JP2005259317A - Method and device for reproducing multivalue information, multivalue information waveform equalizing device, and medium and device for recording multivalue information - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To tune frequency characteristics of a waveform equalizing during reproducing by efficiently and properly reproducing a reproducing signal having nonlinear characteristics of an intercode interference by a short pattern. <P>SOLUTION: A synchronous detecting circuit 21 detects a synchronous signal from a reproducing signal to generate a clock mark window signal, a clock detecting circuit 22 generates a clock signal from the reproducing signal on the basis of the clock mark window signal, and an ADC 23 digitizes and outputs the reproducing signal on the basis of the timing of the clock signal. A waveform equalizing circuit 24 corrects a high band signal attenuated by an intercode interference for a digitized data string output from the ADC23. A most ideal multivalue level is determined by a multivalue determining circuit 25, and multivalue information based on the result of the determination is output. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a multi-value information reproducing method, a multi-value information recording medium, a multi-value information waveform equalizer, a multi-value information reproducing device, and a multi-value information recording device.

When the waveform equalization circuit shapes the playback signal from the recording medium (media), by controlling the tap equalization coefficient according to the error from the preset reference determination level, and correcting the frequency characteristics There is a reproduction method (see, for example, Patent Document 1) in which characteristics are appropriately corrected based on a reproduction signal instead of a fixed waveform equalization characteristic.
In addition, as a method for correcting the characteristics of the waveform equalization circuit in multi-value information recording, a reproduction method using a pattern in which recording marks corresponding to multi-value information are sandwiched between zeros or a continuous pattern of the same recording marks (for example, patents) Reference 2).
Japanese Patent Laid-Open No. 3-178038 JP 2002-319138 A

An optical disc as a recording medium is required to be compatible with optical discs and apparatuses having different manufacturers and specifications. The problem here is that the characteristics of the reproduced signal differ depending on the difference in the shape of the mark and the size of the light spot.
The variation in the mark shape is caused by the difference in the recording method such as the laser emission waveform at the time of recording and the composition of the optical disk recording film. It differs between read-only (ROM) optical discs and recordable optical discs.

Further, the difference in the size of the light spot occurs due to laser characteristics, variation, and optical design, and also changes depending on the temperature at the time of recording.
Even in the case of recording binary information on an optical disk such as a CD or DVD, it is one of the problems to ensure compatibility by changing the characteristics of the reproduction signal.
The general idea is to ensure the quality of the optical disc by prescribing the lower limit of the signal quality in the standard playback system specifications, and to deal with device-specific variations within the margin. .
In this concept, it can be said that optical disks such as CDs and DVDs have excellent compatibility when considering the number of device manufacturers, the number of optical disk manufacturers, and the number of recording speeds selected.

If high recording density is pursued even in binary information recording, the limit of optical resolution is approached, so that adjacent marks also enter the light spot at the same time, and inter-code interference occurs in which adjacent mark components leak into the reproduction signal.
That is, the signal level of the mark to be reproduced changes depending on the size of the adjacent mark. The situation of the intersymbol interference varies depending on the mark shape and the size of the light spot.
However, intersymbol interference does not occur for marks and spaces having a size that sufficiently satisfies the optical resolution.
Conversely, even if the optical resolution limit is exceeded, the reproduction signal does not change. This means that the minimum reproduction signal level is obtained when the mark is large, and the maximum reproduction signal level is obtained when the space is large.

Although it is assumed that the mark has a low reflectivity, depending on the recording film, the reflectivity may increase or the reflectivity may increase due to interference. In this case, the magnitude of the signal is reversed. become.
An AGC (auto gain control) circuit is added to the playback path for fluctuations in the overall amplitude that are proportional to the signal level corresponding to all information, such as changes in the laser power and reflectance changes on the inner and outer circumferences of the optical disk. It can respond by doing.
In this method, the amplitude of the reproduction signal is sampled, and the amplification factor of the reproduction path is controlled so as to keep the magnitude constant. The reproduction signal can be applied to both binary recording and multi-value recording.
However, this AGC cannot cope with the state change of intersymbol interference. Since densification approaches the limit of optical resolution, a technology to cope with this intersymbol interference is required.

Therefore, in the former prior art described above, the characteristics of the waveform equalization (equalizer: EQ) circuit are tuned in an area where predetermined information is recorded. The EQ circuit is a circuit having a characteristic that changes the amplification factor of a specific frequency component. Tuning means making the waveform equalization characteristic appropriate or setting an appropriate value. The appropriate state of the waveform equalization circuit is a state in which intersymbol interference from adjacent marks is removed and an ideal signal level corresponding to the size of the mark can be obtained from the output of the waveform equalization circuit. Therefore, suitability means changing parameters of waveform equalization and converging the parameters to an appropriate state. Examples of parameters include the delay time of the delay element and the amplification factor of each tap output.
Specifically, the inter-symbol interference occurs near the limit of the optical resolution, the gain in the frequency domain where the reproduction signal amplitude is reduced is increased, and the attenuation rate in the frequency domain beyond the optical resolution is increased. .
Since this prior art is considered in the recording of binary information, the binarization timing of the output signal of the EQ circuit is compared with the binarization timing of the ideal reproduction signal that is known in advance, and the error The frequency characteristic of the EQ circuit is changed by controlling the tap coefficient so that the EQ circuit characteristic becomes optimum.

In the case of recording binary information, since the information is stored in the time axis direction, only the error of the binarization timing is reduced in this way.
However, in the recording of multi-value information, not only a deviation in the time axis direction but also a deviation in the amplitude direction cannot be allowed. The reproduction signal level corresponding to each multilevel information is determined from the size and depth of the mark corresponding to the multilevel information, and is the reproduction signal level at the central portion of the normal mark.
Therefore, when the time axis direction is deviated, the reproduction signal level at the mark center portion cannot be acquired, so that accurate multi-value information cannot be reproduced.

Further, if there is a deviation in the amplitude direction, it becomes difficult to separate the reproduction signal level according to the multi-value information, and it is impossible to reproduce the correct multi-value information.
In the recording of binary information, information is stored only at the time crossing the slice level. Therefore, the correction of intersymbol interference has only to be adjusted in the time axis direction.
However, when recording multilevel information, it is necessary to optimally tune the characteristics of the EQ circuit so as to correct the influence of intersymbol interference in the time axis direction and the amplitude direction.

Therefore, as a tuning method of EQ circuit characteristics in multi-level recording, the latter prior art described above records a single waveform by inserting a zero between each mark corresponding to all multi-level information, and a reproduced signal thereof. The errors in the time axis direction and the amplitude direction are obtained collectively from the level, and the frequency characteristics of the EQ circuit are optimized. In addition, the same operation is performed for a continuous pattern of each information.
This also reproduces a predetermined pattern portion, compares it with ideal data known in advance, and controls the tap coefficient of the EQ circuit so that the error is minimized.

As a result, variations between media and devices are corrected. Each of these two operations has a purpose. In multi-value recording, since a recording density with a high optical resolution limit is actively used, the inter-code interference that leaks into the reproduced signal also occurs in the marks of adjacent cells.
First, in EQ circuit tuning with a single waveform, the frequency characteristics of the basic EQ circuit are set in a situation where there is no intersymbol interference by sandwiching zero, and then in the tuning with continuous waveform, the same multi-value information is set. Only when the corresponding marks are continuous, the influence of the intersymbol interference is examined, and the frequency characteristics of the EQ circuit are corrected.

However, even here, the intersymbol interference is not sufficiently corrected. That is, although a series of marks corresponding to the same multi-value information is used as correction of intersymbol interference, it is merely an example of intersymbol interference.
For example, in the case of 8-level recording in which no mark is level 0 and the maximum mark is level 7, the intersymbol interference at the time of continuous level 1 is directed to the adjacent mark of level 1, while the continuous level 7 is performed. Intersymbol interference at the time is for level 7 adjacent marks.
There is no problem as long as the intersymbol interference from level 1 to level 7 is proportional (linear), but the linearity is lost depending on the shape of the mark and the size of the light spot.
The mark is recorded so that the reproduction signal level at the center thereof is linear according to the multilevel information, but the reproduction signal level at the adjacent mark position, that is, the intersymbol interference component is not always linear. That's what it means.

“Linear” means that 1/7 of the signal level at which the level 7 mark leaks at the adjacent mark position is equal to the signal level at which the level 1 mark leaks similarly. In this situation, not the level 1 but the level 2 mark is equal to the leaked signal level. It is impossible for an EQ circuit to correct a multi-value information sequence corresponding to all of the nonlinear components.
Since the EQ circuit can tune the frequency characteristics, the amplification factor at a specific frequency can be set, but the amplification factor can be set individually for patterns with different effects of intersymbol interference and different amplitudes at the same frequency. It is because it is not possible.
For example, if a pattern in which level 0 and level 1 are repeated and a pattern in which level 0 and level 7 are repeated are considered, both have different inter-code interference situations, and therefore there should be an optimum EQ circuit frequency characteristic. is there.
However, since both have the same frequency component, only a specific amplification factor can be set.

The method of obtaining the most appropriate value as this specific amplification factor is a method of tuning the frequency characteristic using a specific pattern that covers all multi-value information strings.
Thereby, even if there is a non-linear component, the frequency characteristic of the EQ circuit should be determined according to the average intersymbol interference component of all the information sequences.
This specific pattern extends not only to the information of the mark position to be reproduced, but also to the information of adjacent mark positions before and after that, and when the influence of intersymbol interference is larger, the range is further extended to information of the previous and subsequent mark positions. Since a combination is included, it becomes a very long information string.

For example, even if it covers only up to the front and rear mark positions in 8-level recording, 512 patterns with the cube of 8 and 3 marks are used for each pattern, so there are 3 times 1536 cells, and there is no intersymbol interference between patterns For example, a very large amount of information is required when considering a replacement area when a flaw is found in this pattern. Adding this to each minimum recording unit must be avoided because it causes a capacity reduction that cannot be ignored, contrary to the improvement in recording density.
Also, if the pattern is long, the amount of calculation until the error is minimized for tuning the tap coefficient of the EQ circuit increases, and the time also increases.
Further, in a system in which intersymbol interference reaches several cells before and after due to high density, the number of taps of the waveform equalization circuit increases and may exceed 10 taps. In this case, it takes more time to tune the tap coefficient. Therefore, there is a need for a method that can determine a more accurate frequency characteristic of an EQ circuit with a short pattern.

As described above, since multi-level recording is performed near the limit of the optical resolution, the amplitude of the high frequency component is attenuated by intersymbol interference. An EQ circuit is used for this correspondence, but in order to detect accurate multi-value information, a learning type EQ circuit that corrects the reproduction signal not only in the time axis direction but also in the amplitude direction is required. If the state of intersymbol interference is linear with respect to multilevel information, the frequency characteristics of the EQ circuit can be tuned with a relatively simple pattern. When the state of intersymbol interference is non-linear with respect to multi-value information, appropriate tuning can be performed using a learning pattern in which all information strings are stored, but it cannot be adopted because it causes capacity reduction.
The present invention has been made in view of the above points, and enables a suitable reproduction to be efficiently performed with a short pattern even on a reproduction signal having nonlinearity of intersymbol interference, and a waveform at the time of reproduction, etc. It is an object of the present invention to be able to tune the frequency characteristics of the conversion.

In order to achieve the above object, the present invention provides the following multilevel information reproducing method, multilevel information recording medium, multilevel information waveform equalizer, multilevel information reproducing apparatus, and multilevel information recording apparatus.
(1) A multi-value information reproducing method for reproducing a recording medium in which marks having different detection signal levels are formed according to multi-value information in a region virtually divided at equal intervals, and for all multi-value information Multi-value information that obtains a reproduction signal from a pattern sequence of multi-value information indicating the average level of intersymbol interference components to adjacent areas generated by the corresponding mark, and optimizes the frequency characteristics of the reproduction means for the reproduction signal Playback method.
(2) In the multi-value information reproducing method of (1), the suitability of the frequency characteristic of the reproducing means includes at least the suitability of the frequency characteristic of the waveform equalizing means.

(3) A multi-value information recording medium in which marks having different detection signal levels according to multi-value information are formed in regions virtually divided at equal intervals, and one mark corresponding to specific multi-value information A correction area including a first area where a mark corresponding to an intermediate signal level between the maximum signal level and the minimum signal level is arranged at least in the cells before and after the cell, different from the user information area. Value information recording medium.
(4) A multi-value information recording medium in which marks having different detection signal levels are formed according to multi-value information in regions virtually divided at equal intervals, and a multi-value corresponding to the upper two levels of the signal level A second region in which information is alternately arranged or multi-value information corresponding to the lower two levels is alternately arranged, and a third region in which marks corresponding to the maximum signal level and the minimum signal level are alternately arranged. A multi-value information recording medium in which the included correction area is arranged differently from the user information area.

(5) A multi-value information recording medium in which marks having different detection signal levels according to multi-value information are formed in regions virtually divided at equal intervals, and an intermediate signal between the maximum signal level and the minimum signal level A correction area including a fourth area in which marks corresponding to two signal levels of a combination that is symmetrical and closest to each other with the reference level sandwiched between the reference level and the reference level is different from the user information area. Multi-value information recording medium arranged.
(6) In the multi-value information recording medium of any one of (3) to (5), a fifth pattern in which a pattern in which two or more marks corresponding to the maximum signal level and the minimum signal level are continuous is arranged. A multi-value information recording medium in which areas are also arranged.

(7) The reproduction signal taken from the correction area of the multilevel information recording medium of any one of (3) to (6) is waveform-equalized by the waveform equalization means, and the waveform-equalized reproduction signal is stored in advance. And a means for setting a waveform equalization characteristic at the time of the waveform equalization by the waveform equalization means so as to reduce an error between the waveform equalized reproduction signal and the target value. Multi-value information waveform equalizer.
(8) In the multi-value information waveform equalizer of (7) above, the waveform equalization means stores the reproduction signal of the correction area in the storage means, and the reproduction signal from the storage means is repeated as a pseudo reproduction signal. A multi-value information waveform equalization apparatus provided with means for outputting and optimizing the waveform equalization characteristics at the time of waveform equalization by the waveform equalization means based on the pseudo reproduction signal.

(9) Optical system means for obtaining a reproduction signal from the recording medium of any one of (3) to (6), servo mechanism system means for moving and maintaining the optical system means to a target position, and the reproduction signal from the above A synchronization detection means for specifying a correction area and outputting a correction signal indicating the correction area, and a waveform equalization characteristic at the time of waveform equalization based on the correction signal output by the synchronization detection means is optimized and maintained. A multi-value information reproducing apparatus comprising waveform equalizing means for equalizing a reproduced signal and multi-value determining means for determining and outputting multi-value information from the reproduced signal waveform-equalized by the waveform equalizing means.
(10) Optical system means for forming a recording mark in accordance with the inputted multi-value information, servo mechanism system means for moving and maintaining the optical system means to a target position, user information and waveform equalization means A multi-value information recording apparatus comprising means for recording the correction pattern in a multi-value information and recording it on a recording medium.

  The multi-value information reproducing method, multi-value information recording medium, multi-value information waveform equalizer, multi-value information reproducing device and multi-value information recording device according to the present invention can be applied to a reproduced signal having non-symbol interference nonlinearity. It is possible to efficiently perform appropriate reproduction with a short pattern, and to tune the frequency characteristics of waveform equalization during the reproduction.

Hereinafter, the best mode for carrying out the present invention will be specifically described with reference to the drawings.
FIG. 1 is a diagram for explaining an outline of recording multi-value information in a multi-value information recording medium according to an embodiment of the present invention.
FIG. 1 shows an example in which multi-value information is recorded as 7-value information from “0” to “6” including no mark, as shown in FIG.
As shown in FIG. 2B, a box-shaped region (referred to as a “cell”) 1 that is virtually divided at equal intervals along a track of an optical disk (medium) that is a recording medium is equal. The recording mark (hereinafter, also simply referred to as “mark”) 2 is formed at the center. The light spot 3 is a spot of laser light irradiated for detecting the mark 2.
The mark 2 may be a phase change mark whose reflectivity mainly changes depending on the type of the optical disk, or a dye mark whose phase difference mainly changes. Further, it may be a pit of a read-only ROM disk.

User information to be recorded on the optical disc is first converted into multilevel information by multilevel modulation. A plurality of marks 2 having a size and a depth at which a reproduction signal level corresponding to the multilevel information can be detected are formed so as to be arranged at the center of the cell 1.
Although FIG. 1 shows an example in which multi-value information is recorded as 7-value information from “0” to “6” including no mark, the number of multi-value information is not particularly limited. In the following description, the case of 8 values in addition to 7 values will be described. As shown in FIG. 5C, the reproduction signal is sampled by a clock (indicated by an arrow in the figure) indicating the center of the cell 1. (D) of the same figure shows an example of a reproduction signal level curve L2 detected from the reproduction signals of the multi-value information “0” to “6”, along with the ideal reproduction signal level line L1. By determining each of the detected reproduction signal levels d, the multi-value information recorded on the optical disk is restored and reproduced.
As will be described in detail later, since the light spot 3 is larger than the cell 1, it is reproduced with the adjacent mark position information leaked, so the reproduction signal level curve L2 is an ideal reproduction signal level line. It is detected at a level different from L1.

Next, a clock detection method is shown as a feature of recording multilevel information.
Multi-value information recording on an optical disk such as a CD or DVD is greatly different from binary information recording in that there is a clock detection method. In the case of recording binary information, since the clock component is included in the data edge, the clock can be extracted as it is by reproducing the normal information area.
However, the clock cannot be extracted by recording multi-value information. Although it is clear that the clock component is not included in the data edge, the peak position that can indicate the mark center is not always detected, so a special pattern (clock mark: CM) for always extracting the clock is constant. Insert at periodic intervals.

FIG. 2 is a diagram showing an example of a data format in which a clock mark which is a special pattern for extracting a clock is inserted.
FIG. 3 is a diagram for explaining clock extraction using clock marks.
As shown in FIG. 2, the clock mark 10 is periodically inserted between user information areas (shown as “data” in the drawing) 13. It is necessary that the clock timing for extracting the cell center can be detected from the waveform. 11 is a synchronization mark, and 12 is a correction area (calibration area: CA).

The clock mark of the type shown in FIG. 3A is “00700” as shown in FIG. 3A, and when the clock mark is reproduced as shown in FIG. The reproduced signal level of the detected CM signal is as shown by a curve L3. Then, as shown in FIG. 5C, the cell center can be extracted by detecting the clock timing of the center level of the curve L3 (the bottom timing of the center level of level 7 sandwiched between levels 0). .
Although this type can directly extract the center of the cell, it has a characteristic that it is vulnerable to noise because it performs bottom detection.

On the other hand, the clock mark of the type shown in FIG. 3B is “000777”, which is a combination of three consecutive levels 0 and seven consecutive levels 7, as shown in FIG. As shown in (b), the reproduction signal level of the CM signal detected when the clock mark is reproduced is shown by a curve L4. Then, as shown in (c) of the figure, the cell boundary is detected as the clock timing by slicing the curve L4 at the center level (slice position shown in (b) of the figure). Based on this cell boundary, it is multiplied by a phase-locked loop (PLL) circuit, and the reverse edge when a cell period signal is generated becomes a clock indicating the cell center. Although this type is complex in terms of circuit, it has relatively strong characteristics against noise.
The CM signal is obtained by extracting the CM portion of the reproduction signal.

In this embodiment, the clock extraction method is not particularly concerned. However, since the clock mark pattern may exist in the user information, a means for specifying the position of the clock mark is also required.
For this purpose, the synchronization mark 11 is used.
FIG. 4 is a diagram for explaining the synchronization mark shown in FIG.
As shown in (a) of the figure, this synchronization mark is, for example, “770000777700”, and as shown in (b) of the same figure, the reproduction signal level detected when the synchronization mark is reproduced is It becomes like a curve L5.

This synchronization mark is a simple pattern that can be detected even if the clock frequency is slightly different. As shown in FIG. 2, it is preferable to arrange at least one for each minimum recording unit.
By detecting this synchronization mark pattern (synchronization pattern) first, the approximate position of the clock mark is specified, and a clock indicating the exact cell center is generated from the clock mark.
The sync signal is obtained by extracting the sync mark portion of the reproduction signal. The correction area (calibration area: CA) will be described later.

FIG. 5 is a block diagram showing a specific configuration example of the multi-value information detection circuit.
First, a synchronization signal is detected from the reproduction signal detected from the optical disk by the synchronization detection circuit 21, and a clock mark window signal indicating the approximate position of the clock mark is generated.
The clock detection circuit 22 generates a clock signal indicating the cell center from the reproduction signal based on the clock mark window signal generated from the synchronization detection circuit 21. The frequency of the clock signal may be higher than the cell frequency, and may be higher than the cell frequency including the cell center timing.
The analog / digital converter (ADC) 23 digitizes and outputs the reproduction signal based on the timing of the clock signal generated from the clock detection circuit 22.

As another method, a fixed clock that is sufficiently higher (for example, eight times or more) than the cell frequency can be used without using the synchronization detection circuit and the clock detection circuit.
In this case, high-speed processing is required for the processing after the ADC. However, the synchronization mark may be detected or the cell center may be specified by digital processing. The digital data sequence captured by the ADC is monitored, and slice level setting and peak / bottom detection are performed based on filter processing and data continuity according to the characteristics of the synchronization mark and the clock mark.
The digitized data sequence output from the ADC 23 by the waveform equalization (equalizer: EQ) circuit 24 is corrected for a high-frequency signal attenuated by intersymbol interference, and the multi-level determination circuit 25 is the closest ideal multiple. The value level is determined, and multi-value information based on the determination result is output. The multi-value information is converted into user information by a subsequent decoding circuit.

Note that the reproduced signal is superimposed with DC components below the servo band and noises above the cell frequency, and it is better to remove them with a filter.
In particular, in order to effectively use the input range to the ADC, it is desirable to cut the DC component of the ADC input by an analog filter or adjust the gain within a range where the maximum amplitude does not exceed the input range. The characteristics of these filters may be changed based on the reproduction signal in the correction area.

Next, the waveform equalization circuit (EQ circuit) 24 is preferably a digital filter whose frequency characteristics can be easily changed.
FIG. 6 is a block diagram illustrating a specific configuration example of the digital filter.
The figure shows a transversal filter, which is a delay circuit (referred to as "T") 30a to 30d called a tap, and an amplifier circuit (referred to as "C1 to C5" in the figure), where C1 to C5 are tap coefficients. Also referred to as 31) to 31e, an addition / subtraction circuit 32, a table 33, an error detection circuit 34, a coefficient setting circuit 35, a memory 36, and a switch SW.
Normally, T of the delay circuits 30a to 30d is set to a cell cycle, but may be an integer multiple of the clock interval.

In FIG. 6, a 5-tap circuit will be described. However, the number of taps is not limited, and more effects can be expected as the number of taps increases. This circuit can also be attenuated, and its amplification factor or attenuation factor is variable.
The delay circuits 30a to 30d delay and output the reproduction signal by T, and the amplification circuits 31a to 31e amplify or attenuate the reproduction signal and the outputs of the delay circuits 30a to 30d, respectively, based on the tap coefficient.
An adder / subtracter circuit (denoted as “+” in the figure) 32 calculates the outputs of the amplifier circuits 31a to 31e. Add or subtract as needed. This output result is sent to a subsequent multi-level (level) determination circuit.

Each of the amplifier circuits 31a to 31e uses a fixed amplification factor during normal data reproduction. However, when a correction signal for specifying the correction area is generated from the synchronization detection circuit or the like, since the known multi-value information is recorded on the optical disc, the error detection circuit 34 outputs the correction signal for the reproduction signal. (EQ output data) and the table (Table) data stored in the table 33 are compared, the comparison result is output, and the coefficient setting circuit 35 determines the correction signal and the table data based on the comparison result of the error detection circuit 34. The tap coefficients of the delay circuits 30a to 30d are sequentially changed so as to minimize the error.
This can continue until the tap coefficients converge if the correction area is sufficiently long. The tap coefficient can be shortened by setting a representative value as an initial value in advance. If the convergence cannot be completed, the same correction area may be accessed again and again until the tap coefficient convergence is completed.

As another method, the reproduction signal of the correction area is taken into the memory 36 which is a storage means, and the pseudo reproduction signal stored from the memory 36 is repeatedly output instead of the reproduction signal until the tap coefficient convergence is completed. It can also be made.
Thus, it is not necessary to repeat access many times, and the convergence operation can be repeated at high speed.
As the table data, an ideal signal level of the multi-value information pattern in the correction area, for example, a level obtained by equally dividing the maximum amplitude by the number of levels may be stored.

FIG. 7 is a flowchart showing an algorithm process for converging the tap coefficients.
First, the tap coefficient is initialized and held in step (indicated by “S” in the figure) 1. This value is preferably set to an average value obtained through experiments or simulations in order to accelerate convergence.
In step 2, it is determined whether a correction signal has been input. If it is determined that a correction signal has been input, i = 1 in step 3, j = 1 in step 4, n = 1 in step 5, and in step 6. A value X1 is set for the tap C1, waveform equalization processing is performed in step 7, the output signal is compared with table data, an error is detected, and in step 8, it is determined whether or not the error is minimum. If not, n = n + 1 or n-1 is set in step 14, and the process returns to step 6 to repeat the above processing. That is, the coefficient setting means sets a value Xn for a specific tap C1, performs waveform equalization processing, compares the output signal with table data, detects an error, and tap C1 until the error is minimized. Change the value of.

  When the error is minimized in step 8, it is determined in step 9 whether j = j_max. If j = j_max is not satisfied, j = j + 1 is set in step 13, and the error is determined in the same manner as described above for the next tap C2. A process of changing the coefficient so as to minimize the value is executed. Thus, after performing the above-described processing up to the final tap Cj_max and optimizing all the way, it is determined in step 10 whether i = i_max, and the above-described processing is repeated i_max times until it is determined that i = i_max. This is because there is a correlation for each tap, so that when another tap is changed, it is adjusted again to the optimum value under that condition.

It should be noted that the closer the inter-code interference component is, the greater the contribution ratio is, and the influence of the tap coefficient is different, so that it always converges. The order of the taps may be started from the ends of the amplifier circuit 31a and the amplifier circuit 31e in FIG. 6, but generally the central amplifier circuit 31c is fixed and the amplifier circuit 31b and the amplifier circuit 31e having a high contribution ratio are fixed. It is desirable to optimize first. In addition, the number of repetitions i_max and j_max is not limited and may be repeated until the minimum error is reached, but may be limited by a predetermined number.
If it is determined in step 10 that i = i_max and all adjustments are completed, this tap coefficient is held in step 11. The tap coefficient convergence method is not limited to the above-described processing, and may be a method as shown in the related art. Further, even when an EQ circuit is arranged in front of the ADC and realized in an analog manner, a variable gain amplifier or the like may be used as the amplifier circuit.

FIG. 8 is a waveform diagram showing the optical resolution and the frequency characteristics of the waveform equalization circuit.
The light spot diameter at the condensing point is determined by the wavelength and the NA (aperture ratio) of the condensing lens. When the mark and space are larger than the light spot diameter, the intensity of the reflected light is saturated, and the signal level is constant regardless of the size.
On the other hand, for a continuation of marks and spaces sufficiently smaller than the light spot diameter, the intensity of reflected light is hardly obtained, and the signal level becomes an average level. For a series of marks and spaces having a size in between, the signal level changes according to the size. This is called optical resolution (MTF), and when the continuation of the mark and space is reproduced at a fixed speed, the magnitude is considered to be the period of the detected signal. Therefore, as shown in FIG. Can be represented in the figure.

Since the amplitude of the signal is constant in the low frequency range of the MTF, it becomes 0 dB, and the attenuation of the signal amplitude starts at the mark / space length in the vicinity of the light spot diameter.
In the recording of multi-value information, this attenuated band is also used. In order to discriminate the signal level with multiple values in the same way as the signal in the non-attenuated band, the attenuation characteristic needs to be corrected. This is waveform equalization (equalizer: EQ).
As shown in FIG. 8, the characteristic of the waveform equalization circuit is such that, as indicated by the waveform equalization characteristic curve (EQ characteristic curve) L6, the attenuation characteristic due to the MTF indicated by the MTF curve L7 is corrected for a specific band. To amplify. The signal after passing through the EQ circuit has a flat amplitude characteristic in the entire data band used in multi-level recording by synthesizing the MTF and the waveform equalization circuit characteristic as shown in the synthesis characteristic curve L8. There is a need.
Note that a very low band, for example, 5 KHz or less, is a servo band for focus and tracking, and is not normally used.

FIG. 9 is a diagram for explaining the attenuation state of the reproduction signal by the MTF.
As shown in (a) of the figure, in the same portion (“0000” and “7777”) in which the multi-value information in the fifth area of (a) is the same for four consecutive cells, the signal of the reproduction signal of (b) A sufficient amplitude of the signal level is obtained as shown by a curve L10 indicating a change in level, but a portion in which multi-value information changes for each cell (“070707”) as in the third region (A). )), The amplitude is attenuated as shown by the curve L10 indicating the change in the signal level of the reproduction signal (b), and deviates from the waveform of the curve L9 indicating the change in the signal level of the ideal reproduction signal.
In particular, in the combination of a large mark and no mark in which intersymbol interference appears remarkably, the deviation is large, and it is determined that the signal level is completely different without the waveform equalization circuit.
By tuning the characteristics of the waveform equalization circuit so as to correct the influence of intersymbol interference, a waveform capable of accurately determining multi-value information such as an ideal waveform can be obtained.
This completes the description of the multi-value information recording and waveform equalization circuit.

FIG. 10 is a diagram for explaining nonlinearity of intersymbol interference.
The figure shows three types of patterns from type 1 to type 3 as mark shapes. Type 1 shown in FIG. 6A is an explanation when the intersymbol interference is linear, and the mark shape is assumed to be a circle.
Of course, this mark shape is not limited to this, and is used as the most straightforward shape for convenience of explanation.
In the cell 1, marks having a plurality of sizes are formed according to multi-value information. Here, the maximum mark (maximum recording mark) 4a, the minimum mark (minimum recording mark) 4b, and a signal that is half the maximum mark 4a. Three types of intermediate marks (intermediate recording marks) 4c capable of obtaining a level are shown. The cells before and after that are not marked in order to easily understand the influence of intersymbol interference. Further, looking at the signal intensity curves L13 to L15 in (d), the signal is bottom at the center of the corresponding cell to be reproduced, and each signal level is half of the signal level of the maximum mark 4a and that of the intermediate mark 4c. A waveform with a signal level is obtained.

In type 1, since the intersymbol interference is also linear, half of the intersymbol interference component “A” from the maximum mark 4a is also intermediate at the center position of the adjacent cell (the broken line on the right side in the figure). The intersymbol interference component “B” from the mark 4 c is obtained. With such a characteristic, the frequency characteristic of the waveform equalization circuit is determined regardless of the multi-value information.
However, the actual mark shape on the optical disk is not necessarily so, but rather it is often type 2 shown in FIG. 10B or type 3 shown in FIG.
Type 2 shown in FIG. 10B is different from type 1 in the shape of the mark, and the size of the mark is mainly small in the direction parallel to the track, that is, the length in the time axis direction. It is a pattern in which the width direction perpendicular to the track direction finally changes to a mark.
That is, three types are shown: the maximum mark 5, the minimum mark 6, and the intermediate mark 7 that can obtain a signal level half that of the maximum mark 5. The maximum mark 5 has a curve L16, the minimum mark 6 has a curve L17, and the intermediate mark has a curve L18.

Even in this case, the intensity of the reproduced signal is the same as the type 1 at the center of the corresponding cell, but is different at the center of the adjacent cell (the position of the broken line on the right side in the figure).
That is, the intersymbol interference component “B” from the intermediate mark is less than half of the intersymbol interference component “A” from the maximum mark.
That is, the change sensitivity of intersymbol interference is high with respect to a large mark, and the intersymbol interference component is in a non-linear state with respect to multilevel information.
Of course, as in the case of Type 1, the shape of the mark is a shape that can be easily imagined for convenience of explanation, but is not limited thereto.

In type 3 shown in FIG. 10C, the shape of the maximum mark of type 2 is the same, but the other marks are different, and the maximum mark 5, the minimum mark 8, and half the signal of the maximum mark 5 are used. Three types of intermediate marks 9 from which levels can be obtained are shown. The maximum mark 5 has a curve L19, the minimum mark 6 has a curve L20, and the intermediate mark has a curve L21.
Type 3 shows the opposite characteristics, and the change sensitivity of intersymbol interference is low for large marks, and the intersymbol interference component is non-linear with respect to multilevel information.
Note that only half of the signal intensity curves of type 2 in FIG. 10B and type 3 in FIG. 10C are shown for ease of explanation.
Here, the case where the spot shape is fixed and the shape of the mark is changed has been described as an example. However, when the spot shape or intensity distribution varies, more complicated nonlinearity appears and this tendency becomes remarkable. .

FIG. 11 is a diagram for explaining that complex nonlinearity appears when the spot shape and intensity distribution vary.
As shown in FIG. 5B, the spot intensity in the signal level curve L22 at the end of the adjacent cell maximum mark 4a when the spot diameter is large is a, and the spot intensity at the end of the intermediate mark 4c is b.
Similarly, the spot intensity at the signal level curve L23 at the end of the adjacent cell maximum mark 4a when the spot diameter is small is a ', and the spot intensity at the end of the intermediate mark 4c is b'. Since the powers of the two are equal, the peak intensity is lowered as the spot diameter is larger.

If the spot is large, a> a ′ and b> b ′, and the intersymbol interference increases. In addition, a / b and a '/ b' are clearly different, and nonlinearity appears in intersymbol interference.
As described above, nonlinear components are generated due to variations in mark shapes and spots.
When all data (mark size) occurs with the same probability, if the intersymbol interference is linear, the intersymbol interference component of all marks is uniquely determined from the intersymbol interference component generated by one mark. Therefore, the waveform equalization characteristic can be determined from any pattern.

However, if the intersymbol interference is non-linear, the intersymbol interference component of each mark cannot be predicted with only the intersymbol interference component from the maximum mark, and the waveform equalization characteristics may be optimized. Can not.
Therefore, it is important to optimize the waveform equalization characteristics with a pattern in which the intersymbol interference becomes an average level. The first to fifth regions are used as a pattern in which the intersymbol interference becomes an average level.

FIG. 12 is a diagram for explaining a problem when the intersymbol interference is the type 2 nonlinearity.
The horizontal dotted lines in the figure indicate ideal multilevel information signal levels. In this case, the eight levels of signal levels are indicated.
When the specific pattern for determining the tap coefficient of the waveform equalization circuit is obtained by the multi-value information “repetition of 0707” shown in FIG. 12A, the intersymbol interference component appropriate for 0707 is removed. However, since the optimization is performed to correct the intersymbol interference component of the maximum mark, the maximum mark is most sensitive to the intersymbol interference as in type 2. Would be overcorrected.

When a curve L26 having a waveform of 0707 indicated by a solid line is input as a reproduction signal, it is amplified as indicated by a line L25 indicated by a dotted line in the figure, and the multi-value information at the cell center is the reproduction signal d1 for “0”. , “7” can be detected as an ideal signal level like the reproduction signal d2.
However, when the waveform equalization circuit having the same tap coefficient is obtained with the reproduction signal of the multi-value information “repetition of 3434” shown in FIG. 12B, the curve of the waveform of 3434 indicated by the solid line as the reproduction signal is obtained. When L28 is input, the intersymbol interference component is not as large as that at the time of the maximum mark, resulting in excessive correction. After the correction, “3” is detected as a reproduction signal d2 and “4” is detected as a reproduction signal d3. The ideal reproduction signal d4 and the ideal reproduction signal d5 for “3” are respectively detected. There is a high possibility that the level will be shifted and mistaken in multi-valued judgment.

This is because the waveform equalization circuit characteristic correction pattern is not limited to 0707, but one piece of multi-value information (for example, 00700) sandwiched between zeros as in the prior art, or a series of the same multi-value information (for example, 77777). ) But the same result.
Thus, when the intersymbol interference is non-linear, even if the characteristics of the waveform equalization circuit are corrected with a pattern in which the intersymbol interference is biased as shown in the conventional example, it is not appropriate.
In this embodiment, it is possible to cope with a system in which the intersymbol interference is nonlinear by optimizing the characteristics of the reproducing means and the waveform equalization circuit using the multi-value information pattern having an average intersymbol interference. .

Next, a pattern of multilevel information that can indicate an average level of intersymbol interference components to adjacent cells generated by marks corresponding to all multilevel information will be described.
FIG. 13 is a diagram for explaining the first correction pattern.
A multi-value information string “00007777... 7xxxx” is shown as the cell data in FIG. The leading “00007777” bears the synchronization mark shown in FIG.
A portion surrounded by a dotted frame C1 in FIG. 13C corresponds to one mark corresponding to specific multi-value information and a signal level intermediate between the maximum signal level and the minimum signal level in the cells before and after the mark. This is a reproduction signal from the correction area including the first area where the mark is arranged.
In the multi-value information indicated by “x” in the cell data shown in FIG. 13A, multi-value information corresponding to a mark in which the intersymbol interference level at the adjacent cell position is an average level in the reproduction signal is entered. However, for example, 7-value recording is equivalent to level 3 and 8-value recording is equivalent to level 3.5, resulting in multi-value information that does not exist in the normal user data area.

However, since it is a desirable value for determining the frequency characteristic of the waveform equalization circuit, it is desirable to use level 3.5.
If level 3 or level 4 is used instead of level 3.5, a certain effect should be obtained to improve the waveform equalization characteristics, but that may be used.
When the occurrence probability of the mark corresponding to each multi-value information is not the same, the mark corresponding to the center value of the multi-value information may not be the average level of intersymbol interference.
Here, one mark corresponding to specific multi-value information and a mark corresponding to a signal level intermediate between the maximum signal level and the minimum signal level are arranged at least in the cells before and after the mark. In this case, the specific multi-value information is sequentially arranged from level 0 to level 7, and the intermediate signal level is level 3.5. In addition, it is assumed that the cells are arranged at least in the front and rear cells, but in FIG.

In this way, when a mark corresponding to specific multi-value information and an intermediate level mark are arranged so as to sandwich the mark, this is set as one set, and the specific multi-value information is arranged to correct the reproduction characteristics by a plurality of sets. More accurate correction.
The number of intermediate level marks is preferably equal to or greater than the number of cells affected by intersymbol interference.
Looking at the signal in which the reproduction signal is superimposed around the sample point indicating the cell center where the mark corresponding to the specific multi-value information is arranged, there are eight types of levels 0 to 7 of all the multi-value information at the sample point. In the dotted line of the adjacent cell position, intersymbol interference having the same difference occurs when the multilevel information of the sample points is level 0 and level 7 with the level 3.5 as the center. I understand that. It can be said that average intersymbol interference occurs.
As a result, the waveform equalization circuit can be efficiently optimized with average intersymbol interference without using redundant correction patterns that cover all intersymbol interference.

FIG. 14 is a diagram for explaining another example of the first correction pattern.
In FIG. 14C, the part surrounded by the dotted line frame C2 corresponds to one mark corresponding to specific multi-value information and a signal level intermediate between the maximum signal level and the minimum signal level in the cells before and after the mark. This is a reproduction signal from the correction area including the first area where the mark is arranged.
On the other hand, the cell data of (a) in FIG. 14 shows a case where level 0 is arranged as “x”.
This is a pattern presented in the prior art. Similarly, when the signal obtained by superimposing the reproduction signals around the sample point is seen, the dotted line at the adjacent cell position is shown in FIG. Thus, if the multi-value information of the sample point is level 7, “A”, and the case between level 3 and 4 is “B”, B is not half of A but greatly biased toward the level 0 side. I understand that.

Further, if the level 7 is used as “x”, the point B is biased to the level 7 conversely. From this, it is understood that when only level 0 or level 7 is used as “x”, the characteristics of the EQ circuit biased according to the nonlinearity of the intersymbol interference are appropriate.
Furthermore, since the correction pattern needs to be configured with a predetermined multi-value information pattern, it needs to be stored separately from the user information.
In particular, prior to user information, a playback position, for example, the beginning of the data area, the area where the media parameter of the innermost circumference is stored, the test area with the optimum recording power, or the like is desirable.

Next, multi-level information corresponding to the upper two levels of the signal level is alternately arranged, or multi-level information corresponding to the lower two levels is alternately arranged, and the maximum signal level and the minimum signal level are set. A correction area including a third area in which corresponding marks are alternately arranged will be described.
This will be described using type A shown in FIG.
Of type A, “00007777” is an example of a synchronization mark.
Here, two types of patterns are required. An example in which marks corresponding to the maximum signal level and the minimum signal level, which are the third region, are alternately arranged is “070707”. This represents the case where intersymbol interference is maximum.

An example in which multi-value information corresponding to the upper two levels as the second area is alternately arranged is “010101”. This represents a pattern when the intersymbol interference is small.
Further, the alternate arrangement of the multilevel information corresponding to the lower two levels of the signal level is “676767”, which represents a case where the intersymbol interference is small like “010101”.
Which type is the smallest between type 2 and type 3 shown in FIG. 10 is different, but in any case, the characteristics of the waveform equalization circuit are optimized including cases where the intersymbol interference is large and small. As a result, the same correction characteristic as that obtained when intersymbol interference exists on average can be obtained.
This can be said to be equivalent to the case where both “x” is 0 and 7 in the above description, but this configuration is possible with a shorter correction area.

Next, a fourth region in which marks corresponding to two signal levels in which the signal level is symmetrical and closest to each other with the reference level as a reference level is set to a signal level intermediate between the maximum signal level and the minimum signal level. The correction area including the will be described.
Description will be made using the correction pattern shown in Type B of FIG.
Like Type A, the head is an example of a synchronization pattern. In the case of 8-level recording, the mark corresponding to the two signal levels in which the signal level is symmetrical and the closest combination is sandwiched between the maximum signal level and the minimum signal level as a reference level. Is “343434”.

Since the reference level is level 3.5, the symmetric and closest combination is level 3 and level 4. In the case of 7-value recording, since the reference level is level 3, the symmetric and closest combinations are levels 2 and 4.
As shown in Type 2 and Type 3 of FIG. 10, when the intersymbol interference is nonlinear, the average is not the half of the level 7 intersymbol interference A, but the intermediate level intersymbol interference B.
Therefore, optimizing the characteristics of the EQ circuit using two types of marks near the intermediate mark can be said to be optimized in an almost average state of intersymbol interference.

Next, a fifth area in which a pattern in which two or more marks corresponding to the maximum signal level and the minimum signal level are continuous is arranged will be described.
A pattern in which two or more marks corresponding to the maximum signal level and the minimum signal level are continuous is arranged. This is indicated by “00007777” shown in type A in FIG. 9A and type B in FIG. 9B. "".
This may be shared if it is the same as the synchronization pattern. In addition, there is an object of optimizing the waveform equalization circuit characteristic in the low frequency, not the waveform equalization circuit characteristic in the cell frequency band.
As described above, the combination of the waveform equalization circuit characteristic and the MTF must be a flat characteristic.
Therefore, the characteristics in the cell frequency band are important, but the characteristics in the low band should be corrected together.

In addition, in the case of multi-value recording, the same meaning of the same multi-value information is the same as making one large mark or space, so that “00007777” reproduces a mark or space that sufficiently satisfies the optical resolution. Thus, no intersymbol interference occurs.
In the MTF, the waveform equalization circuit is corrected in a flat band. That is, the waveform equalization circuit is also required to have flat characteristics.
Therefore, by arranging a pattern in which two or more marks corresponding to the maximum signal level and the minimum signal level are continuously arranged, it is possible to optimize the low-frequency waveform equalizer circuit characteristics.
The number of two or more is sufficient if a mark larger than the optical resolution is formed, and it is sufficient if the continuous length of the cell is larger than the light spot diameter.

Next, embodiments of the multilevel information reproducing apparatus and multilevel information recording apparatus of the present invention will be described.
FIG. 15 is a diagram showing an example of a signal processing block around a light receiving element that receives reflected light of a light beam irradiated on an optical disc (multi-value information recording medium) and extracts various signals.
As shown in FIG. 15 (a), a track 51, which is a guide groove, is carved on a recording type optical disc, and a recording mark 52 is arranged on the track 51. The track 51 has a meandering called a wobble. The optical disk is irradiated with a light beam, and a light spot 53 is formed.
In the case of a read-only optical disc, there is no track, and a guide groove like a track is formed by arranging pits in a straight line. The optical disk is irradiated with a light beam to form a light spot.

As shown in FIG. 15B, the reflected light from the optical disk as described above is received by a four-divided light receiving element (PD) 60. The four-divided light receiving element 60 is optically partitioned into four light receiving elements 60a to 60d by a dividing line corresponding to the track tangential direction of the disk surface and the perpendicular direction thereto. For convenience, the output signals of the light receiving elements 60a to 60d are denoted by A to D, respectively.
Since the outputs of the light receiving elements 60a to 60d are current signals, they are converted into voltage signals by the I / V circuit 61, respectively. Usually, this is also called an optical system. Various signals are extracted from the signal that has undergone voltage conversion by the I / V circuit 61 in a subsequent arithmetic circuit.
First, the adder circuit (adder amplifier) 62a adds the output signals of the light receiving elements 60a to 60d, and outputs an RF signal and a multilevel reproduction signal (A + B + C + D).

The adder circuit 62b adds the output signals of the light receiving elements 60a and 60d to add the added signal (A + D), and the adder circuit 62c adds the output signals of the light receiving elements 60b and 60c to give the added signal (B + C). Then, the addition signal 62A adds the addition signal (A + D) and the addition signal (B + C), passes through the filter 64c, and outputs the low frequency signal of the track cross signal (A + B + C + D).
Further, the addition signal (A + D) and the addition signal (B + C) are subtracted by the subtraction circuit 63a, and the low frequency signal of the track error signal (A + D) − (B + C) is output through the filter 64b. This track error signal is also called a push-pull signal.

The adder circuit 62d adds the output signals of the light receiving elements 60a and 60c to add the added signal (A + C), and the adder circuit 62e adds the output signals of the light receiving elements 60b and 60d to give the added signal (B + D). Then, the subtraction circuit 63b subtracts the addition signal (A + C) and the addition signal (B + D), passes the filter 64d, and outputs a low frequency signal of the focus error signal (A + C)-(B + D).
These track cross signal, track error signal, and focus error signal are called servo signals, and are used to move the light beam and hold it at the target position (focusing, tracking).

Further, the high frequency signal (A + D) − (B + C) output from the subtracting circuit 63a and output from the filter 64a is a wobble signal.
Here, the calculation is performed by the same circuit as the track error signal, but of course, the calculation may be performed by another circuit, or various correction circuits may be inserted before the subtraction circuit (subtraction amplifier).
Usually, the above is called a servo signal detection system and is used to control the mechanism system. Further, since it is desirable to calculate the reproduction (RF) signal by a separate circuit in a high band, the calculation is performed by adding four signals directly after I / V. The CM signal is obtained from this RF signal.

The simplest calculation method for various signals is shown here, but the division shape of the light receiving element (PD) is not limited to this, and it may be divided more finely according to the number of light beams and the optical path. On the contrary, it may be as small as two or three. What is necessary is just to optimize a signal calculation according to each light reception form. It is also possible to detect various signals from a plurality of main and sub light beams.
For example, the track error signal is a case of a three-beam method in which three light beams are received and calculated, a differential push-pull (DPP) method, or the like. The track cross signal can also be calculated with three beams. The track error signal may be a differential phase detection (DPD) method. The focus system may be calculated from another light receiving element such as a knife edge method.
That is, the calculation method may be made appropriate by the detection method, and the method and means for extracting a signal from the disk are not a problem.

FIG. 16 is a block diagram showing the configuration of an information recording / reproducing apparatus as an embodiment of the multi-value information reproducing apparatus and multi-value information recording apparatus of the present invention.
This information recording / reproducing apparatus includes an optical pickup 70 equipped with an optical system, and a plurality of motors such as a seek motor that moves the optical pickup 70 and a spindle motor 71 that rotates the optical disk 50 (motors other than the spindle motor 71 are not shown). And a mechanical system including loading (not shown) for setting the optical disc 50, and various electric systems. However, the electrical system may be included in the mechanism system or the optical system.

The optical pickup 70 includes a semiconductor laser light source 90 such as a laser diode (LD) that generates laser light, a lens 91 that collimates the light beam of the laser light generated by the semiconductor laser light source 90, a prism 92, and quadrant light reception. A lens 95 for guiding laser light to each light receiving element of the element 96, an objective lens 93 for condensing a spot of the laser light on the optical disk 50, and an objective lens 93 for tracking the spot to a desired position on the optical disk 50. An actuator 94 for controlling the position, a four-divided light receiving element 93, and an I / V circuit 97 are mounted.
The four-divided light receiving element 93 corresponds to the four-divided light receiving element 60 in FIG. 15, and the I / V circuit 97 corresponds to the I / V circuit 61 in FIG.

  The electrical system includes the following. At the time of recording, the system controller 81 receives data (information) to be recorded on the optical disc 50 from outside the information recording / reproducing apparatus (for example, a host computer such as a personal computer), and encodes it into an information sequence to be recorded on the optical disc 50 by the encoder 80. Conversion such as modulation. The laser driving unit 73 determines an appropriate laser light emission timing and intensity for recording on the optical disc 50 from the information sequence, and causes the semiconductor laser light source 90 of the optical pickup 70 to emit the laser light. At the time of reproduction, the laser driving unit 73 causes the semiconductor laser light source 90 to emit stable laser light with the intensity for reproduction.

The reflected signal from the optical disk 50 is photoelectrically converted by a four-divided light receiving element (PD) 96 and converted to a voltage signal that can be easily calculated by an I / V circuit 97. The PD 96 and the I / V circuit 97 may be integrated.
Thereafter, signal calculations such as a reproduction signal by the reproduction signal detection unit 74, a wobble signal by the wobble signal detection unit 75, and a servo signal by the servo signal detection unit 76 are performed. Note that various signal operations may be performed in the output (current) state of the PD 96 and then converted into a voltage signal.

The reproduction signal detection unit 74 corresponds to the addition circuit 62a in FIG. 15, the wobble signal detection unit 75 corresponds to the addition circuits 62b and 62c, the subtraction circuit 63a and the filter 64a in FIG. 15, and the servo signal detection unit 76 This corresponds to the addition circuits 62c to 62f, the subtraction circuits 63a and 63b, and the filters 64b to 64d in FIG. Further, the reproduction signal detection unit 74 corresponds to the multi-value information detection circuit shown in FIG.
Although the detection of the wobble signal is described independently, it may be generated from the internal signal of the servo signal detector 76. The detected wobble signal is input to the wobble demodulated signal processing unit 79.

The wobble demodulated signal processing unit 79 includes a recording clock generating unit, a demodulating unit, and the like, and detects a wobble synchronization signal, address information, a recording clock signal, layer information, and the like. These wobble synchronization signal, address information, recording clock signal, and layer information are used by the system controller 81 and the encoder 80 for current position acquisition processing. The recording clock signal is a reference signal that is also used by the encoder 80 and the DSP 77 when recording data.
The servo signal is subjected to various calculations by the servo signal detector 76, and the DSP 77 calculates the amount of movement of the optical pickup 70 and the actuator 94 from the error between the spot position and the target position, thereby causing the laser beam spot to follow the desired position. The seek motor and actuator 94 are operated as much as possible. Further, the optical disk rotation speed is detected based on the clock signal detected from the wobble signal and the reproduction signal, and the rotation speed of the spindle motor 71 with respect to the motor drive unit 72 is controlled in comparison with the target speed.

  During reproduction, the reproduction signal detection unit 74 extracts a reproduction signal that is a high-frequency signal component using a filter. Furthermore, CM detection means is also provided, and quantization is performed by the ADC using the obtained reproduction clock. The reproduced clock signal obtained by the reproduced signal detection unit 74 is used as a reference signal for the reproduction system, but a wobble clock may be used for the coarse synchronization pull-in. Based on the quantized data, waveform equalization is performed by a waveform equalization circuit to determine a multilevel level. Then, the decoder 78 performs various demodulations and decodings, and converts them into reproduction information. The reproduction information is transferred to the outside (for example, a host computer such as a personal computer) through the system controller 81.

In this embodiment, a reproduction signal is obtained from a pattern string of multilevel information indicating the average level of intersymbol interference components to adjacent cells generated by marks corresponding to all the multilevel information, and reproduction means for the reproduction signal is obtained. Since the frequency characteristics of the system are optimized, even in systems where intersymbol interference has non-linearity with respect to multilevel information, the frequency characteristics of the playback system are accurately optimized with a short correction pattern, and the reliability and compatibility are high. A playback system can be constructed.
In addition, if the optimization of the frequency characteristics of the reproduction means includes at least the optimization of the frequency characteristics of the waveform equalization means, the waveform equalization means, which is the main means for correcting intersymbol interference, Since it takes time to optimize the coefficients, a short and efficient correction pattern can be used to greatly reduce the time required for optimization, and a highly reliable and compatible playback system can be constructed. Regeneration can be expected.

Further, a correction area including a first area in which one mark corresponding to specific multi-value information and a mark corresponding to a signal level intermediate between the maximum signal level and the minimum signal level are arranged in at least cells before and after the mark. If it is arranged differently from the user information area, the characteristics of the playback means can be optimized accurately and efficiently with a short pattern even in a system with nonlinearity in intersymbol interference. A highly reliable optical disc (media) can be provided.
Also, the second area where the multi-level information corresponding to the upper two levels of the signal level is alternately arranged, or the multi-level information corresponding to the lower two levels is alternately arranged, and the maximum signal level and the minimum signal level are supported. If the correction area including the third areas in which the marks are alternately arranged is arranged differently from the user information area, even in a system having non-symbol interference, Since the characteristics of the reproducing means can be optimized efficiently with a short pattern, a highly reliable and compatible medium can be provided.

In addition, a fourth region in which marks corresponding to two signal levels in a combination in which the signal level is symmetrical and nearest to each other with the reference level as a reference level is set to a signal level intermediate between the maximum signal level and the minimum signal level. If the included correction area is arranged differently from the user information area, the characteristics of the reproduction means can be optimized accurately and efficiently with a short pattern even in a system having nonlinearity in intersymbol interference. Can provide reliable and compatible media.
In addition to the above-described effects, if a fifth region in which a pattern in which two or more continuous marks corresponding to the maximum signal level and the minimum signal level are arranged is also provided, a waveform in a low region, etc. Optimization can be performed, and reliability and compatibility can be improved.

Further, the reproduction signal captured from the correction area of the multi-value information recording medium is input to the waveform equalization means, the output is compared with the target value stored in advance, and the waveform equalization means of which the error is reduced is reduced. By setting the characteristics, the characteristics of the waveform equalization means can be optimized accurately and efficiently with a short pattern even in a system with non-symbol interference. Multi-value information that can be determined can be generated.
Further, in the waveform equalization means, if the reproduction signal of the correction area is captured by the storage means and the pseudo reproduction signal repeatedly output from the storage means is used to optimize the characteristics of the waveform equalization means, The correction area required to optimize the characteristics of the waveform equalization means can be reduced, and repeated access to the correction area can be avoided, so that the system can ensure storage capacity and satisfy high-speed playback performance. Is possible.

Furthermore, an optical system that obtains a reproduction signal from the multi-value recording medium, a servo mechanism system that moves and maintains the optical system to a target position, and outputs a correction signal that specifies the correction area from the reproduction signal and indicates the correction area. It is provided with synchronization detection means, waveform equalization means for optimizing and maintaining characteristics based on the correction signal, and multi-value determination means for demodulating multi-value information from the output of the waveform equalization means, so as to reproduce user information. Thus, it is possible to provide a multi-value information reproducing apparatus with high reliability and compatibility.
Furthermore, an optical system for forming a recording mark according to the inputted multi-value information, and a servo mechanism system for moving and maintaining the optical system to a target position, not only user information but also waveform equalizing means If the correction pattern is included in the multi-value information and recording is performed on the medium, a multi-value information recording apparatus with high reliability and compatibility can be provided, and further, the medium can be provided.

  The multi-value information reproducing method, multi-value information recording medium, multi-value information waveform equalizer, multi-value information reproducing device, multi-value information recording device according to the present invention does not depend on the laser wavelength or the parameters of the optical system, In the embodiment, a DVD is taken as an example, but a blue laser or a super-resolution technique can be applied.

It is a figure where it uses for description of the outline | summary of recording of the multi-value information in the multi-value information recording medium of the Example of this invention. It is a figure which shows an example of the data format which inserted the clock mark which is a special pattern for extracting a clock. It is a figure where it uses for description of the clock extraction by a clock mark. It is a figure where it uses for description of the synchronous mark shown in FIG. It is a block diagram which shows the specific structural example of a multi-value information detection circuit.

It is a block diagram which shows the specific structural example of a digital filter. It is a flowchart figure which shows the process of the algorithm which converges a tap coefficient. It is a wave form diagram which shows the frequency characteristic of an optical resolution and a waveform equalization circuit. It is a figure where it uses for description of the attenuation | damping state of the reproduction signal by MTF. It is a figure with which it uses for description of the nonlinear of intersymbol interference.

It is a figure with which it uses for description that a complicated nonlinearity appears when the shape and intensity distribution of a spot vary. It is a figure where it uses for demonstrating the malfunction when intersymbol interference is nonlinear of the said type 2. FIG. It is a figure where it uses for description of a 1st correction pattern.

It is a figure where it uses for description of the other example of the 1st correction pattern. It is a figure which shows an example of the signal processing block around the light receiving element which receives the reflected light of the light beam irradiated to the optical disk, and extracts various signals. It is a block diagram which shows the structure of the information recording / reproducing apparatus which is one Example of the multi-value information reproducing | regenerating apparatus and multi-value information recording device of this invention.

Explanation of symbols

1: cell 2, 52: recording mark 3, 53: light spot 4a, 5: maximum recording mark 4b, 6, 8: minimum recording mark 4c, 7, 9: intermediate recording mark 10: clock mark 11: synchronization mark 12: Correction area (calibration area: CA) 13: User information area 21: Synchronization detection circuit 22: Clock detection circuit 23: Analog-digital converter (ADC) 24: Waveform equalization circuit (EQ circuit) 25: Multi-value determination circuit 30a to 30d: delay circuit 31a to 31e: amplifier circuit 32: addition / subtraction circuit 33: table 34: error detection circuit 35: coefficient setting circuit 36: memory 50: optical disk 51: track 60, 96: four-divided light receiving element (PD) 60a ˜60d: light receiving element 61, 97: I / Circuits 62a to 62f: Addition circuits 63a and 63b: Subtraction circuits 64a to 64d: Filter 70: Optical pickup 71: Spindle motor 72: Motor drive unit 73: Laser drive unit 74: Reproduction signal detection unit 75: Wobble signal detection unit 76: Servo signal detector 77: DSP 78: Decoder 79: Wobble demodulated signal processor 80: Encoder 81: System controller 90: Semiconductor laser light source 91, 95: Lens 92: Prism 93: Objective lens 94: Actuator

Claims (10)

A multi-value information reproduction method for reproducing a recording medium in which marks having different detection signal levels are formed in regions virtually divided at equal intervals according to multi-value information,
A reproduction signal is obtained from a pattern sequence of multi-value information indicating an average level of intersymbol interference components to adjacent areas generated by marks corresponding to all the multi-value information, and the frequency characteristic of the reproducing means is obtained with respect to the reproduction signal. A multi-value information reproducing method characterized by adapting.   2. The multi-value information reproducing method according to claim 1, wherein the optimization of the frequency characteristics of the reproducing means includes at least optimization of the frequency characteristics of the waveform equalizing means. A multi-value information recording medium in which marks having different detection signal levels according to multi-value information are formed in regions virtually divided at equal intervals,
A correction area including one mark corresponding to specific multi-value information and a first area in which a mark corresponding to an intermediate signal level between the maximum signal level and the minimum signal level is arranged in at least cells before and after the mark. A multi-value information recording medium characterized by being arranged differently from an information area. A multi-value information recording medium in which marks having different detection signal levels according to multi-value information are formed in regions virtually divided at equal intervals,
A second area in which multilevel information corresponding to the upper two levels of the signal level is alternately arranged or multilevel information corresponding to the lower two levels is alternately arranged, and a mark corresponding to the maximum signal level and the minimum signal level A multi-value information recording medium characterized in that a correction area including third areas alternately arranged is arranged differently from a user information area. A multi-value information recording medium in which marks having different detection signal levels are formed according to multi-value information in regions virtually divided at equal intervals,
It includes a fourth region in which marks corresponding to two signal levels in a combination of the closest and the closest signal level are arranged with the reference signal level between the maximum signal level and the minimum signal level as a reference level. A multi-value information recording medium, wherein the correction area is arranged differently from the user information area.   6. A fifth area in which a pattern in which two or more marks corresponding to the maximum signal level and the minimum signal level are continuously arranged is also arranged. The multi-value information recording medium according to item.   7. A reproduction signal fetched from the correction area of the multi-value information recording medium according to claim 3 is waveform-equalized by a waveform equalization means, and stored in advance as the waveform-equalized reproduction signal. Comparing with a target value, means for setting a waveform equalization characteristic at the time of waveform equalization by the waveform equalization means so as to reduce an error between the waveform equalized reproduction signal and the target value A multi-value information waveform equalizer characterized by.   The reproduction signal of the correction area is stored in the storage means in the waveform equalization means, the reproduction signal is repeatedly output from the storage means as a pseudo reproduction signal, and the waveform by the waveform equalization means based on the pseudo reproduction signal 8. The multi-value information waveform equalization apparatus according to claim 7, further comprising means for optimizing waveform equalization characteristics at the time of equalization.   7. Optical system means for obtaining a reproduction signal from the recording medium according to claim 3, servo mechanism system means for moving the optical system means to a target position and maintaining the correction area from the reproduction signal And a synchronization detection means for outputting a correction signal indicating the correction area and regenerating the waveform equalization characteristics at the time of waveform equalization based on the correction signal output by the synchronization detection means. A multi-value comprising waveform equalizing means for equalizing a signal and multi-value determining means for determining and outputting multi-value information from a reproduction signal waveform-equalized by the waveform equalizing means Information playback device.   Optical system means for forming recording marks in accordance with input multi-value information, servo mechanism system means for moving and maintaining the optical system means to a target position, and correction patterns for user information and waveform equalization means A multi-value information recording apparatus comprising: means for recording multi-value information in a recording medium.

Images