JP2002124042A - Information reproducing device and reproducing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the problem that an error occurs in data decoding when a reproduced signal is distorted due to asymmetry caused by the amplitude fluctuation of a reproduced signal generated by a formed pit too small or too large. SOLUTION: A circuit 8 to temporarily determine a equalization target value outputs temporarily determined state information (temporary determination information) to a nonlinear distortion compensator 12, and outputs a sample data row of waveform equalization corresponding to five equalization target values to a low pass filter(LPF) 10. The LPF 10 equalizes the inputted sample data row. A nonlinear distortion detector 11 detects the error between five equalization target values and corresponding average values from the LPF 10 as nonlinear distortion information. The nonlinear distortion compensator 12 corrects the waveform equalization data extracted from an adaptation waveform equalizer 7 by using a piece of nonlinear distortion information corresponding to a state of current time selected by using temporary determination information from the circuit 8 among five pieces of nonlinear distortion information from the nonlinear distortion detector 11.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information reproducing apparatus and a reproducing method, and more particularly to an information reproducing apparatus and method for decoding a reproduction signal of a recording medium such as an optical disk by using Viterbi decoding (PRML) utilizing a partial response characteristic. Reproduction method.

[0002]

2. Description of the Related Art In an information reproducing apparatus such as an optical disk apparatus, a PRML method has been frequently used as a method for decoding a reproduced signal reproduced from a recording medium with an increase in recording density. This PRML method is a decoding method that combines a partial response (PR) response and a Viterbi decoding method.

The outline of the Viterbi decoding method is as follows. A plurality of states are specified in advance according to a recording method for the recording medium, and a maximum likelihood transition between the plurality of states is selected by a calculation process based on a partial response response of a reproduction signal reproduced from the recording medium. Such a selection is made by an addition, comparison, and selection circuit in a Viterbi decoder that performs the Viterbi decoding method, and estimates the maximum likelihood of the number of state transitions equal to the number of states. Decoded data is generated corresponding to the state transition estimated by the maximum likelihood.

In order to obtain a partial response response of the reproduced signal, a waveform equalization process is performed. The waveform equalization process is performed under predetermined partial response characteristics determined in consideration of a spatial frequency and the like determined by the recording linear density. As described above, the Viterbi decoding method can be classified into several types depending on the number of a plurality of states specified in advance, the setting of the amplitude reference value in each of the plurality of states used in the calculation processing, and the like. Then, a Viterbi decoding method of a type suitable as much as possible to the partial response characteristic determined according to the recording linear density or the like is used.

[0005] The Viterbi decoding method is a method of calculating a likelihood based on an input reproduced signal value and performing decoding by selecting a state transition having the highest likelihood. In order to calculate the likelihood, an amplitude reference value defined by a partial response characteristic used in the waveform equalization processing is used as a value of an identification point of Viterbi decoding. For this reason, in order to lower the error rate of the decoded data by Viterbi decoding, the relationship between the input signal and the identification point is important.

If the reproduced signal is an ideal one having no non-linear distortion, the amplitude of the input signal is controlled by automatic gain control (AG
By keeping the constant C (automatic gain control) or the like, the relationship between the input signal and the discrimination point can be uniquely determined by the partial response characteristics. However, in the case of optical discs, the level of the reproduced signal differs from the discrimination point uniquely determined by the partial response characteristics due to the presence of nonlinear distortion such as a phenomenon called asymmetry in which the eye pattern loses vertical symmetry. , The decoding performance of Viterbi decoding is degraded.

Japanese Patent Laid-Open Publication No. Hei 10-283739 discloses that the value of the discrimination point that does not match the maximum amplitude of the reproduced signal is converted into a signal of a level corresponding to each discrimination point by a band-pass filter (BPF). An information reproducing device and a reproducing method for setting an identification point corresponding to a reproduction signal having asymmetry by extracting an identification signal from the detection of an envelope of the extracted signal are disclosed.

[0008]

However, in this conventional apparatus and method, the nonlinearity of the asymmetry cannot be removed at the discrimination point obtained from the envelope, and the linearity in Viterbi decoding cannot be satisfied.
The above problems cannot be solved.

When the recording medium is an optical disk, a reproduced signal is degraded due to fluctuations in signal amplitude due to birefringence of the substrate, thickness unevenness, reflectivity unevenness of the reflective film, etc., and intersymbol interference between pits before and after. . Also, when the pits formed due to the laser power during recording or the conditions during disk molding become too large or too small, the amplitude of the reproduced signal fluctuates, and as a result, asymmetry causes distortion in the reproduced signal. Or an offset in the reproduction signal due to an electrical offset in the reproduction system causes an error in data decoding.

In the above-described method for determining the value of the discrimination point that does not match the maximum amplitude of the reproduced signal, although the coincidence between the reproduced signal and the discrimination point is checked, the reproduced signal is distorted, offset, and the like due to the above-described causes. In this case, nonlinearity occurs in the relationship between the respective identification points, and the improvement in decoding performance of Viterbi decoding is small, and as a result, the decoding error rate may increase.

The present invention has been made in view of the above points,
An information reproducing apparatus and a reproducing method capable of improving the error rate of decoded data by performing correction for removing errors such as asymmetry in advance from a reproduction signal input to Viterbi decoding and improving the accuracy of Viterbi decoding. The purpose is to provide.

[0012]

In order to achieve the above object, an information reproducing apparatus according to the present invention comprises: reading means for reading a digital signal from a recording medium on which the digital signal is precoded and recorded; A bit clock is generated from the read digital signal, a sampling interpolator for sampling and interpolating the digital signal with the bit clock, and a digital signal taken out from the sampling interpolator based on an externally input equalization error. Adaptive waveform equalization means for performing waveform equalization of an adaptive partial response characteristic to output waveform equalization data; and a recording digital signal which is precoded with respect to the waveform equalization data output from the adaptive waveform equalization means. Calculating an equalization target value based on a predetermined state transition of the signal and a plurality of zero point information; The difference between the equalization target value and the waveform equalization data is calculated and supplied as an equalization error to the adaptive waveform equalization means, and a plurality of waveform equalization samples corresponding to a plurality of equalization target values based on state transitions Equalization target value provisional determination means for outputting a data string and information indicating a current transition state, and averaging means for averaging a plurality of waveform equalization sample data strings output from the equalization target value provisional determination means, respectively. A nonlinear distortion detecting means for detecting an error between an output signal of the averaging means and a plurality of equalization target values as nonlinear distortion information; and adaptively based on information indicating a transition state and nonlinear distortion information from the nonlinear distortion detecting means. And a nonlinear distortion correcting means for correcting and outputting the waveform equalized data from the waveform equalizing means.

According to another aspect of the present invention, there is provided an information reproducing method comprising: generating a bit clock from a signal read from a recording medium; and sampling and interpolating the signal read with the bit clock. And performing a waveform equalization of an adaptive partial response characteristic on the signal subjected to the sampling interpolation based on an equalization error input from the outside to output waveform equalized data.
And a predetermined state transition of the pre-coded recording digital signal with respect to the waveform
A third step of calculating an equalization target value based on the point information, calculating a difference between the equalization target value and the waveform equalization data and outputting the result as an equalization error, and receiving the waveform equalization data as input; A fourth step of outputting a plurality of waveform equalization sample data strings corresponding to a plurality of equalization target values based on state transitions and information indicating a current transition state, and averaging the plurality of waveform equalization sample data strings, respectively A fifth step of converting, a sixth step of detecting an error between the signal averaged in the fifth step and a plurality of equalization target values as nonlinear distortion information, information indicating a transition state, and nonlinear distortion information And correcting and outputting the waveform equalized data based on the second step.

According to the apparatus and the method of the present invention, the waveform equalization data obtained by performing the waveform equalization of the adaptive partial response characteristic based on the equalization error inputted from the outside is used for a plurality of equalization targets. The waveform equalization sample data sequence corresponding to the value is averaged, an error between each averaged signal and the corresponding equalization target value is detected as a nonlinear distortion amount, and a current transition state of the plurality of nonlinear distortion amounts is detected. The waveform equalization data is corrected by the amount of nonlinear distortion selected in accordance with, so that errors such as asymmetry are removed in advance from the digital signal reproduced from the recording medium. An identification point determined by the partial response characteristic and the signal amplitude level can be used as it is as the identification point for decoding.

[0015]

Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows a block diagram of an embodiment of an information reproducing apparatus according to the present invention. This embodiment is an optical disc reproducing apparatus, and an information signal recorded on the optical disc 1 by applying the PR method is constituted by a semiconductor laser diode (LD), a pin photodiode (PD), an optical system, and the like. Reproduced by the optical head 2.

The output side of the optical head 2 is
An amplifier 3 for amplifying the reproduction signal detected by the A / D converter, an A / D converter 4 for converting the reproduction signal into a digital signal,
An AGC circuit 5 for controlling the amplitude of the digital signal from the converter 4 to a predetermined level, a digital phase locked loop (DPLL) circuit 6 for generating a clock from a reproduction signal output from the AGC circuit 5, and an equalization target value An adaptive waveform equalizer 7 having a partial response characteristic which operates adaptively with respect to the above, an equalization target value provisional decision circuit 8, a subtractor 9,
Low pass filter (LPF) 10 and nonlinear distortion detector 11
, A nonlinear distortion corrector 12 and a Viterbi decoder 13.

The above-mentioned equalization target value provisional decision circuit 8 provisionally decides the equalization target value for the sample data from the output signal of the adaptive waveform equalizer 7 and the partial response characteristics, and based on the provisional decision result. A sample data string corresponding to each provisional judgment value is output. The LPF 10 averages the sample data sequence corresponding to the provisional determination value. The nonlinear distortion detector 11 detects nonlinear distortion of the output signal of the adaptive waveform equalizer 7 based on the signal output from the LPF 10. The nonlinear distortion corrector 12 corrects the nonlinear distortion of the output signal of the adaptive waveform equalizer 7 from the output signal of the nonlinear distortion detector 11. The Viterbi decoder 13 outputs the decoded data obtained by Viterbi decoding the output signal of the nonlinear distortion corrector 12 to a demodulation / error correction means (not shown) to perform error detection and error correction, thereby converting user data. Demodulate.

Next, the operation of signal reproduction according to the embodiment of the present invention will be described. The optical head 2 irradiates the optical disk 1 with laser light, and receives reflected light generated by the PD through an optical system in the optical head 2 to generate a reproduction signal. The reproduction signal output from the optical head 2 is adjusted in gain by the amplifier 3 and supplied to the A / D converter 4.
The reproduction signal is also sent to a focus error detection circuit and a tracking error detection circuit (not shown), which are respectively operated to generate a focus error signal and a tracking error signal, which are used for focus servo and tracking servo.

The A / D converter 4 samples the input reproduction signal based on the basic clock and generates sampling data, that is, digital data. The digital data is supplied to the DPLL circuit 6 after the data level is controlled to a predetermined level by the AGC circuit 5. D
The PLL circuit 6 receives the digital data output from the AGC circuit 5 as an input, generates a bit clock serving as a reference timing of the information reproducing apparatus from the input digital data, and further samples and interpolates the input digital data by the bit clock. The data is supplied to the adaptive waveform equalizer 7 after thinning interpolation.

The adaptive waveform equalizer 7 is a circuit that adaptively performs a waveform equalization process on an equalization target value with a partial response characteristic suitable for Viterbi decoding to be adapted. For example, as shown in the block diagram of FIG. As shown, it is composed of a transversal filter 71 and a multiplier / LPF 72 for adaptively changing the filter coefficient, and adaptively changes the filter coefficient based on the equalization error to equalize the waveform.

In the adaptive waveform equalizer 7, a signal output after being sampled and interpolated by the DPLL circuit 6 is received by a transversal filter 71 as an input signal.
Here, a filtering process of characteristics according to the filter coefficient from the multiplier / LPF 72 is performed, and the obtained signal is output to the equalization target value provisional determination circuit 8, the subtractor 9, and the nonlinear distortion corrector 12, respectively.

The equalization target value provisional decision circuit 8 compares the waveform equalization output data from the transversal filter 71 with an equalization target value determined by a partial response characteristic described later, and corresponds to the waveform equalization output data. The equalization target value is provisionally determined. The provisionally determined equalization target value is supplied to the subtractor 9 and is subtracted from the waveform equalization output data from the transversal filter 71 to obtain an equalization error as a difference value between them. The equalization error is supplied to a multiplier / LPF 72 in the adaptive waveform equalizer 7, where the multiplier multiplies the output from the transversal filter 71 to detect a correlation, and is integrated by the LPF. It is supplied to the transversal filter 71 as a filter coefficient that makes the equalization error zero.

Further, the equalization target value provisional judgment circuit 8 outputs a waveform equalization sample data sequence corresponding to the provisional judgment value. This waveform-equalized sample data string corresponds to the LPF 10 shown in FIG.
The average value of the variance is obtained by the formula (1), and is supplied to the nonlinear distortion detector 11.

The adaptive waveform equalizer 7, the tentative equalization target value judgment circuit 8 and the subtractor 9 constitute a feedback loop, and the tentative equalization target value judgment circuit 8 in the feedback loop performs partial response equalization. This circuit outputs a waveform equalization sample data sequence corresponding to the equalization target value by performing a tentative decision based on the premise. For example, the shortest bit length of a CD or the like is set to 3T (T is the bit period of data). System, the partial response characteristics PR (a, b, b,
When a) is applied, quinary state transition occurs, and five equalization target values and five waveform equalization sample data strings corresponding to the equalization target values are output.

Here, the virtual response (PR) characteristics will be further described. The user data is converted into a codeword as recording data by an appropriate encoding method according to the characteristics of the recording medium and the recording / reproducing method. In an optical disk device, a run length limited (RLL) coding method is often used in block coding. Generally, an m / n block code in which the number of consecutive '0's between' 1 'and' 1 'is at least d and at most k is called an RLL (d, k; m, n) code. .

For example, a DVD (Digital Versatile Dis
In the code adopted in (k), an 8/16 block encoding method (RLL (2) in which the number of consecutive '0's between' 1 'and' 1 'is at least two and at most ten. , 10; 8,
16) code). The DVD 8/16 block code is a coding method in which 8 bits are converted into 14 bits based on a conversion table, and 8 bits are converted into 16 bits as a whole by using 2 bits as a combination bit of each code.

A Viterbi decoding method can be used to decode a reproduced signal to be reproduced and obtain read data from data recorded by the combination of the RLL encoding method and the mark edge recording method. . The mark edge recording method is a recording method in which the inversion of the polarity at the boundary of each bit of the encoded recording data is expressed by the edge of each pit on the recording medium. Is performed, and pits are formed on the recording medium with respect to the precode output. On the other hand, for example, a pit is formed on the recording medium with respect to “1” of the encoded recording data, and “0” is formed.
A recording method that does not form pits is called a mark position recording method.

Here, in the RLL (2, 10) block code, the partial response characteristic PR (1, 1,
(1) will be described. First, the precode will be described. Regarding the signal recorded on the optical disk, in the mark edge recording method, prior to the actual recording on the optical disk, precoding based on the recording data encoded by the RLL encoding or the like is performed. Assuming that the recording data sequence at each time point t is x [t] and the precode output based on this is y [t], the precoding is performed as follows.

Y [t] = mod2 {x [t] + y [t−1]} (1) Such a precode output y [t] is actually recorded on the optical disc.

The adaptive waveform equalizer 7 performs a waveform equalization process on the partial response characteristic PR (1, 1, 1, 1). From this, the value of the ideal reproduction signal that matches the PR characteristic is represented by r [t], and the actual reproduction signal including noise (ie,
The reproduced signal reproduced from the recording medium) is denoted by n [t].

First, all possible states are specified based on the encoding method and the recording method for the recording medium.
Next, starting from each state at a certain time point, all possible state transitions at the next time point, and the recording data x [t] and the ideal reproduction signal value r [t] at the time of each state transition are specified. I do. All the specified states and state transitions, and “record data x [t]” when each state transition occurs
/ Ideal reproduced signal value r [t] ”in the form of a diagram is called a state transition diagram. FIG. 3 shows RLL (2, 10),
The state transition diagram in PR (1,1,1,1) is shown. Then, to perform a decoding operation based on this state transition diagram,
The Viterbi decoder 13 shown in FIG. 1 is configured.

Next, the partial response will be described in detail. The waveform equalization characteristic is PR (a, b, b, a) without normalizing the signal amplitude. PR (a, b, b,
In a), the contribution of the amplitude at the time t to the value of the reproduced signal at a certain time t is a of the amplitude of the signal at that time.
The amplitude contribution at times t-1 and t-2 is b times the signal amplitude at each time, and the amplitude contribution at time t-3 is a times the signal amplitude at that time. It is. Accordingly, the maximum value of the value of the reproduction signal is as follows when a pulse is detected from time t-3 to time t.

A + b + b + a = 2a + 2b The minimum value of the reproduced signal is zero. However, in the actual handling, the following, which is obtained by subtracting a + b of the DC component, is used as r [t]. Here, y [t] represents the precode output at the time point t as described above.

R [t] = a × y [t−3] + b × y [t−2] + b × y [t−1] + a × y [t] −ab (2) Therefore, noise is considered. Playback signal r [t] when not
Takes one of the values a + b, b, 0, -b, and -ab.

In the optical disk device of the present embodiment,
N subjected to waveform equalization under PR (a, b, b, a)
FIG. 4 shows an example of the eye pattern for [t]. From FIG. 4, the value of the reproduced signal n [t] at each time point has a variation due to noise, but is substantially a + b, b, 0, −.
b or -ab can be confirmed. These values of a + b, b, 0, -b, and -ab are used as equalization target values.

First, as a state used here, a state at a certain time t is defined as follows using the precode output before the time t and before. That is, h = a
[T], i = b [t-1], j = b [t-2], k = a
The state at the time [t-3] is defined as Shijk.

According to such a definition, 16 (= 2 ⁴ )
Although it is considered that there are individual states, the states that can actually occur are limited based on an encoding method or the like. RLL (2,
10) In the recording data sequence x [t] encoded as a code, at least two '0's are included between' 1 'and' 1 ', so two or more'1's are continuous. There is nothing. Based on such conditions imposed on the recording data sequence x [t], certain conditions are imposed on the precode output y [t], and the resulting states are restricted.

As described above, two or more '1's are included in the recording data string generated by the RLL (2, 10) encoding.
Are consecutive, or there is only one '0' between '1'. x [t] = 1, x [t-1] = 1, x [t-2] = 1, x [t-3] = 1 x [t] = 1, x [t-1] = 1, x [t-2] = 1, x [t-3] = 0 x [t] = 1, x [t-1] = 1, x [t-2] = 0, x [t-3] = 1 x [t] = 1, x [t-1] = 1, x [t-2] = 0, x [t-3] = 0 x [t] = 1, x [t-1] = 0, x [ t-2] = 1, x [t-3] = 1 x [t] = 1, x [t-1] = 0, x [t-2] = 1, x [t-3] = 0 x [ t] = 0, x [t-1] = 1, x [t-2] = 1, x [t-3] = 1 x [t] = 0, x [t-1] = 1, x [t -2] = 1, x [t-3] = 0 x [t] = 0, x [t-1] = 1, x [t-2] = 0, x [t-3] = 1 x [t ] = 0, x [t-1] = 0, x [t-2] = 1, x [t-3] = 1

Considering the conditions imposed on the precode output y [t] according to the equation (1) based on such conditions imposed on the recording data string, the state S001
0, S0100, S0101, S0110, S100
It can be seen that eight states of 1, S1010, S1011 and S1101 cannot occur. Therefore, eight states (= 2 ⁴ -8) can occur.

In order to obtain a state that can occur at the next time point s + 1 starting from the state at a certain time point s,
It is necessary to check separately when the value x [s + 1] of the recording data at the time point s + 1 is 1 and when it is 0.

Here, the state S0000 will be described as an example. According to the above equation (1), S0000, that is, h =
y [s] = 0, i = y [s−1] = 0, j = y [s−
2] = 0 and k = y [s−3] = 0, the following two data are conceivable.

X [s] = 0, x [s-1] = 0, x [s-2] = 0, x [s-3] = 1 x [s] = 0, x [s-1] = 0, x [s-2] = 0, x [s-3] = 0

[When x [s + 1] = 1] At this time, (1)
According to the equation, y [s + 1] is calculated as follows. y [s + 1] = mod2 {x [s + 1] + y [s]} = mod2 {1 + 0} = 1 Therefore, the value of the reproduced signal r [s + 1] is calculated as follows according to the equation (2).

R [s + 1] = a × y [s−2] + b × y [s−1] + b × y [s + 1] −a × y [s + 1] −ab = ｛a × 0 + b × 0 + b × 0 + a × 1｝ -ab = -b

The state Shijk at the next time point s + 1 is
h = y [s + 1], i = y [s], j = y [s−1],
k = y [s-2]. And y as described above
[S + 1] = 1, y [s] = 0, y “s−1” = 0, y
Since [s−2] = 0, the state at the next time point s + 1 is S1000. Therefore, when y [s + 1] = 1, it can be specified that a transition of S0000 → S1000 occurs.

[When x [s + 1] = 0] At this time, (1)
According to the equation, y [s + 1] is calculated as follows. y [s + 1] = mod2 {x [s + 1] + y [s]} = mod2 {0 + 0} = 0 Therefore, the value of the reproduced signal r [s + 1] is calculated as follows according to the equation (2).

R [s + 1] = a × y [s−2] + b × y [s−1] + b × y [s] + a × y [s + 1] −ab = ｛a × 0 + b × 0 + b × 0 + a × 0｝ −ab = −ab

The state Shijk at the next time point s + 1 is
Are h = y [s + 1], i = y [s], j = y [s−
1], k = y [s-2]. Then, as described above, y [s + 1] = 0, y [s] = 0, y “s−1” =
Since 0, y [s-2] = 0, the state at the next point in time is S0000. Therefore, when x [s + 1] = 0, it can be specified that a transition from S0000 to S0000 occurs.

Thus, the state S0 at the time point s is obtained.
Regarding each state other than 000, the state transition that can occur at the next time point s + 1 starting from them and the recording data value x [s + 1] and the reproduction signal value r [s + 1] when such a state transition occurs occur. A response can be requested.

As described above, for each state, the correspondence between the state transition that can occur starting from the state, the value of the recording data and the value of the reproduction signal at the time when each state transition occurs, is obtained, and the correspondence is shown in the form of FIG. FIG. 5 shows the result.

The above time points s and s + 1 are not special time points. Therefore, the correspondence between the possible state transitions obtained as described above and the accompanying values of the recording data and the values of the reproduction signal can be applied at any time. For this reason, in FIG. 5, the value of the recording data accompanying the state transition occurring at an arbitrary time point t is represented as x [t], and the value of the reproduced signal is represented as r [t].

In FIG. 5, the state transition is represented by an arrow. Further, the sign given to each arrow indicates (recorded data value x [t] / reproduced signal value r [t]). State S
While there are two types of state transition starting from 0000 and S1111, states S0011, S0111, S1
000, S1110, S0001, and S1100 can have only one transition.

Further, in FIG. 3, states S0000 and S0000
0001 is r for x [t + 1] = 0
The value [t + 1] =-ab is taken, and the state S0000
Has transitioned to On the other hand, for x [t + 1] = 1, a transition from the state S0001 cannot occur.

Similarly, states S1111 and S1110 take the same value of r [t] for the same x [t + 1] = 0, and transit to the same state. Therefore, state S
0000 and S0001 are collectively expressed as S0, and the state S
1111 and S1110 can be collectively expressed as S3. Further, the state S0011 is changed to S4, and the state S011 is set.
1 as S5, state S1000 as S1, state S1100
Is expressed as S2, and the state transition diagram arranged is shown in FIG.

FIG. 3 is a state transition diagram used in the five-valued six-state Viterbi decoding method. In FIG. 3, S0
6 to S5, and five values of -ab, -b, 0, b, and a + b as the value of the reproduction signal r [t + 1]. While there are two types of state transitions starting from the states S0 and S3, there is only one state transition starting from the states S1, S2, S4, and S5.

Here, in FIG. 3, the states S2 and S3 of the right half are shown.
And S5 are both positive values (either b or a + b)
, And the left half states S0, S1, and S4 all follow a path of a negative value (either -b or -ab). By referring to the value, it can be determined whether the route is a positive route or a negative route.

Moreover, if the interval from one zero cross point to the next zero cross point is known, that is, if the number of transitions from state S2 to state S4 or from state S4 to state S2 is known, the route is determined. And it becomes clear for each possible sample point.

In the above state transition diagram, the value of "0" is
As shown in FIG. 4, it indicates a zero cross point, and a value other than “0” indicates that it is not a zero cross point. Information indicating whether or not this is a zero cross point is referred to as zero point information Z. , Z = 1, it is assumed that the zero cross point is obtained. From the state transition diagram, two zero cross points (Z = 1) are not taken out consecutively, and RLL (2, 10)
In the case of, there are at least two Z = 0 between adjacent Z = 1 (that is, the state S1 → S2 → S3 → S5 → S
4 or the state S5 → S4 → S0 → S1 →
When transitioning to S2).

Paying attention to the above points, the equalization target value provisional judgment circuit 8 in FIGS. 1 and 2 compares the input waveform equalization data from the adaptive waveform equalizer 7 with a fixed threshold value, For every five sample points of the continuously input waveform equalization data, the values of those five sample points are all either Z = 0 or 5
Whether only the first of the five values is Z = 1, if only the last of the five consecutive values is Z = 1, the first and last of the five consecutive values are Z = 1, The remaining three values determine whether Z = 0.

In these patterns, when the value of the 0-point information of interest is Z = 0, the values Z of the 0-point information on both sides are both "0". Is attached to the positive side or the negative side, and when any of these patterns is satisfied, a large value P1 is calculated.

In addition, when neither of the above patterns is used, the value Z of the zero point information of five consecutive sample points is adjacent on both sides when the value Z of the zero point information of interest is “0”. One of the values Z of the 0 point information is "
It is determined whether it is 1 ", and when this condition is satisfied, a value P2 of an intermediate level between 0 and P1 is calculated.

When the value of the current time of the input waveform equalized data is 0 or more, the final provisional judgment value Q is set to the value of P1 or P2 at that time, and when the input waveform equalization data is negative, the final provisional judgment value Q is set to that value. Invert the value and polarity of P1 or P2. If none of the above, the final provisional judgment value Q is set to 0.
And This final provisional determination value Q is used as an equalization target value in FIG.
And output to the subtractor 9 in FIG. Note that the algorithm of the above-described provisional determination circuit 8 is described in Japanese Patent Application No.
No. 372392 (inventor: Junichiro Tonami).

The equalization target value provisional decision circuit 8 converts the provisionally decided state information (provisional decision information) into the nonlinear distortion corrector 12.
And outputs to the LPF 10 five waveform equalization sample data strings corresponding to the five equalization target values. If the waveform equalization sample data sequence is an ideal reproduction signal, the equalization target value and the waveform equalization sample data sequence will match, but if the waveform equalization sample data sequence has asymmetry, the equalization target The value and the waveform equalization sample data sequence have different values, and the amount and polarity of the nonlinear distortion can be obtained based on the difference between the waveform equalization sample data sequence and the equalization target value.

Here, the operation of the present invention for correcting nonlinear distortion will be described with reference to FIG. Equalization target value provisional judgment circuit 8
Are temporarily determined into five states (a + b, b, 0, -b, -ab), and the average value of the variance of the five waveform equalized sample data strings corresponding to each state is separately obtained by the LPF 10 ( The LPF 10 is composed of five LPFs for obtaining an average value of each waveform equalized sample data sequence), and the five equalization target values known by the nonlinear distortion detector 11 and the LPF
An error from the corresponding average value from F10 is detected as nonlinear distortion information, and the detected five nonlinear distortion information is supplied to the nonlinear distortion corrector 12.

The nonlinear distortion corrector 12 converts the waveform equalization data extracted from the adaptive waveform equalizer 7 from the equalization target value provisional decision circuit 8 out of the five pieces of nonlinear distortion information from the nonlinear distortion detector 11. One nonlinear distortion information corresponding to the state at the current time selected by the selector 15 based on the tentative decision information is corrected by adding it by the adder 16, and the corrected waveform equalized data is added to the Viterbi decoder 13. To generate decoded data by the Viterbi decoding method. Viterbi decoder 1
3 is the maximum likelihood decoded sequence for the recorded data recorded. Therefore, when there is no decoding error, the decoded data matches the recorded data.

As described above, according to the present embodiment, since Viterbi decoding is performed using data in which non-linear distortion due to asymmetry or the like has been corrected in advance, the identification point of Viterbi decoding includes the partial response characteristic and the signal amplitude level. Thus, the discrimination point determined by this can be used as it is, and more reliable decoding can be performed without deteriorating the Viterbi decoding performance.

It should be noted that the present invention is not limited to the above-described embodiment, and the reproduction of the RLL (1, X) digital signal is performed by the equalization target value calculating algorithm of the equalization target value provisional decision circuit 8. The present invention can be applied by making slight changes to.

[0068]

As described above, according to the present invention,
By performing correction to remove errors such as asymmetry from the digital signal reproduced from the recording medium in advance, when decoding by the Viterbi decoding method, the identification point of Viterbi decoding is kept as it is by the identification point determined by the partial response characteristics and the signal amplitude level. Since it can be used, more reliable decoding can be performed without deteriorating Viterbi decoding performance.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of the device of the present invention.

FIG. 2 is a block diagram of an embodiment of an adaptive waveform equalizer in FIG.

FIG. 3 is a state transition diagram of a five-valued six-state partial response.

FIG. 4 shows RLL (2, 10) and PR (1, 1, 1, 1)
FIG. 4 is a diagram showing an eye pattern of a signal at the time of FIG.

FIG. 5 is a schematic diagram for explaining a process of creating a state transition diagram of the 5-valued 6-state Viterbi decoding method.

FIG. 6 is a configuration diagram of a main part of FIG. 1;

[Explanation of symbols]

 Reference Signs List 1 optical disk 2 optical head 4 A / D converter 5 AGC circuit 6 digital PLL (DPLL) circuit 7 adaptive waveform equalizer 8 equalization target value provisional judgment circuit 9 subtractor 10 low-pass filter (LPF) 11 nonlinear distortion detector 12 Nonlinear distortion corrector 13 Viterbi decoder 71 Transversal filter 72 Multiplier / LPF

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G11B 20/18 570 G11B 20/18 570F 572 572C 572F H03M 13/41 H03M 13/41

Claims (3)

[Claims] 1. A reading means for reading a digital signal from a recording medium on which a digital signal is pre-coded and recorded; a bit clock generated from the digital signal read by the reading means; Sampling interpolation means for sampling and interpolating a signal; and performing waveform equalization of an adaptive partial response characteristic on the digital signal extracted from the sampling interpolation means based on an equalization error input from the outside to form waveform equalization data. Adaptive waveform equalizing means for outputting a predetermined state transition of the pre-coded recording digital signal and a plurality of 0-point information with respect to the waveform equalized data output from the adaptive waveform equalizing means. An equalization target value is calculated, and a difference between the equalization target value and the waveform equalization data is calculated. And calculates the equalization error and supplies the same as the equalization error to the adaptive waveform equalization means, and indicates a current transition state and a plurality of waveform equalization sample data strings corresponding to a plurality of equalization target values based on the state transition. Information and an averaging means for averaging each of the plurality of waveform equalization sample data strings output from the equalization target value tentative determination means; and A nonlinear distortion detecting unit that detects an error between an output signal and the plurality of equalization target values as nonlinear distortion information; andthe adaptive waveform based on the information indicating the transition state and the nonlinear distortion information from the nonlinear distortion detecting unit. An information signal reproducing device, comprising: a nonlinear distortion correction unit that corrects and outputs the waveform equalization data from the equalization unit. 2. The information signal reproducing apparatus according to claim 1, further comprising a Viterbi decoder that Viterbi-decodes the waveform equalized data corrected by the non-linear distortion correction unit. 3. A digital signal reproducing method for reading a signal from a recording medium in which a digital signal is precoded and recorded and reproducing a digital signal from the read signal, wherein a bit clock is generated from the read signal. A first step of sampling and interpolating the read signal with the bit clock; and a waveform of an adaptive partial response characteristic with respect to the sampled and interpolated signal based on an equalization error input from the outside. Second to perform waveform equalization and output waveform equalized data
Calculating an equalization target value based on a predetermined state transition of the pre-coded recording digital signal and a plurality of zero-point information with respect to the waveform equalization data, and calculating the equalization target value and the waveform. A third step of calculating a difference with the equalized data and outputting the same as the equalization error; and a plurality of waveforms which receive the waveform equalized data as input and correspond to a plurality of equalized target values based on the state transition. A fourth step of outputting the equalized sample data sequence and information indicating the current transition state; a fifth step of averaging the plurality of waveform equalized sample data sequences, respectively; and an averaging by the fifth step A sixth step of detecting an error between the converted signal and the plurality of equalization target values as non-linear distortion information, and based on the information indicating the transition state and the non-linear distortion information, Information signal reproducing method characterized by comprising a seventh step of outputting the corrected data.