spread of the distribution is small, and therefore the predetermined threshold value SL shown by hatching in FIG.
When the relative frequency of the following part is small and the signal quality is poor, that is, when the noise is large, the distribution spread is large, and therefore the relative frequency of the part below the shaded threshold SL in FIG. Has become. After all, since the relative frequency of the portion below SL reflects the spread of the distribution, that is, the magnitude of noise, the relative frequency corresponds to the signal quality.

By the way, in order to accurately obtain the SAM value, it is necessary to know the correct answer bit string (recording data pattern). In the above-described embodiment, the correct bit string necessary for obtaining the SAM value is obtained from the Viterbi decoding result, but strictly speaking, the Viterbi decoding result does not completely match the correct bit string because it contains a decoding error.

However, since the bit error rate to be evaluated is about 1E-3 at the worst, the influence of the decoding error is very small. Further, if an appropriate threshold value is selected for the following reason, there is almost no effect on the SAM relative frequency.

The SAM value calculation when a decoding error occurs is the one in which the sign of ΔM is reversed from the definitions of equations (11) to (14). That is, when a decoding error occurs, ΔM <0, but since the wrong path is regarded as the correct path, the polarity is reversed, and −ΔM is set as the SAM value (since the own decoding result is the correct answer, , Which means that the SAM value ≥ 0 always holds).

However, normally, since the skirt portion of the normal distribution centered on the SAM ideal value 1.5 is 0 or less and an error occurs, even if the SAM value <0, the absolute value |
The SAM value | is not so large. Therefore, SAM value <predetermined threshold value SL holds for most of the SAM values corresponding to the error bits, and therefore SL of the SAM frequency distribution
The SAM relative frequency, which is the integral of the following part, is hardly affected.

As described above, it can be said that there is almost no effect even when the SAM relative frequency is calculated using the Viterbi decoding result of itself, but when more strict evaluation is desired,
When it is desired to evaluate the state in which the bit error rate is extremely bad, a known data pattern is recorded in advance when the signal quality is evaluated, and the SAM value calculation is performed by referring to the data pattern. I don't mind.

As described above, in the magneto-optical disk reproducing apparatus, unlike the conventional example, a complicated structure for selecting only a bit pattern in which the SAM value has a predetermined ideal value and calculating the SAM value is not required, Only a simple circuit consisting of a comparator and a counter makes it possible to detect the reproduced signal quality easily and with high accuracy.

In the present embodiment, the reproduction signal quality is detected on the basis of the relative frequency equal to or less than SL of the SAM frequency distribution for the predetermined threshold value SL, but the present invention is not limited to this. Not a thing. For example, for the second threshold value SL2 equal to or less than SL, the S
Even when using a relative frequency in the range of L2 or more and SL or less,
There is essentially no significant difference, and the reproduced signal quality can be detected in the same manner as above.

[Second Embodiment] The following will describe another embodiment of the present invention with reference to FIGS. In the present embodiment, constituent elements having the same functions as those of the constituent elements in the first embodiment are designated by the same reference numerals, and the description thereof will be omitted.

FIG. 3 is a block diagram when the present invention is applied to a magneto-optical disk reproducing apparatus, and FIG. 4 is a flow chart for explaining a test read operation by this reproducing apparatus.

The magneto-optical disk reproducing apparatus of the present embodiment comprises a magneto-optical disk 1, a semiconductor laser 2, a photodiode 3, a reproduction clock extraction circuit 4, an A / D converter 5, a path metric calculation circuit 6, and a Viterbi decoding circuit. 7, the threshold register 8, the comparator 9, the counter 10, the counter 11, the divider 12, and the controller 13 are the same as those in the first embodiment. In the present embodiment, the laser power control circuit 1 for controlling the drive current of the semiconductor laser 2 is further provided.
It is equipped with 4.

The laser power control circuit 14 has a function as a reproducing power changing means of the present invention, and the controller 13 has a function as an optimum reproducing power determining means of the present invention.

A test read operation by the magneto-optical disk reproducing apparatus having the above structure will be described.

In step S1, the controller 13 sets the drive current of the conductor laser 2 to a predetermined initial value via the laser power control circuit 14. The semiconductor laser 2 is irradiated onto the magneto-optical disk 1 with the reproduction power set to the initial value, and the reflected light is reproduced into digital data through the photodiode 3 and the A / D converter 5 (step S2).
After the frequency distribution of the SAM value is obtained by the path metric calculation circuit 6, the reproducing operation is performed until the relative frequency of the portion smaller than the predetermined threshold SL is output from the divider 12 (step S3). The same as the first embodiment.

The controller 13 at step S4
The relationship between the set reproduction power and the obtained SAM relative frequency is stored. Subsequently, in step S5, the reproduction power is set higher by a predetermined amount, and then it is determined whether or not the power upper limit for performing the test is exceeded (step S6). If the upper limit is not exceeded yet, steps S2 to S5 are performed. The process of is repeated.

As a result, the SAM relative frequencies for the plurality of reproduction power values changed in the predetermined step are stored as a table. Eventually, when the reproduction power exceeds the test range in step S6, the process proceeds to step S7 and S
A reproduction power range in which the AM relative frequency is smaller than a predetermined reference value is obtained by the controller 13, and finally the center value of the reproduction power range is determined as the optimum reproduction power in step S8.

FIG. 5 shows a graph of the result of actual measurement of the relationship of the SAM relative frequency with respect to each reproducing power, which is obtained by the reproducing operation. Here, the threshold SL for obtaining the SAM relative frequency is SL = 0.5, and the test range of the reproducing power is Pr0, Pr1, ..., Pr8. Also, the bit error rate measured at the same reproducing power is shown by the dotted line graph with respect to the SAM relative frequency shown by the solid line graph. However, in this bit error rate actual measurement result, the number of measurement bits that is one digit or more larger than the SAM relative frequency is used because it is necessary to reduce the measurement error.

From this result, the reproduction power is set to the initial value Pr0.
It can be seen that the SAM relative frequency gradually decreases with increasing from, and then gradually increases with Pr4 as the boundary. Furthermore, it can be seen that this change corresponds to the bit error rate. That is, it can be said that the SAM relative frequency represents the signal quality that corresponds very well to the bit error rate.

Therefore, when 0.005 is used as an appropriate reference value of the SAM relative frequency, the reproduction power range below this reference value is obtained as PrL to PrH, so the optimum reproduction power is the center value ( PrL + PrH)
/ 2 (reproduction power represented by a solid arrow in FIG. 5). This sets the reference value of the bit error rate to 1E-4.
It is a value very close to the center power (reproduction power shown by the dotted arrow in FIG. 5) obtained as, and the bit error rate is extremely poor even if the optimum reproduction power changes due to the tilt of the disk or temperature change. The risk of becoming a value can be kept low.

As described above, in the magneto-optical disk reproducing apparatus, the SAM relative frequency is used as the evaluation value that corresponds very well to the bit error rate.
It is possible to accurately obtain the optimum reproduction power with a simple circuit configuration without requiring a large number of measurement bits as the bit error rate.

Particularly, in the case where the bit error rate is directly evaluated and the test read is performed, the measurement bit for one track of one reproduction power is required. Therefore, the reproduction power change in the test range × the disc rotation time. Only takes test lead time. On the other hand, if the test read is performed by the SAM relative frequency, the reproduction power can be changed for each sector arranged in a plurality of tracks. For this reason, the test read can be completed in only one rotation time of the disk, and a significant reduction in time can be realized.

In the above embodiment, the SAM
The center value of the reproduction power range in which the relative frequency is less than or equal to the predetermined reference value is determined as the optimum reproduction power.
The reproduction power that minimizes the M relative frequency may be determined as the optimum reproduction power. For example, in the case of FIG. 5, Pr4 may be set as the optimum power.

[Third Embodiment] The following will describe still another embodiment of the present invention with reference to FIGS. In the present embodiment, constituent elements having the same functions as those of the constituent elements in the first embodiment are designated by the same reference numerals, and the description thereof will be omitted.

FIG. 6 is a block diagram when the present invention is applied to a magneto-optical disk recording / reproducing apparatus, and FIG. 7 is a flow chart for explaining a test write operation by this recording / reproducing apparatus.

The magneto-optical disk recording / reproducing apparatus of the present embodiment comprises a magneto-optical disk 1, a semiconductor laser 2, a photodiode 3, a reproduction clock extraction circuit 4, an A / D converter 5,
The path metric calculation circuit 6, the Viterbi decoding circuit 7, the threshold value register 8, the comparator 9, the counter 10, the counter 11, the divider 12, and the controller 13 are the same as those in the first embodiment. The present embodiment further includes a laser power control circuit 14, which controls the drive current of the semiconductor laser 2, a test pattern generation circuit 15, and a magnetic head 16.

The laser power control circuit 14 has a function as a recording power changing means of the present invention, the test pattern generating circuit 15 and the magnetic head 16 have a function as a recording means of the present invention, and the controller 13 has a main function. It has a function as an optimum recording power determining means of the invention.

A test write operation by the magneto-optical disk recording / reproducing apparatus having the above configuration will be described.

At step S10, the controller 13
The drive current of the semiconductor laser 2 is set to a predetermined initial value via the laser power control circuit 14. When the semiconductor laser 2 is irradiated to the magneto-optical disk 1 with the recording power set to the initial value and the magnetic head 16 is driven by the test pattern output from the test pattern generating circuit 15, the magneto-optical disk 1 is driven. A test pattern is magneto-optically recorded on the disk (step S11).

Here, since the optimum recording power can be obtained with higher accuracy by reflecting the influence of cross write and cross talk from the adjacent tracks, in addition to recording the test pattern on the test track,
It is desirable to record a different test pattern on both adjacent tracks.

Next, in step S12, the recording power is set higher by a predetermined amount, and then it is judged whether or not the upper limit of the power to be tested is exceeded (step S13). The processing from S11 to S12 is repeated. As a result, a plurality of test patterns for each recording power are recorded.

When the recording power exceeds the test range in step S13, the process proceeds to step S14, the laser power control circuit 14 returns the drive current of the semiconductor laser 2 to an appropriate reproducing power, and then the recording is performed for each recording power. The read test pattern is read from the magneto-optical disk 1, reproduced as digital data through the photodiode 3 and the A / D converter 5, and the frequency distribution of the SAM value is obtained by the path metric calculation circuit 6, and then the predetermined pattern is determined. The relative frequency of the portion smaller than the threshold value SL is output from the divider 12 (step S
The reproducing operation up to 15) is the same as that in the first embodiment.

After storing the relationship between the set recording power and the obtained SAM relative frequency in step S16, the controller 13 determines whether or not all the test patterns for each recording power have been reproduced (step S17). ), If there are still unreproduced test patterns, step S14
The process from S16 to S16 is repeated. As a result, the SAM for the plurality of recording power values changed in the predetermined step
The relative frequencies are stored as a table.

When all the test patterns are reproduced in step S17, the process proceeds to step S18, the recording power range in which the SAM relative frequency becomes smaller than a predetermined reference value is obtained by the controller 13, and finally the step S1 is performed.
At 9, the center value of the recording power range is determined as the optimum recording power.

FIG. 8 shows a graph of the result of actual measurement of the relationship of the SAM relative frequency with respect to each recording power, which is obtained by the above recording operation. Here, the threshold value SL for obtaining the SAM relative frequency is SL = 0.5, and the recording power test range is Pw0, Pw1, ..., Pw6.

Also, the bit error rate measured at the same recording power with respect to the SAM relative frequency shown by the solid line graph is shown by the dotted line graph. However, as in the second embodiment, the measurement result of the bit error rate uses a measurement bit number that is one digit or more larger than the SAM relative frequency because it is necessary to reduce the measurement error.

From this result, it can be seen that the relative frequency gradually decreases as the recording power is increased from the initial value, and gradually increases at Pw3. Furthermore, it can be seen that this change corresponds to the bit error rate. That is, it can be said that the SAM relative frequency represents the signal quality that corresponds very well to the bit error rate.

Therefore, when 0.014 is used as an appropriate reference value of the SAM relative frequency, the reproduction power range below this reference value is obtained as PwL to PwH, so the optimum recording power is the center value ( PwL + PwH)
/ 2 (recording power indicated by a solid arrow in FIG. 5). This sets the reference value of the bit error rate to 1E-3.
The value is extremely close to the center power (recording power shown by the dotted arrow in FIG. 5) when calculated as, and the bit error rate is extremely bad even if the optimum recording power changes due to the tilt of the disk or temperature change. The risk of becoming a value can be kept low.

As described above, in the magneto-optical disk reproducing apparatus, the SAM relative frequency, which is the relative frequency of the SAM distribution, is used as the evaluation value that corresponds very well to the bit error rate. As a result, the optimum recording power can be accurately obtained with a simple circuit configuration without requiring a large number of measurement bits as much as the bit error rate.

Particularly, when the bit error rate is directly evaluated and the test write process is performed, it is necessary to provide the measurement bits for one track of one recording power. For this reason, recording and reproduction must be repeated each time the recording power is changed, and the number of rotations that is four times the recording power to be changed (3 rotations for recording and 1 rotation for reproduction) is required, which is extremely long for test writing. It takes time.

On the other hand, if the test write is performed with the SAM relative frequency, the signal quality can be accurately evaluated with a small number of measurement bits. Therefore, for example, if the recording power is changed in units of a plurality of sectors existing in one track, the test write can be completed only by one recording and reproduction, that is, only four rotations. Significant reduction in time can be realized.

In the above embodiment, the SAM
The center value of the recording power range in which the relative frequency is less than or equal to the predetermined reference value is determined as the optimum recording power.
The recording power that minimizes the M relative frequency may be determined as the optimum recording power. For example, in the case of FIG. 8, Pw3 may be the optimum power.

In the second and third embodiments, SA
The reproducing power or the recording power is optimized based on the M relative frequency, but the present invention is not limited to this, and the SAM relative is also used for other parameters that cause deterioration of the reproduced signal quality. By optimizing these parameters based on the frequency, the same effect as described above can be obtained.

It is well known that the quality of the reproduced signal changes according to the servo offset state of the tracking servo, the focus servo, etc. (Y. Tanaka, E.
valuation of a 120mm Sized Magneto optical Disk Sy
stem of Over 6GB Capacity; Japanese Journal of Appl
ied Physics, Vol.37, No.4B, 1998, p2150-2154, or Fujimoto et al. "Application of PR (1,2,1) ML Viterbi Decoder to Magneto-Optical Recording"; Proceedings of the 5th Son
y Research Forum, 1995, p465-469). Therefore, SAM
If these servo offsets are optimized so that the reproduction signal quality is improved based on the relative frequency, an accurate signal quality evaluation can be realized with a small number of measurement bits by a simple circuit configuration, and as a result, The servo offset optimization process can be performed accurately, and the processing time can be greatly shortened.

In addition to this, it is well known that the quality of the reproduced signal changes depending on the equalization coefficient at the time of waveform equalization of the reproduced signal (Fujimoto et al. "PR (1,2,1) ML system. Application of Viterbi Decoder to Magneto-optical Recording "; Proceedings of
the 5th Sony Research Forum, 1995, p465-469). Therefore, by optimizing the equalization coefficient of the waveform equalization so that the reproduction signal quality is improved based on the SAM relative frequency as in the above, accurate signal quality can be obtained with a small number of measurement bits by a simple circuit configuration. You can make an evaluation and, as a result,
It is possible to realize accurate optimization processing of the equalization coefficient and greatly shorten the processing time.

Further, it is well known that the reproduction signal quality is deteriorated also by the tilted state of the recording medium (Fujimoto et al. "Pr (1,2,1) ML Viterbi Decoder for magneto-optical recording). Application ”; Proceedings of the 5th Sony
Research Forum, 1995, p465-469). Therefore, with respect to the tilt of the recording medium, the tilt correction is performed based on the SAM relative frequency so that the reproduction signal quality is improved in the same manner as described above. It is possible to perform quality evaluation, and as a result, it is possible to realize accurate tilt correction processing and a significant reduction in tilt correction processing time.

In addition to the above, it becomes a factor of deteriorating the reproduction signal quality such as optical pulse waveform control of a light beam (so-called write strategy) performed in an optical disc device of a phase change medium, alignment of various optical parts and the like. It is needless to say that the same effect as described above can be obtained by the optimization processing based on the SAM relative frequency of the present invention described above for all the obtained parameters.

Now, let us consider the predetermined reference value used in the second to third embodiments. As described with reference to FIG. 15 in the conventional example, the frequency distribution of SAM values has a distribution shape in which a plurality of SAM ideal values have variations due to noise, and thus a plurality of distributions overlap each other. If the noise is close to the white noise, each distribution can be approximated to a normal distribution. Therefore, for a portion smaller than the minimum value of SAM ideal value of 1.5, the normal distribution having a value close to 1.5 as the mode value μ. It can be approximated to.

At this time, the standard deviation σ representing the variation of the approximated normal distribution and the bit error rate BER have a one-to-one correspondence, and this relationship is expressed by the equation (15). FIG. 9 is a graph in which the actual measurement result (solid line) of the SAM frequency distribution in the actual optical disk reproducing device and the normal distribution of σ (dotted line) corresponding to the bit error rate are superimposed.

[0139]

[Equation 5]

The latter half of the right side of the equation (15) is known in statistics as a distribution function obtained as the integral of the probability density function of the normal distribution, and the normal distribution determined by the mode μ and the standard deviation σ. , 0 represents the relative frequency of 0 or less.

On the other hand, since an error bit is generated when the SAM value <0, the bit error rate is S
It is considered to be equal to the ratio of the part of 0 or less to the total frequency of the AM frequency distribution. Therefore, the value obtained by multiplying the relative frequency of 0 or less of the normal distribution by the coefficient K of the parameter conversion matches the bit error rate. Specifically, the coefficient K is the total frequency of the SAM frequency distribution is N, and the pattern in which the SAM ideal value is the minimum in all the measured bit sequences, that is, 1.5 (the distribution generated only by the SAM value of this pattern is about When the number of normal distributions having a mode value of 1.5) is n, then K = n / N.

The modulation method is, for example, (1,7) RLL.
As described above, if the run length limited code is d = 1,
The pattern in which the SAM ideal value is 1.5 can be specified by examining all paths and searching for a pattern in which the square of the Euclidean distance between the correct path and the incorrect path is 1.5.

Specifically, "00111", "0001"
The four types of patterns "1", "11000", and "11100" correspond to this. Therefore, when the modulation method is the run-length limited code with d = 1, the appearance probabilities of the above four types of patterns in the pattern obtained by modulating the random data are obtained in advance, or from the bit string decoded by the reproducing device. The coefficient K can be specifically and accurately obtained by directly detecting the appearance frequencies of the four types of patterns.

Further, in addition to the method of obtaining the number n of patterns having an SAM ideal value of 1.5, a means for detecting the frequency less than or equal to the mode value μ in the SAM frequency distribution is additionally provided, and the detected frequency is By using a doubled numerical value as n, the same effect as above can be obtained by the method of specifically obtaining the coefficient K. This is SA
Since the part of the M frequency distribution with the mode value μ or less approximates to the normal distribution with the mode value of about 1.5, twice the frequency with the mode value μ or less almost matches the frequency n of the normal distribution. It is based on the fact that

As a result, in the case where the data pattern for which the SAM is obtained is a special pattern that is not a random pattern, that is, when the appearance frequencies of the above-mentioned four types of patterns obtained by random patterns in advance differ greatly from the actual appearance frequency. Also, the coefficient K can be accurately obtained with a simple configuration without performing a complicated process of detecting the appearance frequencies of the four types of patterns based on the decoded bit string.

Since the standard deviation σ corresponding to the bit error rate BER can be obtained from the equation (15),
By using this σ, the relative frequency of the portion of the reproduced signal having the quality equivalent to the desired bit error rate, which is smaller than the arbitrary threshold value SL in the SAM frequency distribution, can be obtained by the equation (2).

[0147]

[Equation 6]

For example, if the bit error rate required for the reproducing device is 1E-4 or less from the viewpoint of the error correction capability of the reproducing device, the optimum reproducing power and the optimum recording power determined by the test read and the test write are: It is desirable to obtain it as the center value of the power range where the bit error rate becomes 1E-4 or less. Therefore, BER = 1E-
If the corresponding SAM relative frequency is obtained from equations (15) and (2) as 4, and the numerical value is used as a reference value for obtaining the power range in the above second and third embodiments, high reliability is obtained. It is possible to realize test read and test write.

Next, the predetermined threshold value SL used in the first to third embodiments will be considered. In the above consideration, it is premised that the part smaller than 1.5 which is the minimum value of the SAM ideal value can be approximated to the normal distribution having the mode value μ close to 1.5.

However, as can be seen from FIG. 9, although this holds true for the portion close to 0 of the distribution, the SAM ideal value = 2.
Since the distributions of 5 are mixed, it gradually becomes unrealizable. This means that if the predetermined threshold value SL is increased to some extent or more, the error in the calculation of the SAM relative frequency calculated by the equations (15) and (2) becomes large.

FIGS. 10A to 10F show the SAM when the threshold SL is changed, using the same actual measurement result as in FIG.
It is a graph of the relationship between the relative frequency (vertical axis) and the bit error rate (horizontal axis). Each plot point in the graph represents each relationship between the SAM relative frequency and the bit error rate measured by variously changing the recording power. Further, what is indicated by a dotted line is a theoretical relationship between the SAM relative frequency and the bit error rate, which is calculated by the equations (15) and (2).

From this graph, when the threshold value SL is close to 0, the actual measurement result and the calculation result match exactly, but as the threshold value SL is increased, the error between the actual measurement result and the calculation result is increased. I understand that it will occur. This error is small when SL ≦ 0.6, but begins to increase when SL ≧ 0.7, so the threshold SL is set to 0.6.
The error can be suppressed to be small by the following.

On the other hand, if the threshold value SL is too small,
Since the number of SAM values equal to or less than SL in the SAM frequency distribution becomes too small, it is likely to be affected by the defect, which is a drawback similar to the bit error rate.

FIGS. 11A to 11F show FIGS.
In the same measurement result as (f), 0.1% of the total number of bits
FIG. 10 is a graph showing the relationship between the SAM relative frequency and the bit error rate when a certain degree of defect is mixed, and the meaning of the graph is the same as in FIG. From this graph, it can be seen that the closer the threshold SL is to 0, the larger the error between the actual measurement result and the calculation result.

Generally, from the viewpoint of the error correction capability of the reproducing apparatus, the bit error rate required of the reproducing apparatus is required to be 1E-3 or less, that is, 1E-3 or less even if it is bad. The standard of signal quality used in the test write is also 1E-3 or less. Therefore, it is necessary that the error in the relationship between the SAM relative frequency and the bit error rate is suppressed to be small in at least this range.

As can be seen from FIGS. 11A to 11D,
If the bit error rate is 1E-3 or less, the error is SL ≧
Although it is small at about 0.4, it starts to increase at SL ≦ 0.3. Therefore, by setting the threshold value SL to be 0.4 or more, it is possible to suppress the influence of the defects.

When the above examination results are combined, the threshold value SL
It is desirable that the range of 0.4 ≦ SL ≦ 0.6. However, this value is
The ideal sample level of PR (1,2,1) characteristic is
It is a numerical value under conditions normalized to -1, -0.5, 0, +0.5, +1.

When generalized and the PR characteristic is specified by the impulse response (a, 2a, a), the ideal sample levels are 0, a, 2a, 3a, 4a. At this time, SA
The value corresponding to the M ideal value of 1.5 is the pattern "0001
It is calculated as the square of the Euclidean distance between the ideal waveforms of "1" and "00111".

Since the ideal waveform of the pattern "00011" is (0, a, 3a) and the ideal waveform of the pattern "00111" is (a, 3a, 4a), the Euclidean distance is (a-0) ² + (3a- a) ² + (4a-3a) ² =
6a ² .

Therefore, 0.4 is 6a ² × (0.4 / 1.
5) = 1.6a ² and 0.6 is 6a ² × (0.
6 / 1.5) = 2.4a ² . After all, 1.6
By setting a ² ≦ SL ≦ 2.4a ² , it is possible to simultaneously suppress the effects of errors and defects caused by the deviation of the SAM frequency distribution and the normal distribution, and realize highly accurate test reading and test writing.

[Fourth Embodiment] The following description will explain still another embodiment of the present invention with reference to FIG.

As for the mode value μ, if the noise is white noise, it matches the SAM ideal value, so μ = 1.5.
The above formula may be calculated with a fixed value of, but in practice, the mode changes due to the influence of colored noise. Therefore, in addition to the configuration of the reproducing apparatus according to the first embodiment, a configuration in which the apparatus directly detects the mode value of the SAM frequency distribution is added, and the mode value thus detected is used to (15), ( By calculating the equation (2), the reference value can be determined more accurately. FIG. 12 shows this configuration.

The path metric calculation circuit 6 shown in FIG.
The controller 13 is the same as that of the first embodiment, and other configurations similar to those of the first embodiment are omitted. In the present embodiment, further, the comparator 17 that detects S0 or more and less than S1, the comparator 18 that detects S1 or more and less than S2, ..., Sk−1 or more S
A counter 19, a counter 21, a counter 21, ..., A counter 22, and a maximum value detection circuit 23 for detecting that the value is less than k are provided, and these configurations function as the mode detection means of the present invention. have. In addition, S
0, S1, ..., Sk represent numerical value sequences at equal intervals.

When the magneto-optical disk 1 (see FIG. 1) (not shown) is reproduced by this reproducing apparatus, the path metric difference ΔM is output from the path metric calculation circuit 6 by the same processing as in the first embodiment. The path metric difference ΔM is simultaneously input to the comparator 17, the comparator 18, ..., The comparator 19, and it is determined whether or not ΔM is within a predetermined range.

The counter that receives the output of the comparator that is determined to be within the predetermined range is incremented. For example, if S1 ≦ ΔM <S2, the counter 21 that receives the output of the comparator 18 is incremented. By repeating this process, the number within each range of the SAM frequency distribution is counted.

Then, when the calculation of the path metric difference ΔM for all the measurement target bits is completed, the maximum value detection circuit 23 selects the maximum output from the counter 20, counter 21 ,. Select and output the center value of the corresponding range as the mode value μ.

For example, if the output of the counter 21 is maximum, μ = (S1 + S2) / 2. Controller 13
Performs the calculation based on the equations (4) and (5) using the input mode value μ to determine the reference value of the SAM relative frequency.

As described above, since the reference value of the SAM relative frequency is determined based on the mode detected in real time,
Always correspond to the mode that changes depending on the type of noise,
It is possible to determine the reference value more accurately.

In the description of the above embodiment,
(1,7) RLL as the run length limited code of d = 1
Although symbols are used, it is needless to say that they are not limited to these.

Further, in the description of the above embodiment,
Although the magneto-optical disk reproducing device has been described as an example of the reproducing device, the reproducing device is not limited to this, and PR
The effect should be equally exerted in a device for reproducing a signal of the ML system. That is, the present invention can be applied to all of the phase change type optical disk devices, magnetic recording devices, communication data receiving devices, and the like.

[0171]

As described above, the signal quality evaluation method and reproducing apparatus of the present invention do not require a complicated structure for selecting only a bit pattern in which the SAM value has a predetermined ideal value and calculating the SAM value. It is a thing. Therefore, it is possible to detect the reproduction signal quality simply and with high accuracy by using a simple circuit configuration.

Further, the reproducing apparatus of the present invention uses the SAM relative frequency as an evaluation value that corresponds very well to the bit error rate. Therefore, there is an effect that the optimum reproduction power can be accurately obtained with a simple circuit configuration without requiring a large number of measurement bits as much as the bit error rate. Further, there is an effect that the test read time can be significantly reduced as compared with the case where the bit error rate is directly evaluated and the test read is performed.

Further, the reproducing apparatus of the present invention may determine the center value of the reproducing power range in which the SAM relative frequency is equal to or lower than the predetermined reference value as the optimum reproducing power in the test read. Therefore, there is an effect that the risk of the bit error rate becoming an extremely bad value can be suppressed to a low level even if the optimum reproduction power changes due to the tilt of the disc, the temperature change, or the like.

Further, the reproducing apparatus of the present invention uses the relative frequency of the SAM distribution as an evaluation value that corresponds very well to the bit error rate. Therefore, without requiring as many measurement bits as the bit error rate,
With the simple circuit configuration, it is possible to obtain the optimum recording power accurately. Further, there is an effect that the test write time can be significantly shortened as compared with the case of performing the test write by directly evaluating the bit error rate.

Further, the reproducing apparatus of the present invention may determine the center value of the recording power range in which the SAM relative frequency is equal to or less than the predetermined reference value as the optimum recording power in the test write. Therefore, it is possible to reduce the risk of the bit error rate becoming an extremely bad value even if the optimum recording power changes due to the inclination of the disk, the temperature change, or the like.

Further, the reproducing apparatus of the present invention is provided with the servo means for optimizing the servo offset and performing the servo on the reproduced signal reproduced by the reproducing means based on the signal quality evaluated by the quality evaluating means. May be. Therefore, by using the SAM relative frequency as an evaluation value that corresponds very well to the bit error rate, it is possible to use a SAM relative frequency without requiring a large number of measurement bits as the bit error rate.
With a simple circuit configuration, it is possible to accurately obtain the optimized servo offset, and it is possible to significantly reduce the processing time compared with the case where the servo offset optimization processing is performed by directly evaluating the bit error rate. Produce an effect.

Further, the reproducing apparatus of the present invention optimizes the equivalent coefficient of the reproduced signal reproduced by the reproducing means on the basis of the signal quality evaluated by the signal quality evaluating means, and waveform-equivalents the reproduced signal. Waveform equalizing means may be provided. Therefore, by using the SAM relative frequency as an evaluation value that corresponds very well to the bit error rate, it can be accurately optimized by a simple circuit configuration without requiring a large number of measurement bits as the bit error rate. It is possible to obtain the equalization coefficient, and further, it is possible to significantly reduce the processing time as compared with the case where the bit error rate is directly evaluated and the equalization coefficient is optimized.

Further, the reproducing apparatus of the present invention is provided with the tilt servo means for correcting the tilt of the recording medium on the basis of the signal quality evaluated by the signal quality evaluating means for the reproduced signal reproduced by the reproducing means. Good. Therefore, as an evaluation value that corresponds very well to the bit error rate, S
By using the AM relative frequency, the tilt of the recording medium can be accurately corrected with a simple circuit configuration without requiring a large number of measurement bits as the bit error rate, and the bit error rate can be directly measured. As compared with the case where the tilt correction is performed after evaluation, the tilt correction time can be significantly shortened.

Further, in the reproducing apparatus of the present invention, the modulation method of the recording medium is a run length limited signal in which d = 1,
The impulse response of the isolated mark assumed by the path metric difference detecting means for PRML decoding is represented as (a, 2a, a), and when the PRML decoding considers the run length, it is a predetermined value for obtaining the SAM relative frequency. Threshold value is 1.
It may be 6a ² or more and 2.4a ² or less. Therefore, it is possible to achieve high-accuracy test reading and test writing by simultaneously suppressing the effects of errors and defects caused by the deviation between the SAM frequency distribution and the normal distribution.

Further, in the reproducing apparatus of the present invention, the SAM relative frequency corresponding to the desired bit error rate is calculated using the theoretically derived formula, and the numerical value is used for the power range in the test read and test write. It may be used as a reference value of. As a result, it is possible to achieve highly reliable test read and test write.

In the reproducing apparatus of the present invention, the modulation method is d.
When the run length limited code is = 1, the predetermined reference value may be obtained based on the appearance probability of the specific pattern. As a result, the SAM frequency distribution and the approximate normal distribution can be accurately converted, so that the SAM relative frequency corresponding to the desired bit error rate can be derived with high accuracy, and highly reliable test leads can be obtained. ,
The effect that the test light can be realized is achieved.

Further, in the reproducing apparatus of the present invention, the frequency detecting means may detect the frequency of the path metric difference equal to or less than the mode value, and use a value obtained by doubling the detected frequency as n. Therefore, if the frequency below the mode is 2 in the SAM frequency distribution,
Since the parameter conversion of the SAM frequency distribution and the approximate normal distribution can be accurately performed based on the multiplied value, the SAM relative frequency corresponding to the desired bit error rate can be derived with high accuracy with a simple configuration. The effect is that a highly reliable signal quality evaluation can be realized.

Further, the reproducing apparatus of the present invention comprises mode value detecting means for detecting the mode value of the frequency distribution of the path metric difference. Therefore, since the reference value of the SAM relative frequency can be determined based on the mode detected in real time, the reference value can be determined more accurately by always corresponding to the mode changing depending on the type of noise. There is an effect that it becomes possible.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a schematic configuration of a magneto-optical disk reproducing device according to a first embodiment of the present invention.

2A and 2B are graphs for explaining the relationship between the relative frequency of SAM frequency distribution and signal quality.

FIG. 3 is a configuration diagram showing a schematic configuration of a magneto-optical disk reproducing device according to a second embodiment of the present invention.

FIG. 4 is a flowchart illustrating a test read operation of the magneto-optical disk reproducing apparatus of FIG.

5 is an explanatory diagram showing a graph of an actually measured result of a test read operation of the magneto-optical disk reproducing apparatus of FIG.

FIG. 6 is a configuration diagram showing a schematic configuration of a magneto-optical disk reproducing device according to a third embodiment of the present invention.

7 is a flowchart illustrating a test write operation of the magneto-optical disc reproducing apparatus in FIG.

8 is an explanatory diagram showing a graph of actual measurement results of a test write operation of the magneto-optical disk reproducing apparatus of FIG.

FIG. 9 is a graph in which the actual measurement result of the SAM frequency distribution and the normal distribution corresponding to the bit error rate are overlaid.

10A to 10F are explanatory diagrams showing graphs of actual measurement results and theoretical calculation results of the relationship between the SAM relative frequency and the bit error rate.

11A to 11F are explanatory diagrams showing graphs of actual measurement results and theoretical calculation results of the relationship between the SAM relative frequency and the bit error rate when affected by a defect.

FIG. 12 is a configuration diagram of a magneto-optical disk reproducing device according to a fourth embodiment of the present invention.

FIG. 13 is a schematic diagram of a reproduction signal waveform according to a PR (1,2,1) characteristic.

FIG. 14 is a schematic diagram showing a trellis diagram.

FIG. 15 is a SAM frequency distribution graph with an actually measured waveform and an ideal waveform.

[Explanation of symbols]

1 Magneto-optical disk (recording medium) 2 Semiconductor laser (reproducing means) 3 Photodiode (reproducing means) 4 Reproduced clock extraction circuit (reproducing means) 5 A / D converter (reproducing means) 6 Path metric calculation circuit (path metric difference) Detection means) 7 Viterbi decoder (path metric difference detection means) 8 Threshold register (path metric difference detection means) 9 Comparator (relative frequency detection means) 10, 11 Counter (relative frequency detection means) 12 Divider ( Relative frequency detecting means) 13 Controller (signal quality evaluating means, optimum reproducing power determining means, optimum recording power determining means) 14 Laser power control circuit (reproducing power changing means, recording power changing means) 15 Test pattern generating circuit (recording means) 16 magnetic head (recording means) 17, 18, 19 comparator (mode value detection means) ) 20, 21, 22 Counter (mode value detection means) 23 Maximum value detection circuit (mode value detection means)

Front page continuation (51) Int.Cl. ⁷ Identification code FI theme code (reference) G11B 7/005 G11B 7/005 Z 5J065 7/09 7/09 A G 7/125 7/125 C 11/105 553 11 / 105 553B 553C 556 556A 586 586V 20/10 321 20/10 321Z 20/14 341 20/14 341B H03M 13/25 H03M 13/25 (72) Inventor Jun Akiyama 22-22 Nagaikecho, Abeno-ku, Osaka-shi, Osaka In-house (72) Inventor Shigeki Maeda 22-22 Nagaike-cho, Abeno-ku, Osaka-shi, Osaka F-term (in reference) DD03 EE20 FF42 5D118 AA14 AA18 BA01 BF02 CA01 CA08 CB05 CD08 CD11 5D119 AA09 BA01 DA05 HA16 HA54 5J065 AA01 AB01 AC03 AD10 AE06 AF02 AF03 AG05 AH05 AH08 AH15

Claims (14)

[Claims] 1. A method of reproducing a recording medium, a step of obtaining a path metric difference between two paths input to a correct state of a trellis diagram in a PRML decoding process of a reproduction signal from the recording medium, A signal quality characterized by including the step of obtaining the relative frequency of one side part of which the frequency distribution of the metric difference is divided by a predetermined threshold value, and the step of evaluating the quality of the reproduced signal based on the relative frequency. Evaluation methods. 2. A reproducing means for reproducing a recording medium, and a path metric difference for obtaining a path metric difference between two paths to be inputted in a correct state of a trellis diagram in a PRML decoding process of a reproduced signal reproduced by the reproducing means. A detection means; a relative frequency detection means for obtaining the relative frequency of one side part of the frequency distribution of the path metric difference divided by a predetermined threshold; and a signal quality evaluation means for evaluating the quality of the reproduced signal by the relative frequency, A reproducing device characterized by being provided. 3. The recording medium is an optical recording medium, and in addition to the above configuration, a reproduction power changing means for changing a reproduction power of a light beam, and a reproduction signal reproduced by the reproduction means at each reproduction power. The reproduction apparatus according to claim 2, further comprising: optimum reproduction power determination means for determining optimum reproduction power based on the signal quality evaluated by the quality evaluation means. 4. The optimum reproduction power determining means determines the center value of a reproduction power range in which the quality of the reproduction signal is better than a predetermined reference value, as the optimum reproduction power. Playback device. 5. The recording medium is an optical recording medium, and in addition to the above configuration, recording power changing means for changing the recording power of a light beam, recording means for recording a test pattern at each recording power, and the reproduction. An optimum recording power determining means for determining an optimum recording power based on the signal quality evaluated by the signal quality evaluating means for the reproduced signal of the recorded test pattern reproduced by the means. The reproducing device according to claim 2. 6. The optimum recording power determining means determines the center value of a recording power range in which the quality of a reproduced signal is better than a predetermined reference value as the optimum recording power. Playback device. 7. In addition to the above structure, a servo means is provided for optimizing a servo offset based on the signal quality evaluated by the signal quality evaluation means, and performing servo control on the reproduction signal reproduced by the reproduction means. The reproducing apparatus according to claim 2, wherein the reproducing apparatus is provided. 8. In addition to the above configuration, a waveform equalizing means for optimizing the equalization coefficient based on the signal quality evaluated by the signal quality evaluating means and equalizing the waveform of the reproduced signal reproduced by the reproducing means is provided. The reproducing apparatus according to claim 2, wherein 9. In addition to the above structure, a tilt servo means for performing tilt correction of a recording medium based on a signal quality evaluated by the signal quality evaluation means for a reproduction signal reproduced by the reproduction means is provided. The reproducing apparatus according to claim 2, wherein 10. The modulation method of the recording medium is a run length limited code in which d = 1, and the impulse response of the isolated mark assumed by the path metric difference detection means for performing the PRML decoding is (a, 2a, a).
, And when PRML decoding considers run-length restrictions,
The predetermined threshold value for obtaining the relative frequency is 1.6a ²
The reproducing device according to any one of claims 2 to 9, characterized in that it is 2.4a ² or less. 11. The total number of bits for obtaining the path metric difference is SL, the predetermined threshold value is SL, the desired bit error rate reference value is BER, the mode of the frequency distribution of the path metric difference is μ. Let N be the number of patterns for which the ideal value of the path metric difference is the minimum in all bit sequences, then From the equation (1), σ is obtained, and this σ is given by the equation (2) 7. The reproducing apparatus according to claim 4 or 6, wherein a value obtained by substituting is used as the predetermined reference value. 12. When the modulation method of the optical recording medium is a run length limited code with d = 1, the number of patterns of “00111”, “00011”, “11000” or “11100” in all bit sequences is n. The reproducing apparatus according to claim 11, wherein: 13. In addition to the above configuration, a frequency detecting means for detecting a frequency of the path metric difference equal to or less than a mode value is provided, and a numerical value obtained by doubling the detected frequency is used as n. The playback device according to 1. 14. The reproducing apparatus according to claim 11, further comprising, in addition to the above configuration, a mode detecting means for detecting a mode of the frequency distribution of the path metric difference.