JP5623097B2 - Information reproducing method and apparatus, and optical disc apparatus - Google Patents

Description

  The present invention relates to an adaptive filter, an information reproduction method and apparatus for reproduction using a PRML (Partial Response Maximum Likelihood) method, and an optical disc apparatus including the information reproduction apparatus.

  Increasing the capacity of various optical discs reduces the size of recording marks for information written on the tracks of the disc, reduces the wavelength of laser light used for recording and reproduction, and adopts an objective lens with a large numerical aperture. This has been achieved by reducing the size of the focused spot on the surface.

  For example, in a CD (compact disc), the thickness of a disc substrate that is a light transmission layer (a transparent protective layer and a space layer provided on the information recording layer; also referred to as a transparent substrate) is about 1.2 mm, and the laser beam wavelength is About 780 nm, the numerical aperture of the objective lens was 0.45, and the recording capacity was 650 MB. At this time, the resolution of the data (pits) is limited by the diffraction limit. This diffraction limit is given by λ / (4 × NA) using the laser light wavelength λ and the numerical aperture NA. The limit can be calculated to be about 430 nm. On the other hand, the shortest data length (shortest pit length) is about 830 nm, and the size of the focused spot is about 1.93 times.

  In a DVD (Digital Versatile Disc), the thickness of the light transmission layer is about 0.6 mm, the laser light wavelength is about 650 nm, and the NA is 0.6, which is a recording capacity of 4.7 GB. The diffraction limit is calculated as in the case of CD, and is about 270 nm. On the other hand, the shortest data length is about 400 nm, and the size of the focused spot is about 1.48 times.

  For a higher density BD (Blu-ray Disc), an optical disc with a light transmission layer thickness of 0.1 mm is used, the laser light wavelength is about 405 nm, and the NA is 0.85. Capacitance is realized. The diffraction limit can be calculated to be about 120 nm. On the other hand, the shortest data length is about 150 nm, and the size of the focused spot is about 1.25 times.

  From the above, apart from the increase in capacity by reducing the focused spot, the shortest data length for the focused spot has also been reduced from about 1.93 to about 1.25 times, which is also high capacity. Has led to This is due to the evolution of signal processing technology, and it is possible to reduce the SNR (Signal to Noise Ratio) required for the read reproduction signal.

  As described above, the increase in the capacity of the optical disc is accompanied by an improvement in the signal processing technique. Among them, the reproduction waveform of the optical disc has a known partial response characteristic, and this is the maximum likelihood by the Viterbi decoding method. A PRML method combining an estimation method has been developed, and this technique contributes to a large error rate improvement (for example, Patent Document 1).

  Specifically, for example, in the case of a BD, PRML with a partial response class of (1, 2, 2, 1) is generally used. (1, 2, 2, 1) expresses the optical response (intersymbol interference) to recorded binary data with seven gradations (amplitude values), and can be expressed close to the actual reproduced waveform. is there. By using a maximum likelihood estimation method (Viterbi decoding method) for this, an ideal optical response close to the reproduced waveform is derived, and the recorded binary data is estimated.

  Further, as an example in which the contribution of signal processing is large, there is an HD DVD that realizes a large capacity by PRML with the same optical system as a DVD. This HD DVD has a shortest data length of about 200 nm and is smaller than the diffraction limit of about 270 nm, so the shortest data cannot be read. Therefore, the problem is solved by setting the partial response class to (1, 2, 2, 2, 1). This class represents an optical response in which the shortest data does not appear in the reproduced waveform with nine gradations (amplitude values).

  As described above, as the capacity of optical disks continues to increase, physical improvements in resolution that are limited by the diffraction limit are becoming increasingly difficult, and the role of signal processing is increasing. In particular, shortening the wavelength of the laser beam to be used to be shorter than 405 nm of BD is not considered practical because it causes deterioration of the optical element and concerns about adverse effects on the body. Therefore, in the future, it will change from the flow up to the BD, and it is moving in the direction using near-field light, multilayering, and holography. At the same time, asymmetry of the reproduction waveform deteriorates or the signal intensity near the shortest data length decreases, the quality of the reproduction signal further deteriorates, and further signal processing technology needs to be improved.

  One example is adaptive PRML (see, for example, Patent Document 2). This is a method of adapting each amplitude value of the partial response to the reproduced waveform, and is particularly effective for a reproduced waveform having a large asymmetry. Specifically, for each amplitude value of the partial response, the center value of the distribution of points corresponding to each amplitude value in the reproduced waveform is derived, and the center value is newly set as the value of each amplitude value of the partial response. is there.

  As another method different from the above, there is a method in which the tap coefficient update step of the adaptive filter is made adaptive, and the mark and space are separated and learned (see, for example, Patent Document 3). This has the effect of improving the convergence characteristics of the adaptive filter and the stability of the convergence state for a reproduced waveform having a large asymmetry.

JP-A-2005-93033 JP-A-10-261273 JP 2007-73099 A

  However, the above solution has a problem that it cannot cope with a decrease in signal intensity near the shortest data length. Specifically, when the method of Patent Document 2 is used, in the Viterbi decoding method, a path metric (sum of Euclidean distances) of possible paths is compared to estimate a path with the maximum likelihood. It only reduces the metric as a whole, and when considering the difference between the path metrics, it does not increase or decrease, and does not affect the determination of which path is most likely. On the other hand, when the method of Patent Document 3 is used, the maximum likelihood estimation determination is stabilized because the convergence characteristics of the adaptive filter and the stability of the convergence state can be improved, but the signal strength decreases near the shortest data length. There is no effect of reducing errors due to.

An information reproduction method according to one aspect of the present invention includes:
In an information reproduction method for generating binarized data by a PRML method with respect to a reproduction signal from a recording medium,
Outputting a partial response waveform from the binarized data generated by the PRML method;
Adaptively filtering the reproduced signal with the partial response waveform as a target waveform;
When the difference between the target waveform and the waveform filtered by the adaptive filter process is fed back to the adaptive filter process as an error signal, the error signal having the shortest data length is reduced and the second adjacent to the shortest data length is reduced. An adjustment step for adjusting the partial response waveform so as to increase an error signal having a short data length .
In the adjustment step, a value obtained by subtracting a first fixed value from an absolute value of an amplitude value appearing in the partial response waveform when the data length is the shortest data length is an absolute value, which is the same as the amplitude value before adjustment. By making the amplitude value having a sign an amplitude value after adjustment, the error signal is made smaller,
In the adjustment step, an absolute value is obtained by adding a second fixed value to the absolute value of the amplitude value appearing in the partial response waveform when the data length is the second shortest data length. with amplitude value of the adjusted amplitude value having the same sign as value, characterized in that you larger the error signal.
An information reproduction method according to another aspect of the present invention includes:
In an information reproduction method for generating binarized data by a PRML method with respect to a reproduction signal from a recording medium,
Outputting a partial response waveform from the binarized data generated by the PRML method;
Adaptively filtering the reproduced signal with the partial response waveform as a target waveform;
When the difference between the target waveform and the waveform filtered by the adaptive filter process is fed back to the adaptive filter process as an error signal, the error signal having the shortest data length is reduced and the second adjacent to the shortest data length is reduced. An adjustment step for adjusting the partial response waveform so as to increase an error signal having a short data length.
The adjustment step makes the error signal smaller by multiplying the error signal of the shortest data length by a first fixed coefficient smaller than 1.
The adjustment step makes the error signal larger by multiplying the error signal having the second shortest data length by a second fixed coefficient greater than 1.
It is characterized by that.

  According to the present invention, a local error near the shortest data length can be reduced with a small amount of computation (circuit scale).

1 is a diagram illustrating an overall configuration of an optical disc device according to a first embodiment of the present invention. It is the schematic which shows the structure of an information reproduction apparatus. It is explanatory drawing which shows the misjudgment of the reproduction signal in a PRML circuit. It is explanatory drawing which shows the misjudgment in a PRML circuit. 1 is a schematic diagram illustrating a configuration example of an information reproducing device according to Embodiment 1. FIG. FIG. 3 is an explanatory diagram illustrating a target waveform in the first embodiment. FIG. 6 is a schematic diagram showing an effect of the information reproducing apparatus in the first embodiment. FIG. 10 is a schematic diagram illustrating another configuration example of the information reproducing apparatus in the first embodiment. It is the schematic which shows the structural example of the information reproduction apparatus in Embodiment 2 of this invention. It is the schematic which shows the effect of the information reproduction apparatus in Embodiment 2 of this invention. FIG. 10 is a schematic diagram illustrating another configuration example of the information reproducing apparatus in the second embodiment.

Embodiment 1 FIG.
FIG. 1 is a diagram showing an overall configuration of the optical disc apparatus according to the first embodiment. Note that the arrows in FIG. 1 indicate the flow of typical signals and information, and do not represent all the connection relationships of the blocks constituting the optical disc apparatus 50.

  In FIG. 1, an optical disk device 50 irradiates a laser beam to a spindle motor 51 for rotating the optical disk 40 and the optical disk 40, receives the return light beam reflected by the information recording layer of the optical disk 40, and outputs a signal. Optical head device 52, a sled motor 53 for driving the optical head device 52 in the radial direction of the optical disc 40, a laser control circuit 54, a servo control circuit 55, a reproduction signal processing circuit 56, a modulation circuit 64, a RAM (Random Access Memory). ) 80, and MPU (Micro Processing Unit) 81.

  The servo control circuit 55 includes a spindle motor control circuit 63 that controls the spindle motor 51, a thread motor control circuit 62 that controls the thread motor 53, and an optical head control circuit 61 that controls the optical head device 52. The operation is performed according to the instruction.

  The reproduction signal processing circuit 56 detects a reproduction signal based on the signal detected by the optical head device 52 and sent via the transmission line L3, and detects the reproduction signal to the transmission line L1. There is a reproduction signal detection circuit 58 that outputs as an output signal, and a wobble signal detection circuit 57 that detects a wobble signal obtained from reflected light from the meandering guide track groove of the optical disk 40.

  The MPU 81 determines the operation of the entire optical disc apparatus based on the output signal of the transmission line L1 such as the signal amplitude value data and the status signal detected by the reproduction signal detection circuit 58, or the output signal from each other part, and to each part. Control data (for example, a signal on the transmission line L2 from the MPU 81 to the reproduction signal detection circuit 58) is sent to control them.

  Note that a part of the components of the reproduction signal processing circuit 56 may be processed by software inside the MPU 81.

  The RAM 80 includes a program area and a data area. The MPU 81 controls the operation of each part in accordance with the program recorded in the RAM 80, and makes a control judgment from signals sent from each part.

  The optical head control circuit 61 outputs a control signal to the optical head device 52 via the transmission line L4 based on the servo error signal sent from the servo signal detection circuit 59 and the operation command from the MPU 81, and the optical head device 52 receives the optical disk from the optical head device 52. The light irradiated onto 40 is controlled.

  The thread motor control circuit 62 and the spindle motor control circuit 63 control the thread motor 51 and the spindle motor 53 based on the servo error signal and the operation command from the MPU 81.

  The output signal (reproduction signal) of the reproduction signal detection circuit 58 is demodulated into binary data by the information reproduction device 60 through the transmission line L1.

  Part of the data output from the MPU 81 is converted into a recording signal suitable for recording on the optical disk 40 by the modulation circuit 64 and sent to the laser control circuit 54. Based on this recording signal, a control signal is sent from the laser control circuit 54 to the optical head device 52 via the transmission line L5, and the light emission power of the semiconductor laser mounted on the optical head device 52 is controlled.

  The optical head device 52 collects a light beam from a semiconductor laser on the optical disc 40 and receives a return light beam reflected by the information recording layer of the optical disc 40 to detect a signal for generating a reproduction signal or a servo signal. .

  FIG. 2 is a diagram illustrating a general configuration example of the information reproducing device 60. As described above, this is a block that binarizes the reproduction signal input from the reproduction signal detection circuit 58 through the transmission line L1.

  The input reproduction signal is converted into a digital value by the A / D converter 1. At this time, a PLL (Phase-Lock Loop) circuit 8 generates a clock synchronized with the reproduction signal, and uses this as a clock for the entire information reproduction apparatus 60 including the A / D converter 1. For example, when the signal modulation method is 1-7 RLL (Run Length Limit) modulation, the length of one cycle of the clock is T, and the reproduced signal includes a length of 2T (shortest data length) to 8T, and is reproduced. The PLL circuit 8 works so that the point at which the signal crosses zero and the rising or falling point of the clock coincide. Below, the case where this 1-7 RLL modulation is used is demonstrated.

  The digitized reproduction signal is adjusted to a desired amplitude level by the digital amplifier 2. At this time, if the amplifier 2 is an AGC (Auto Gain Control) circuit and is configured to automatically adjust so that the amplitude level becomes a constant value, the fluctuation of the amplitude level due to the reflected light fluctuation of the optical disk, etc. This circuit can be ignored.

  The reproduced waveform adjusted to a desired amplitude level is mainly amplified by the pre-equalizer circuit 3 for high frequency components around the shortest data length. Due to the MTF (Modulation Transfer Function) characteristics of the light, the high frequency component around the shortest data length tends to become extremely small. Therefore, if the pre-equalizer circuit 3 amplifies to some extent, the processing in the subsequent filter becomes easy.

The adaptive filter circuit 4 is used in combination with the PRML circuit 5 at the subsequent stage, and performs a filter process using the ideal reproduction waveform estimated by the PRML circuit 5 as a target waveform. For example, a typical adaptive algorithm is an LMS (Least Mean Square) algorithm. This is because the input signal data at time n is x (n), the coefficient at the filter tap k is w _k (n), the filter coefficient update step is μ, and the coefficient update is
w _k (n + 1) = w _k (n) + 2 · μ · e (n) · x (n) (1)
It is represented by At this time, e (n) is an error signal, and the data string of the reproduced waveform after filtering is y (n) and the data string of the target waveform is d (n).
e (n) = d (n) -y (n) (2)
It is represented by The error signal e (n) is a subtracter 9 that receives the target waveform data string d (n) output from the PRML circuit 5 and the reproduced waveform data string y (n) output from the adaptive filter 4 as inputs. Is required. Equations (1) and (2) are derived from an algorithm that minimizes the instantaneous square error e ² (n) at time n. Since this instantaneous square error is averaged over time, LMS ( Called the least mean square).

  In addition to this LMS algorithm, a normalized LMS algorithm, an RMS (Recursive Last Square) algorithm, a projection algorithm, or the like may be used. In the following description, an example using a simple and low-computation LMS algorithm will be described.

  As described above, the adaptive filter circuit 4 only functions to mathematically minimize the error signal. On the other hand, the ease of decoding by the subsequent PRML circuit 5 exists separately, and when the reproduced signal waveform deteriorates, there arises a problem that errors in the vicinity of a short T occur frequently. As an example, the case where the partial response class is (1, 2, 2, 1) will be described below.

  The PRML circuit 5 includes a Viterbi decoding circuit 6 and a PR decoder 7. In the Viterbi decoding circuit 6, maximum likelihood estimation and binarization processing are performed. Further, the PR decoder 7 performs a process of decoding the binarized data as a target waveform of the adaptive filter circuit 4. FIG. 3 shows an example of erroneous determination in the Viterbi decoding circuit 6. At this time, the reproduction waveform RS in the figure is a waveform after filtering. The estimated waveform ES in the figure corresponds to the waveform that has undergone the estimation process in the PRML circuit 5. That is, the waveform estimated by the Viterbi decoding circuit 6 is the most correct, and its binary data string is (1, 1, 0, 0, 1, 1, 1). On the other hand, what is decoded from binary data written on the optical disk is an ideal waveform TS, which corresponds to an ideal waveform that the Viterbi decoding circuit 6 can process most accurately and without error, and its binary data string is ( 1, 1, 0, 0, 0, 1, 1). That is, the portion that is 3T is erroneously determined as 2T.

  The ideal waveform TS is equal to the waveform estimated by the Viterbi decoding circuit 6 to be second correct. In the figure, when attention is paid to three points A, B, and C which are erroneous 3T portions, the ideal waveform TS is close to the reproduced waveform RS at the point A, and the estimated waveform ES and the ideal waveform TS are reproduced waveforms at the points B and C. The difference from RS is comparable, and the total error appears closer to the ideal waveform TS than to the estimated waveform ES. However, in Viterbi decoding, since the maximum likelihood path is found in consideration of a certain length, the error before and after is also considered. That is, 3T (left side in the figure) and 4T (right side in the figure) are adjacent to 3T, but these errors are also integrated, and the estimated waveform ES is finally selected as the maximum likelihood.

  The reason why errors occur frequently with a short T as described above is that 1T cannot exist due to the limitation of 1-7 RLL modulation, and 2T is often erroneously determined to be 3T. An example is shown in FIG. As a case where 3T is erroneously determined as 2T or 4T with respect to the reference waveform Sc, four waveforms of Sa, Sb, Sd, and Se are illustrated. Thus, it can be seen that adjacent long T edges are shifted in the waveforms other than the waveform Sb. This edge shift is so-called jitter itself. In general, a longer T has a smaller jitter value, that is, a longer T has a smaller edge shift amount. Conversely, consider a waveform in which the edge shift increases as T decreases. From this, it is considered that erroneous determination is rarely performed on the waveforms Sa, Sd, and Se in which the long T edge is shifted, and erroneous determination is often performed on the waveform Sb confusing 2T and 3T.

  From the above, it is considered that avoiding erroneous determination of 2T and 3T facilitates decoding in a system using 1-7 RLL modulation and PRML with a partial response class of (1, 2, 2, 1). .

  Here, 1-7 RLL modulation is taken as an example, but the same idea applies to EFM (Eight-to-Fourteen Modulation) modulation used in DVD. In this case, since there is no 2T or less, the erroneous determination of 3T and 4T is dominant, and the erroneous determination of 3T and 4T may be avoided.

  Further, the partial response class is not limited to (1, 2, 2, 1), and for example, (1, 2, 2, 2, 1) has the same concept.

  Next, means for avoiding erroneous determination of 2T and 3T will be described. Simply, an equalizer function that suppresses the 2T component and emphasizes the 3T component may be provided. For example, an adjuster (equalizer) may be inserted immediately after the pre-equalizer 3. However, the adaptive filter 4 at the subsequent stage works in a direction to cancel the equalization effect by the inserted adjuster (equalizer), so that the equalization effect is reduced.

  Therefore, in the present embodiment, as shown in FIG. 5, an adjuster 12 composed of an arithmetic unit for adjusting the output of the PR decoder 7 is provided. This is to equalize the target signal output from the PR decoder 7 to change the filter coefficient of the adaptive filter 4 and indirectly give an equalizing effect to the reproduced waveform.

  Specifically, for example, when the partial response class is (1, 2, 2, 1), the amplitude values that 2T can take are three points of 0, -1, and +1 as shown in FIG. On the other hand, the amplitude values that 3T can take are three points of 0, -2, and +2. Therefore, by using this difference, the adjuster 12 sets the point of the target signal corresponding to −1, +1 to a smaller value and sets the point of the target signal corresponding to −2, +2 to a larger value. An equalizing function that emphasizes 3T while suppressing 2T can be expected.

  For example, as shown in FIG. 6, the target signal points corresponding to −1 and +1 are +0.5 and −0.5, respectively, and the target signal points corresponding to −2 and +2 are +0.5 and −2, respectively. If 0.5, 2T is suppressed and the target signal is enhanced with 3T. The effect at this time is shown in FIG. The figure shows the variance in each amplitude value of the reproduced waveform after the adaptive filter, and the amplitude difference between the two points of −1 and +1 and the two points of −2 and +2 is widened by the equalizing process by the adjuster 12. I understand that. It can also be seen that, in each amplitude value, the dispersion peak is higher than before the equalization processing by the adjuster 12, and the dispersion is small.

  In the above example, the values of a total of four points of -1, +1, -2, +2 are changed, but only two points of -1, +1 may be changed (smaller values), -2, +2 It is possible to change only two points (set to a larger value), or to change six points -1, +1, -2, +2, -3, and +3. As a result, 2T and 3T are separated. In other words, an arithmetic process that increases the difference in the error signal value due to the difference in data length may be added.

  Moreover, although the case where the class of the partial response is (1, 2, 2, 1) is shown as an example, the present invention is not limited to this. For example, the same effect can be obtained when (1, 2, 2, 2, 1), etc. It is done.

  In the example shown in FIG. 5, the amplitude value is simply changed. However, as shown in FIG. 4, the two points −2 and +2 are also amplitude value points that can be taken by 4T or more. Therefore, when the values of −2 and +2 are changed, 4T or more is affected. On the other hand, in order to emphasize 3T more purely, it is preferable to change the values of −2 and +2 only when it is determined to be 3T. Specifically, a data length determination circuit 13 for determining the data length using the binarized data sequence estimated by the Viterbi decoding circuit 6 is provided as shown in FIG. Only when the determination is made, the regulator 12 may perform the same arithmetic processing as described above on the target signal.

  As shown in FIGS. 5 and 8, instead of performing the above addition / subtraction process on the target signal output from the PRML circuit 5, the above addition / subtraction is performed when the target signal is generated in the PRML circuit 5. It is also possible to perform processing corresponding to. In the latter case, for example, it is conceivable to use a conversion table in which the output of the PR decoder is input in the PRML circuit 5.

  According to the information reproducing apparatus and the information reproducing method implemented by the apparatus, there is a local error reducing effect near the shortest data length.

  In order to realize the above equalization function, it is only necessary to change the value output from the PR decoder 7 as a load. Therefore, the amount of calculation increase or circuit scale increase due to the addition of the adjuster 12 is almost not. The effect of the information reproducing method described above can be obtained.

  Also in the optical disc apparatus provided with this information reproducing apparatus, the effect of the information reproducing method described above can be obtained.

Embodiment 2. FIG.
Next, an information reproduction apparatus according to Embodiment 2 and an information reproduction method implemented using the apparatus will be described. FIG. 9 shows an information reproducing apparatus according to the second embodiment of the present invention. The illustrated information reproducing apparatus is generally the same as the information reproducing apparatus described with reference to FIG. 5 with respect to the first embodiment, but differs in the following points. The information reproducing apparatus according to the second embodiment is used as a part of the optical disk apparatus as in the first embodiment.

  The information reproducing apparatus according to Embodiment 2 changes the filter coefficient of the adaptive filter 4 by equalizing the error signal fed back to the adaptive filter 4 using the adjuster 14, and indirectly with respect to the reproduced waveform. Gives equalization effect.

  As described in the first embodiment, as a cause of frequent occurrence of errors around short T, a short T jitter value is generally worse than a long T jitter value. On the contrary, here, consider a waveform in which the jitter value deteriorates as T becomes shorter. Further, as shown in the first embodiment, in Viterbi decoding, it is easy to determine a long T, but it is relatively difficult to determine a short T.

  Therefore, the adaptive filter 4 performs the adaptation process so that the error signal is stabilized at the convergence value without depending on the length of T. On the other hand, when the data length is 3T, the error signal is increased and the data length is increased. The error signal in the 2T case is made smaller and given to the adaptive filter.

  Specifically, for example, when the class of the partial response is (1, 2, 2, 1), as shown in FIG. 4, when the data length is relatively short (in the case of 2T), it appears in the partial response waveform. Amplitude values are three points of 0, -1, and +1, whereas when the data length is comparatively long (in the case of 3T), amplitude values that appear in the partial response waveform are three points of 0, -2, and +2. is there. Therefore, by using this difference, the regulator 14 suppresses the error signal at two points −1 and +1, and emphasizes the error signal at the two points −2 and +2, thereby suppressing 2T and reducing 3T. The emphasis equalization function can be expected.

As a method for suppressing the error signal (making the error signal smaller), it is conceivable to multiply by a coefficient smaller than 1. As a method for enhancing the error signal (making the error signal larger), a coefficient larger than 1 is multiplied. It is possible.
For example, FIG. 10 shows the effect of doubling the error signal at points −1 and +1 and doubling the error signal at points −2 and +2. The figure shows the dispersion at each amplitude value of the reproduced waveform after the adaptive filter, and it can be seen that the dispersion at −2 and +2 particularly related to 3T is reduced by the equalization processing.

  In the above example, values corresponding to a total of four points of -1, +1, -2, +2 were changed, but only two points of -1, +1, or -1, +1, -2, +2,- The value corresponding to 6 points of 3, +3 may be changed, and as a result, an arithmetic process that separates 2T and 3T may be added.

  Moreover, although the case where the class of the partial response is (1, 2, 2, 1) is shown as an example, the present invention is not limited to this. For example, the same effect can be obtained when (1, 2, 2, 2, 1), etc. It is done.

  Further, here, the error signals at the four points of -1, +1, -2, +2 are simply changed, but as shown in FIG. 4, the two points of -2, +2 have amplitudes that can be taken by 4T or more. It is also a point of value. Therefore, when the error signal at -2 and +2 is changed, 4T or more is affected. On the other hand, in order to emphasize 3T more purely, it is preferable to change the error signal at −2 and +2 only when it is determined to be 3T. Specifically, as shown in FIG. 11, the Viterbi decoding circuit 6 uses the binarized data string estimated by the Viterbi decoding circuit 6 to determine the data length as shown in FIG. The determination circuit 13 may be provided, and the error signal may be doubled and output by the adjuster 14 only when 3T is estimated by the data length determination circuit 13.

  According to the information reproducing apparatus and the information reproducing method implemented by the apparatus, there is a local error reducing effect near the shortest data length.

  Further, in order to realize the above equalization function, as a load, it is only necessary to multiply the error signal output from the subtracter 9 by a fixed value. There is almost no increase in computation or circuit scale due to the addition of 14, and the effect of the information reproducing method described above can be obtained.

  In the first and second embodiments, it has been described that the adjustment is performed so that the difference between the error signal values becomes larger when the data length is 2T and when the data length is 3T, but when the shortest data length is 3T. May be adjusted so that the difference in error signal value due to the difference between 3T and 4T becomes larger. Generally speaking, if the data length is adjusted near the shortest data length (that is, within the range including the shortest data length and data length close thereto), the difference in the error signal due to the difference in the data length is adjusted. good.

  In addition, when there is a defect in the optical disk, the reproduction waveform is greatly disturbed locally, and the error signal is also locally disturbed, thereby causing a problem of destabilizing the adaptive filter. By using the equalization process described above, this problem can be dealt with. Even when the data length is near the longest data length, the error signal is made larger, and the destabilization suppressing effect of the adaptive filter can be obtained.

  Overall, the feature of the present invention is that the magnitude of the error signal is adjusted in accordance with the data length.

  According to the first and second embodiments of the present invention described above, there is an effect that a local error near the shortest data length can be reduced without significantly increasing a calculation load or a circuit scale. is there.

  4 adaptive filter, 5 PRML circuit, 6 Viterbi decoding circuit, 7 PR decoder circuit, 9 subtractor, 12 adjuster, 13 data length judgment circuit, 14 adjuster, 40 optical disc, 50 optical disc device, 60 information reproducing device.

Claims (5)

In an information reproduction method for generating binarized data by a PRML method with respect to a reproduction signal from a recording medium,
Outputting a partial response waveform from the binarized data generated by the PRML method;
Adaptively filtering the reproduced signal with the partial response waveform as a target waveform;
When the difference between the target waveform and the waveform filtered by the adaptive filter process is fed back to the adaptive filter process as an error signal, the error signal having the shortest data length is reduced and the second adjacent to the shortest data length is reduced. An adjustment step for adjusting the partial response waveform so as to increase an error signal having a short data length.
In the adjustment step, a value obtained by subtracting a first fixed value from an absolute value of an amplitude value appearing in the partial response waveform when the data length is the shortest data length is an absolute value, which is the same as the amplitude value before adjustment. By making the amplitude value having a sign an amplitude value after adjustment, the error signal is made smaller,
In the adjustment step, an absolute value is obtained by adding a second fixed value to the absolute value of the amplitude value appearing in the partial response waveform when the data length is the second shortest data length. with amplitude value of the adjusted amplitude value having the same sign as value, information reproducing method characterized in that a larger the error signal. In an information reproduction method for generating binarized data by a PRML method with respect to a reproduction signal from a recording medium,
Outputting a partial response waveform from the binarized data generated by the PRML method;
Adaptively filtering the reproduced signal with the partial response waveform as a target waveform;
When the difference between the target waveform and the waveform filtered by the adaptive filter process is fed back to the adaptive filter process as an error signal, the error signal having the shortest data length is reduced and the second adjacent to the shortest data length is reduced. An adjustment step for adjusting the partial response waveform so as to increase an error signal having a short data length.
The adjustment step makes the error signal smaller by multiplying the error signal of the shortest data length by a first fixed coefficient smaller than 1.
The adjusting step, the the second shortest data length error signal is multiplied by a second fixed coefficient greater than 1, information reproducing method characterized in that a larger the error signal. In an information reproducing apparatus for generating binarized data by a PRML method with respect to a reproduction signal from a recording medium,
A partial response decoder that outputs a partial response waveform from the binarized data generated by the PRML method;
An adaptive filter that adaptively filters the reproduced signal with the partial response waveform as a target waveform;
When the difference between the target waveform and the waveform filtered by the adaptive filter is fed back to the adaptive filter as an error signal, the data length is calculated using the shortest data length and the second shortest data length adjacent to the shortest data length. Adjusting means for adjusting so as to increase the difference of the error signal according to the difference of
The adjustment means sets the absolute value to a value obtained by subtracting the first fixed value from the absolute value of the amplitude value appearing in the partial response waveform when the data length is the shortest data length, and is the same as the amplitude value before adjustment By making the amplitude value having a sign an amplitude value after adjustment, the error signal is made smaller,
The adjusting means uses an absolute value obtained by adding a second fixed value to an absolute value of an amplitude value appearing in the partial response waveform when the data length is the second shortest data length, and an amplitude before adjustment. with amplitude value of the adjusted amplitude value having the same sign as value, information reproducing apparatus characterized in that a larger the error signal. In an information reproducing apparatus for generating binarized data by a PRML method with respect to a reproduction signal from a recording medium,
A partial response decoder that outputs a partial response waveform from the binarized data generated by the PRML method;
An adaptive filter that adaptively filters the reproduced signal with the partial response waveform as a target waveform;
When the difference between the target waveform and the waveform filtered by the adaptive filter is fed back to the adaptive filter as an error signal, the data length is calculated using the shortest data length and the second shortest data length adjacent to the shortest data length. Adjusting means for adjusting so as to increase the difference of the error signal according to the difference of
The adjustment unit multiplies the error signal having the shortest data length by a first fixed coefficient smaller than 1, thereby reducing the error signal.
It said adjusting means is a short data length error signal to the second, by multiplying the second fixed coefficient greater than 1, information reproducing apparatus you characterized in that a larger the error signal. An optical disc apparatus comprising the information reproducing apparatus according to claim 3 .

Images