WO2010106810A1 - Equalizer and equalization method - Google Patents

Equalizer and equalization method Download PDF

Info

Publication number
WO2010106810A1
WO2010106810A1 PCT/JP2010/001971 JP2010001971W WO2010106810A1 WO 2010106810 A1 WO2010106810 A1 WO 2010106810A1 JP 2010001971 W JP2010001971 W JP 2010001971W WO 2010106810 A1 WO2010106810 A1 WO 2010106810A1
Authority
WO
WIPO (PCT)
Prior art keywords
unit
signal
equalization
linear
delay
Prior art date
Application number
PCT/JP2010/001971
Other languages
French (fr)
Japanese (ja)
Inventor
速水淳
糸長誠
Original Assignee
日本ビクター株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from JP2009072812A external-priority patent/JP5136489B2/en
Priority claimed from JP2010049358A external-priority patent/JP5136577B2/en
Application filed by 日本ビクター株式会社 filed Critical 日本ビクター株式会社
Priority to US13/256,608 priority Critical patent/US8611411B2/en
Priority to KR1020127033156A priority patent/KR101495979B1/en
Priority to CN201080012536.5A priority patent/CN102356432B/en
Publication of WO2010106810A1 publication Critical patent/WO2010106810A1/en

Links

Images

Classifications

    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B20/00Signal processing not specific to the method of recording or reproducing; Circuits therefor
    • G11B20/10Digital recording or reproducing
    • G11B20/10009Improvement or modification of read or write signals
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B20/00Signal processing not specific to the method of recording or reproducing; Circuits therefor
    • G11B20/10Digital recording or reproducing
    • G11B20/10009Improvement or modification of read or write signals
    • G11B20/10037A/D conversion, D/A conversion, sampling, slicing and digital quantisation or adjusting parameters thereof
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B20/00Signal processing not specific to the method of recording or reproducing; Circuits therefor
    • G11B20/10Digital recording or reproducing
    • G11B20/10009Improvement or modification of read or write signals
    • G11B20/10046Improvement or modification of read or write signals filtering or equalising, e.g. setting the tap weights of an FIR filter
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B20/00Signal processing not specific to the method of recording or reproducing; Circuits therefor
    • G11B20/10Digital recording or reproducing
    • G11B20/10009Improvement or modification of read or write signals
    • G11B20/10046Improvement or modification of read or write signals filtering or equalising, e.g. setting the tap weights of an FIR filter
    • G11B20/10055Improvement or modification of read or write signals filtering or equalising, e.g. setting the tap weights of an FIR filter using partial response filtering when writing the signal to the medium or reading it therefrom
    • G11B20/1012Improvement or modification of read or write signals filtering or equalising, e.g. setting the tap weights of an FIR filter using partial response filtering when writing the signal to the medium or reading it therefrom partial response PR(1,2,2,2,1)
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B20/00Signal processing not specific to the method of recording or reproducing; Circuits therefor
    • G11B20/10Digital recording or reproducing
    • G11B20/10009Improvement or modification of read or write signals
    • G11B20/10046Improvement or modification of read or write signals filtering or equalising, e.g. setting the tap weights of an FIR filter
    • G11B20/10212Improvement or modification of read or write signals filtering or equalising, e.g. setting the tap weights of an FIR filter compensation for data shift, e.g. pulse-crowding effects
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B20/00Signal processing not specific to the method of recording or reproducing; Circuits therefor
    • G11B20/10Digital recording or reproducing
    • G11B20/10009Improvement or modification of read or write signals
    • G11B20/10268Improvement or modification of read or write signals bit detection or demodulation methods
    • G11B20/10287Improvement or modification of read or write signals bit detection or demodulation methods using probabilistic methods, e.g. maximum likelihood detectors
    • G11B20/10296Improvement or modification of read or write signals bit detection or demodulation methods using probabilistic methods, e.g. maximum likelihood detectors using the Viterbi algorithm
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B2220/00Record carriers by type
    • G11B2220/20Disc-shaped record carriers
    • G11B2220/25Disc-shaped record carriers characterised in that the disc is based on a specific recording technology
    • G11B2220/2537Optical discs
    • G11B2220/2541Blu-ray discs; Blue laser DVR discs
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B2220/00Record carriers by type
    • G11B2220/20Disc-shaped record carriers
    • G11B2220/25Disc-shaped record carriers characterised in that the disc is based on a specific recording technology
    • G11B2220/2537Optical discs
    • G11B2220/2562DVDs [digital versatile discs]; Digital video discs; MMCDs; HDCDs
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B2220/00Record carriers by type
    • G11B2220/20Disc-shaped record carriers
    • G11B2220/25Disc-shaped record carriers characterised in that the disc is based on a specific recording technology
    • G11B2220/2537Optical discs
    • G11B2220/2579HD-DVDs [high definition DVDs]; AODs [advanced optical discs]

Definitions

  • the present invention relates to an equalizer, and more particularly to an equalizer and an equalization method for equalizing nonlinear distortion.
  • the disk control circuit rotates the optical disk at a predetermined rotation speed, and the optical pickup reads a reproduction signal recorded on the optical disk.
  • the reproduction signal is amplified by a preamplifier and then amplified to a predetermined amplitude by an AGC circuit or the like.
  • the reproduction signal is A / D converted, waveform equalized by a linear waveform equalization circuit, and then decoded by Viterbi decoding.
  • image data and music data recorded on the optical disc are reproduced.
  • a nonlinear waveform equalization circuit is used to reduce nonlinear distortion.
  • a neural network is used to realize a nonlinear waveform equalization circuit (see, for example, Patent Document 1).
  • a training signal is recorded at a predetermined location on the optical disc, and a coefficient in the neural network is determined using an output corresponding to the training signal as a teacher signal. Therefore, since the training signal is recorded in advance on the optical disc, the utilization efficiency of the optical disc is reduced. In addition, since the coefficient is fixed after the learning operation is finished, it is difficult to follow the fluctuation of the reproduction waveform characteristic in the plane of the optical disc. Furthermore, there are power fluctuations and the like depending on the recorder that records data on the optical disc, but it is also difficult to follow them. For this reason, it is required to adaptively reduce the linear distortion without using a training signal as well as adaptively reducing the non-linear distortion of the reproduction signal caused by the improvement of the recording density and the recording power fluctuation.
  • the present invention has been made in view of such circumstances, and an object of the present invention is to provide a technique for reducing non-linear distortion of a reproduction signal caused by improvement in recording density, recording power fluctuation or the like without a training signal.
  • an equalizer sequentially linearly equalizes a signal to be processed and linearly equalizes a signal that is linearly equalized in the linear equalization unit.
  • the temporary determination unit may execute a temporary determination according to the partial response rule. In this case, the provisional determination according to the partial response rule is executed, so that the partial response process can be handled.
  • a delay unit that delays the signal that is nonlinearly equalized in the nonlinear equalizer may be further provided.
  • the delay unit executes delay over a period corresponding to the difference between the processing delay in the temporary determination unit and the processing delay in the nonlinear equalization unit, and the nonlinear equalization unit performs the temporary determination in the signal delayed in the delay unit and the temporary determination unit.
  • a plurality of coefficients may be derived based on the difference from the signal. In this case, since the delay is executed over a period corresponding to the difference between the processing delay in the provisional determination unit and the processing delay in the nonlinear equalization unit, the timing of the delayed signal and the provisionally determined signal can be matched.
  • the non-linear equalization unit may newly derive a plurality of coefficients when the difference becomes larger than the threshold value. In this case, since a plurality of coefficients are newly derived when divergence is detected, deterioration of equalization characteristics can be suppressed.
  • Another aspect of the present invention is an equalization method.
  • an input signal is sequentially linearly equalized, a linearly equalized signal is sequentially provisionally determined, a plurality of coefficients are derived using the provisionally determined signal as a teacher signal, and a plurality of coefficients are included. And sequentially performing nonlinear equalization on the linearly equalized signal.
  • the equalizer includes an input unit that sequentially inputs a signal to be processed, a linear equalization unit that sequentially linearly equalizes the signal input at the input unit, and an input that is performed in parallel with the linear equalization in the linear equalization unit.
  • An adaptive nonlinear equalization unit that sequentially performs nonlinear equalization on the input signal, an addition unit that adds the nonlinear equalization signal in the adaptive nonlinear equalization unit, and the linear equalization signal in the linear equalization unit, and addition
  • a provisional determination unit that sequentially provisionally determines signals added in the unit.
  • the adaptive nonlinear equalization unit derives a plurality of coefficients using the signal provisionally determined by the provisional determination unit as a teacher signal, and performs nonlinear equalization based on the plurality of coefficients.
  • linear equalization and non-linear equalization are performed in parallel, the equalization signals from both are added, and the result of provisional determination of the addition signal is used as a teacher signal, so that no training signal is used.
  • a coefficient for nonlinear equalization can be derived.
  • the multi-stage tap included in the linear equalization unit and the multi-stage tap in the adaptive nonlinear equalization unit may be shared. In this case, since multistage taps are shared, an increase in circuit scale can be suppressed.
  • the temporary determination unit may execute a temporary determination according to the partial response rule. In this case, the provisional determination according to the partial response rule is executed, so that the partial response process can be handled.
  • a determination unit that determines convergence of a plurality of coefficients in the adaptive nonlinear equalization unit may be further provided.
  • the adder outputs the signal linearly equalized by the linear equalizer to the temporary determiner until convergence is determined by the determiner, and after the convergence is determined by the determiner, the added signal is temporarily determined. You may output to a part. In this case, since the nonlinear equalized signal is not output to the provisional determination unit until the nonlinear equalization coefficient is converged, deterioration of the accuracy of the provisional determination can be suppressed.
  • a linear delay unit may further include a first delay unit that delays the linearly equalized signal, and a second delay unit that delays the nonlinear equalized signal in the adaptive nonlinear equalization unit.
  • the first delay unit executes a delay over a period according to the processing delay in the temporary determination unit
  • the second delay unit executes a delay over a period according to the processing delay in the temporary determination unit
  • the adaptive nonlinear equalization unit The linear equalization unit derives a plurality of coefficients based on the difference between the sum of the signal delayed in the first delay unit and the signal delayed in the second delay unit and the signal provisionally determined in the temporary determination unit.
  • An addition unit that adds a signal that is linearly equalized in the linear equalization unit and a signal that is nonlinearly equalized in the adaptive nonlinear equalization unit, and a delay unit that delays the signal output from the addition unit may be further provided.
  • the delay unit performs a delay over a period corresponding to the processing delay in the provisional determination unit, and the adaptive nonlinear equalization unit is based on a difference between the signal delayed in the delay unit and the signal provisionally determined in the provisional determination unit.
  • the plurality of coefficients may be derived, and the linear equalization unit may perform the linear equalization using the plurality of coefficients, and may use a fixed value as the plurality of coefficients. In this case, since the coefficient for linear equalization is a fixed value, the stability of the equalization process can be improved.
  • the adaptive nonlinear equalization unit may newly derive a plurality of coefficients when detecting the divergence of the plurality of coefficients. In this case, since a plurality of coefficients are newly derived when divergence is detected, deterioration of equalization characteristics can be suppressed.
  • Still another aspect of the present invention is an equalization method.
  • the step of linearly equalizing the input signal, the step of sequentially nonlinearly equalizing the input signal in parallel with the linear equalization, and the nonlinear equalized signal and the linearly equalized signal are added.
  • a step of tentatively determining the added signals sequentially.
  • a plurality of coefficients are derived using the temporarily determined signal as a teacher signal, and nonlinear equalization is executed based on the plurality of coefficients.
  • the present invention it is possible to reduce the non-linear distortion of the reproduction signal caused by the improvement of the recording density, the recording power fluctuation or the like without the training signal.
  • FIG. 6 is a diagram illustrating a configuration of a path memory unit in FIG. 5. It is a figure which shows the data structure of the table memorize
  • Embodiment 1 of the present invention reproduces a signal recorded on a recording medium such as an optical disc, equalizes a reproduced signal (hereinafter referred to as “reproduced signal”) by a partial response method, and equalizes the signal.
  • the present invention relates to a playback apparatus for decoding (hereinafter referred to as “equalized signal”).
  • equalized signal As described above, as the recording capacity of the optical disk increases, the influence of nonlinear distortion that cannot be removed by the linear waveform equalizer increases. In order to remove the nonlinear distortion, a neural network as a nonlinear equalizer is effective, but it is necessary to learn and converge with a training signal. Therefore, in order to reduce the non-linear distortion of the reproduction signal without the training signal, the reproduction apparatus according to the present embodiment executes the following processing.
  • the playback device arranges a linear waveform equalizer in series before the nonlinear waveform equalizer.
  • the reproduction apparatus inputs an equalized signal from a linear waveform equalizer (hereinafter referred to as “linear equalized signal”) to a nonlinear waveform equalizer, and then receives an equalized signal (hereinafter referred to as a nonlinear waveform equalizer). , Referred to as “nonlinear equalized signal”) to the Viterbi decoder.
  • the linear equalization signal is also input to the temporary determination unit, and is temporarily determined by the temporary determination unit.
  • the provisionally determined signal (hereinafter referred to as “temporary determination signal”) is input to the linear waveform equalizer and the nonlinear waveform equalizer as a teacher signal.
  • the linear waveform equalizer and the nonlinear waveform equalizer perform equalization processing by deriving tap coefficients based on the teacher signal.
  • a neural network is used for the nonlinear equalizer, but according to the above configuration, the neural network is learned without using a training signal.
  • a linear equalization signal, a non-linear equalization signal, and a temporary determination signal are used for derivation of tap coefficients in the linear waveform equalizer and the non-linear waveform equalizer.
  • the output timing is different from the provisional determination signal. Therefore, in order to match these timings, the playback device delays the linear equalization signal and the nonlinear equalization signal in order to derive the tap coefficient.
  • FIG. 1 shows the configuration of a playback apparatus 100 according to Embodiment 1 of the present invention.
  • the playback device 100 includes an optical disc 10, an optical disc drive unit 12, an optical pickup 14, a preamplifier unit 16, an AGC unit 18, a PLL (Phase Locked Loop) unit 20, an A / D conversion unit 22, a processing unit 24, and a control unit 26.
  • an optical disc 10 an optical disc drive unit 12
  • an optical pickup 14 includes an optical pickup 14
  • a preamplifier unit 16 an AGC unit 18, a PLL (Phase Locked Loop) unit 20, an A / D conversion unit 22, a processing unit 24, and a control unit 26.
  • PLL Phase Locked Loop
  • the optical disc 10 is a recording medium configured to be detachable from the playback device 100.
  • the optical disk 10 corresponds to various types such as CD, DVD, BD, and HD DVD.
  • the optical disk 10 is particularly targeted for a case where the nonlinear distortion is large enough to affect reproduction.
  • the optical disk drive unit 12 is a motor for rotating the optical disk 10 at a predetermined rotation speed.
  • the optical pickup 14 reads a signal to be processed from the optical disc 10 and performs photoelectric conversion and amplification on the signal. The resulting signal corresponds to the “reproduction signal” described above.
  • the optical pickup 14 outputs a reproduction signal to the preamplifier unit 16.
  • the preamplifier unit 16 amplifies the reproduction signal
  • the AGC unit 18 amplifies the reproduction signal from the preamplifier unit 16 to a predetermined amplitude.
  • the AGC unit 18 outputs the amplified reproduction signal to the PLL unit 20, and the PLL unit 20 detects a clock from the reproduction signal.
  • the A / D converter 22 performs analog / digital conversion of the reproduction signal based on the clock detected by the PLL unit 20.
  • the processing unit 24 performs equalization processing and decoding processing on the reproduction signal (hereinafter also referred to as “reproduction signal”) analog / digital converted by the A / D conversion unit 22. Details of the processing unit 24 will be described later.
  • This configuration can be realized in terms of hardware by a CPU, memory, or other LSI of any computer, and in terms of software, it can be realized by a program loaded in the memory, but here it is realized by their cooperation.
  • Draw functional blocks Accordingly, those skilled in the art will understand that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof.
  • FIG. 2 shows the configuration of the processing unit 24.
  • the processing unit 24 includes a linear equalization unit 44, a first delay unit 32, a provisional determination unit 30, a nonlinear equalization unit 46, a second delay unit 34, a first addition unit 40, a second addition unit 42, and a Viterbi decoding unit 38. including.
  • the signal includes a linear equalization error signal 300, a non-linear equalization error signal 302, and a provisional determination signal 306.
  • the reproduction signal sampled for each bit clock in the A / D conversion unit 22 in FIG. 1 is sequentially input to the linear equalization unit 44.
  • the linear equalizer 44 sequentially performs linear equalization on the input reproduction signal.
  • the linear equalization unit 44 is configured by a transversal filter, delays the reproduction signal by a multistage tap, multiplies the output from the multistage tap and a plurality of tap coefficients, and adds the multiplication results.
  • the addition result corresponds to the linear equalization signal described above.
  • the linear equalization unit 44 receives the linear equalization error signal 300 from the first addition unit 40 described later, and derives a plurality of tap coefficients based on the linear equalization error signal 300.
  • an adaptive algorithm such as a LMS (Least Mean Square) algorithm is used to derive a plurality of tap coefficients.
  • the linear equalization unit 44 outputs a linear equalization signal to the first delay unit 32, the provisional determination unit 30, and the non-linear equalization unit 46.
  • the non-linear equalization unit 46 receives the linear equalization signal from the linear equalization unit 44 and sequentially performs non-linear equalization on the linear equalization signal.
  • the nonlinear equalization unit 46 is configured by a neural network. The result of nonlinear equalization in the nonlinear equalization unit 46 corresponds to the aforementioned nonlinear equalization signal. Further, the nonlinear equalization unit 46 receives the nonlinear equalization error signal 302 from the second addition unit 42 described later, and based on the nonlinear equalization error signal 302, a plurality of taps used in the neural network. Deriving coefficients.
  • the non-linear equalization unit 46 uses the temporary determination signal as the teacher signal. It can be said that a plurality of coefficients are derived.
  • the nonlinear equalization unit 46 outputs the nonlinear equalization signal to the second delay unit 34 and the Viterbi decoding unit 38.
  • the Viterbi decoding unit 38 receives the nonlinear equalization signal from the nonlinear equalization unit 46 and executes Viterbi decoding on the nonlinear equalization signal.
  • the Viterbi decoding unit 38 has a branch metric calculation circuit that calculates a branch metric from a non-linear equalization signal, a path metric calculation circuit that calculates a path metric by accumulating the branch metrics every clock, and a path metric is minimized.
  • a path memory for selecting and storing the data series as the most probable candidate series.
  • the path memory stores a plurality of candidate sequences, and selects candidate sequences according to a selection signal from the path metric calculation circuit. In addition, the selected candidate series is output as a data series.
  • the provisional determination unit 30 receives the linear equalization signal from the linear equalization unit 44 and executes Viterbi decoding on the linear equalization signal to sequentially provisionally determine the linear equalization signal.
  • the provisional determination unit 30 is configured in the same manner as the Viterbi decoding unit 38.
  • the path memory stores a plurality of candidate series, and temporary determination is executed according to the partial response rule based on the selection signal from the path metric calculation circuit. More specifically, the provisional determination unit 30 provisionally determines the output level for a predetermined input bit when the partial response equalization is normally performed, and the provisional determination signal indicates the level temporarily determined for the input bit. Output as 306.
  • the temporary determination unit 30 and the Viterbi decoding unit 38 are configured to have different path memory lengths. For example, when the path memory length of the Viterbi decoding unit 38 is 64 bits, the path memory length of the temporary determination unit 30 is 24 bits or 32 bits.
  • the first delay unit 32 receives the linear equalization signal from the linear equalization unit 44.
  • the first delay unit 32 delays the linear equalization signal and then outputs the delayed linear equalization signal (hereinafter referred to as “linear equalization signal” or “delay signal”) to the first addition unit 40.
  • the first delay unit 32 executes the delay over a period corresponding to the processing delay in the provisional determination unit 30. That is, the timing of the provisional determination signal 306 output from the provisional determination unit 30 and the linear equalization signal from the linear equalization unit 44 are matched in the first addition unit 40.
  • the first delay unit 32 is configured by, for example, a latch circuit driven by a bit clock.
  • the first addition unit 40 receives the linear equalization signal and the temporary determination signal 306 from the first delay unit 32.
  • the first addition unit 40 generates a linear equalization error signal 300 based on the difference between the linear equalization error and the provisional determination signal 306. For example, the linear equalization error signal 300 is derived by subtracting the provisional determination signal 306 from the linear equalization error. The first addition unit 40 outputs the linear equalization error signal 300 to the linear equalization unit 44.
  • the second delay unit 34 inputs the nonlinear equalization signal from the nonlinear equalization unit 46.
  • the second delay unit 34 delays the nonlinear equalized signal and then outputs the delayed nonlinear equalized signal (hereinafter referred to as “nonlinear equalized signal” or “delayed signal”) to the second adder 42.
  • the second delay unit 34 executes the delay over a period corresponding to the difference between the processing delay in the provisional determination unit 30 and the processing delay in the nonlinear equalization unit 46.
  • the second adder 42 generates a non-linear equalization error signal 302 based on the difference between the non-linear equalization signal from the second delay unit 34 and the provisional determination signal 306.
  • the non-linear equalization error signal 302 is derived by subtracting the provisional determination signal 306 from the non-linear equalization signal sum.
  • the second addition unit 42 outputs the non-linear equalization error signal 302 to the non-linear equalization unit 46.
  • the non-linear equalization unit 46 derives a plurality of coefficients based on the non-linear equalization error signal 302. That is, the nonlinear equalizer 46 uses the temporary determination signal 306 as a teacher signal. Further, the non-linear equalization unit 46 calculates an integrated value that is a sum obtained by successively adding the square values of the non-linear equalization error signal 302, thereby converging a plurality of tap coefficients in the non-linear equalization unit 46. To monitor. That is, the non-linear equalization unit 46 determines the convergence of a plurality of tap coefficients when the non-linear equalization error signal 302 changes from a state where the integrated value is larger than a threshold value to a small state.
  • the nonlinear equalization unit 46 newly derives a plurality of tap coefficients.
  • FIG. 3 shows the configuration of the linear equalization unit 44.
  • the linear equalization unit 44 includes a multistage tap 50 and a linear processing unit 52.
  • the multistage tap 50 includes a first delay tap 54a, a second delay tap 54b, a third delay tap 54c, and an Nth delay tap 54n, which are collectively referred to as a delay tap 54.
  • the linear processing unit 52 includes a first multiplication unit 56a, a second multiplication unit 56b, a third multiplication unit 56c, an N + 1 multiplication unit 56n + 1, a tap coefficient derivation unit 58, and an integration unit 60, which are collectively referred to as a multiplication unit 56.
  • the multistage tap 50 is formed by serially connecting a plurality of delay taps 54. More specifically, the first delay tap 54a inputs the reproduction signal, and outputs the reproduction signal after delay. The second delay tap 54b receives the reproduction signal from the first delay tap 54a, and outputs the reproduction signal after delay. The third delay tap 54c to the Nth delay tap 54n perform the same processing. An input portion and an output portion to the delay tap 54 are output signals from the multistage tap 50. For example, when four delay taps 54 are arranged, there are five output signals. These output signals are output to the multiplication unit 56.
  • the multiplication unit 56 inputs the output signal from the delay tap 54 and also receives the tap coefficient from the tap coefficient deriving unit 58.
  • the tap coefficient is derived in association with each output signal.
  • the multiplier 56 multiplies the output signal and the tap coefficient.
  • the multiplication unit 56 outputs each multiplication result to the integration unit 60.
  • the accumulating unit 60 adds the multiplication results from the multiplying unit 56 one after another to obtain an accumulated value as an addition result.
  • the integrated value as the addition result corresponds to the above-described linear equalization signal.
  • the integrating unit 60 outputs a linear equalization signal.
  • the tap coefficient deriving unit 58 receives the linear equalization error signal 300.
  • the tap coefficient deriving unit 58 controls a plurality of tap coefficients using the linear equalization error signal 300 and the multiplication result of the multiplication unit 56 so that the reproduction signal matches the partial response characteristic.
  • the tap coefficient is derived by using an adaptive algorithm such as the LMS algorithm so that the linear equalization error signal 300 is controlled to be small. Since the LMS algorithm is a known technique, the description thereof is omitted here.
  • FIG. 4 shows the configuration of the nonlinear equalization unit 46.
  • the nonlinear equalization unit 46 includes a multistage tap 70 and a nonlinear processing unit 72.
  • the multistage tap 70 includes a first delay tap 74a, a second delay tap 74b, and an Nth delay tap 74n, which are collectively referred to as a delay tap 74.
  • the nonlinear processing unit 72 includes an eleventh multiplication unit 76aa, a twelfth multiplication unit 76ab, a first M multiplication unit 76am, a twenty-first multiplication unit 76ba, a twenty-second multiplication unit 76bb, a second M multiplication unit 76bm, (N + 1) 1 multiplier 76 (n + 1) a, (N + 1) 2 multiplier 76 (n + 1) b, (N + 1) M multiplier 76 (n + 1) m, and first integrator 78a collectively referred to as integrator 78.
  • a 1-multiplier 82a, a second multiplier 82b, an M-th multiplier 82m, an integrator 84, a function calculator 86, and a tap coefficient derivation unit 88 are included.
  • the non-linear equalization unit 46 is configured by a three-layer perceptron type neural network as shown in the figure.
  • the input layer corresponds to the multistage tap 70
  • the hidden layer corresponds to the function calculation unit 80
  • the output layer corresponds to the function calculation unit 86.
  • the multistage tap 70 is formed by serially connecting a plurality of delay taps 74. More specifically, the first delay tap 74a receives a linear equalization signal and outputs a linear equalization signal after delaying.
  • the second delay tap 74b receives the linear equalization signal from the first delay tap 74a, and outputs the linear equalization signal after the delay.
  • the Nth delay tap 74n also performs the same process.
  • An input portion and an output portion to the delay tap 74 are output signals from the multistage tap 70. These output signals are output to the multiplier 76.
  • the multiplication unit 76 multiplies the output signal from the multistage tap 70 and the tap coefficient from the tap coefficient deriving unit 88. More specifically, the IJ multiplication unit 76ij multiplies the i-th output signal S (i) from the top of the multistage tap 70 by the tap coefficient W1 (i, j), thereby obtaining a multiplication result U (i , J).
  • the accumulating unit 78 performs accumulating by sequentially adding the multiplication results in the multiplying unit 76. More specifically, the J-th integrating unit 78j adds the multiplication results U (1, j), U (2, j), U (3, j),..., U (n + 1, j). To generate an integration result V (j).
  • the function calculation unit 80 calculates a sigmoid function on the integration result V (j) in the integration unit 78.
  • the integration result V (j) is input to x in Expression 1.
  • the calculation result in the J-th function calculation unit 80j is denoted as X (j), and the calculation result corresponds to the output from the hidden layer.
  • the multiplication unit 82 multiplies the calculation result in the function calculation unit 80 and the tap coefficient from the tap coefficient derivation unit 88. Specifically, the J-th multiplication unit 82j generates a multiplication result Y (j) by multiplying the calculation result X (j) in the J-th function calculation unit 80j by the tap coefficient W2 (j). .
  • the accumulating unit 84 performs an accumulation by sequentially adding the multiplication results in the multiplying unit 82.
  • the multiplication results in all the multiplication units 82 are integrated, and an integration result Z is generated.
  • the function calculation unit 86 calculates a sigmoid function on the integration result in the integration unit 84.
  • the integration result Z is input to x in Equation 1.
  • the calculation result of the function calculation unit 86 corresponds to the output from the output layer, and corresponds to the above-described nonlinear equalization signal.
  • the tap coefficient deriving unit 88 derives tap coefficients W1 (i, j) and W2 (j) used in the multiplying unit 76 and the multiplying unit 82. A random value or a value close to that after convergence is set as the initial value of W1 (i, j) and W2 (j). Further, the tap coefficient deriving unit 88 updates W1 (i, j) and W2 (j) by the LMS algorithm similarly to the tap coefficient deriving unit 58 of FIG. Here, learning of W1 (i, j) and W2 (j) is performed by back propagation.
  • A corresponds to the linear equalization signal
  • D corresponds to the provisional determination signal 306. That is, AD corresponds to the error signal 302 for nonlinear equalization.
  • the tap coefficient deriving unit 88 controls W1 (i, j) and W2 (j) so that E is minimized.
  • the result of back propagation at the output layer is shown as follows.
  • ( ⁇ E) / ( ⁇ Y (j)) f ′ (Y (j)) ⁇ 2 (AD) (Formula 3)
  • the tap coefficient deriving unit 88 updates the tap coefficient W2 (j) as follows.
  • W2 (j) W2 (j) old ⁇ ⁇ ( ⁇ E) / ( ⁇ W2 (j)) (Formula 4)
  • W2 (j) old indicates the tap coefficient W2 (j) at the previous timing.
  • the tap coefficient deriving unit 88 updates the tap coefficient W1 (i, j) as follows.
  • W1 (i, j) W1 (i, j) old ⁇ ⁇ ( ⁇ E) / ( ⁇ W1 (i, j)) (Formula 6)
  • W1 (i, j) old indicates the tap coefficient W1 (i, j) at the previous timing.
  • FIG. 5 shows a configuration of the provisional determination unit 30.
  • the provisional determination unit 30 includes a branch metric calculation unit 90, a path memory unit 92, and a specifying unit 96. Further, the selection signal SEL is included as a signal.
  • the branch metric calculation unit 90 performs branch metric calculation and path metric calculation based on a linear equalization signal from a linear equalization unit 44 (not shown). Therefore, the branch metric calculation unit 90 includes the aforementioned branch metric calculation circuit and path metric calculation circuit.
  • the partial response method is applied in the present embodiment, but before the configuration of the provisional determination unit 30 is described, here, state transition in the partial response method will be described.
  • FIG. 6 shows a state transition when the temporary determination unit 30 corresponds to a partial response (1, 2, 2, 2, 1).
  • the partial response (1, 2, 2, 2, 1) the amplitude falls within a range of ⁇ 4.
  • 10 states from S0 to S9 are defined according to the values included in the combination.
  • the state transitions as shown in the figure according to the next input bit value. For example, when a bit value “1” is input to the state S0, a transition to the state S1 is made.
  • a value such as “x / y” is shown in the arrows connecting the states, where x indicates an input bit value and y indicates a new bit value added to the original state.
  • the temporary decision value for 5 bits is shown.
  • FIG. 7 shows a state transition when the provisional determination unit 30 corresponds to a partial response (1, 2, 2, 2, 1).
  • FIG. 7 shows a state at two consecutive timings, and each state is the same as FIG.
  • FIG. 8 shows the configuration of the branch metric calculation unit 90.
  • the branch metric calculation unit 90 includes a first addition unit 110a, a second addition unit 110b, a third addition unit 110c, a fourth addition unit 110d, a fifth addition unit 110e, a sixth addition unit 110f, which are collectively referred to as an addition unit 110.
  • SEL0 a 0th selection signal SEL0, a first selection signal SEL1, a second selection signal SEL2, a seventh selection signal SEL7, an eighth selection signal SEL8, and a ninth selection signal SEL9, which are collectively referred to as a selection signal SEL.
  • the addition unit 110 subtracts a predetermined target value from the linear equalization signal.
  • the square circuit 112 calculates the square value of the subtraction result in the addition unit 110.
  • the ACS circuit 114 performs a metric operation on the square from the square circuit 112 by addition, comparison, and selection. Further, the ACS circuit 114 outputs a 0th selection signal SEL0, a first selection signal SEL1, a second selection signal SEL2, a seventh selection signal SEL7, an eighth selection signal SEL8, and a ninth selection signal SEL9 as a result of the metric calculation. To do. There is also a square value that is not input to the ACS circuit 114 due to the partial response characteristic. The adder 116 adds such a square value. Returning to FIG.
  • the path memory unit 92 receives the selection signal SEL from the branch metric calculation unit 90 and stores a path corresponding to the selection signal SEL.
  • FIG. 9 shows the configuration of the path memory unit 92.
  • the path memory unit 92 includes an eleventh memory 120aa, a twelfth memory 120ab, a thirteenth memory 120ac, a fourteenth memory 120ad, a fifteenth memory 120ae, a sixteenth memory 120af, a seventeenth memory 120ag, and an eighteenth memory.
  • one path is stored in the L + 1 memory 120, and 10 types of paths are stored so as to correspond to each of the 10 types of states shown in FIGS.
  • the selection unit 122 selects one of the paths according to the selection signal SEL.
  • the selected path corresponds to the survival path.
  • the majority decision unit 124 inputs bit values stored in the (L + 1) th memory 120 (l + 1) a to the (L + 1) th memory 120 (l + 1) j, respectively, and executes the majority decision.
  • the majority decision unit 124 outputs the selection result.
  • the specifying unit 96 inputs the selection value from the majority decision unit 124 (not shown) and holds the selection value in a latch.
  • the specifying unit 96 selects one combination from selection values corresponding to five timings including past selection values.
  • the combination is updated by removing the oldest selection value from the combination.
  • FIG. 10 shows the data structure of the table stored in the specifying unit 96.
  • memory value column 200 b (k) column 202, b (k-1) column 204, b (k-2) column 206, b (k-3) column 208, b (k-4) column.
  • 210 and a provisional determination output column 212 are included.
  • b (k) corresponds to the most recently input selection value
  • b (k ⁇ 1) corresponds to the selection value input at the previous timing
  • b (k ⁇ 4) This corresponds to the selection value input at the previous four timings. As described above, these are held by the latch.
  • the b (k) column 202 to the b (k-4) column 210 combinations of values that the selection values held in the latch can take are shown.
  • the memory value column 200 memory values corresponding to possible values are shown, and in the temporary determination output column 212, temporary determination values corresponding to possible values are shown. For example, if the content of the path memory is “00000”, the temporary determination value “ ⁇ 4” is associated, and if “00001”, the temporary determination value “ ⁇ 3” is associated.
  • the specifying unit 96 specifies a provisional determination value corresponding to the combination while referring to the table shown in FIG.
  • the specifying unit 96 outputs the temporary determination value as the temporary determination signal 306.
  • FIG. 11 is a flowchart showing a procedure for deriving coefficients in the nonlinear equalization unit 46.
  • the nonlinear equalization unit 46 continues to derive the magnitude of the nonlinear equalization error signal 302 even after the magnitude of the nonlinear equalization error signal 302 has converged.
  • the nonlinear equalization unit 46 newly derives a tap coefficient (S42). If the size does not become larger than the threshold value (N in S40), the process is terminated.
  • FIG. 12A shows a histogram of signals equalized by a conventional linear waveform equalizer.
  • the bit error rate at this time is 1.1 ⁇ 10 ⁇ 2 .
  • FIG. 12B shows a histogram of signals equalized by the playback apparatus 100.
  • the target value at this time is 9 values of partial response (1, 2, 2, 2, 1) as described above. Further, the bit error rate at this time is 1.5 ⁇ 10 ⁇ 4 . It is estimated that the characteristic deterioration in the conventional linear waveform equalizer is due to the fact that the waveform does not converge to the Viterbi target value because the waveform includes a nonlinear component.
  • the result of provisional determination of the linear equalization signal is used as the teacher signal, so that the provisional determination signal can be used as the teacher signal instead of the training signal.
  • the temporary determination signal is used as the teacher signal instead of the training signal
  • a coefficient for nonlinear equalization can be derived without using the training signal.
  • the coefficient for nonlinear equalization is derived without using the training signal, nonlinear equalization can be performed without using the training signal.
  • a coefficient for nonlinear equalization is derived without using a training signal, nonlinear distortion of a reproduction signal caused by improvement in recording density, recording power fluctuation, or the like can be reduced without a training signal.
  • provisional judgment is executed according to the partial response rule, so it can support partial response processing.
  • the delay is executed over a period corresponding to the difference between the processing delay in the provisional determination unit and the processing delay in the nonlinear equalization unit, the timing of the nonlinear equalization signal and the provisional determination signal can be matched.
  • the timing of the nonlinear equalization signal and the provisional determination signal are matched, it is possible to improve the estimation accuracy of tap coefficients for nonlinear equalization. Further, since a plurality of coefficients are newly derived when the divergence of the non-linear equalization error signal is detected, deterioration of the equalization characteristics can be suppressed.
  • the second embodiment of the present invention reproduces a signal recorded on a recording medium such as an optical disk, equalizes the reproduced signal (hereinafter referred to as “reproduced signal”) using a partial response method, and equalizes the signal.
  • the present invention relates to a playback apparatus for decoding (hereinafter referred to as “equalized signal”).
  • the recording capacity of the optical disk increases, the influence of nonlinear distortion that cannot be removed by the linear waveform equalizer increases.
  • a neural network as a nonlinear equalizer is effective, but it is necessary to learn and converge with a training signal. Therefore, in order to reduce the non-linear distortion of the reproduction signal without the training signal, the reproduction apparatus according to the present embodiment executes the following processing.
  • the playback device has a linear waveform equalizer and a nonlinear waveform equalizer arranged in parallel, and inputs a playback signal to both.
  • the reproducing apparatus includes an equalized signal from a linear waveform equalizer (hereinafter referred to as “linear equalized signal”) and an equalized signal from a nonlinear waveform equalizer (hereinafter referred to as “nonlinear equalized signal”).
  • the synthesized signal hereinafter referred to as “addition signal” is input to the Viterbi decoder.
  • the Viterbi decoder functions as a provisional determination unit of the present invention that sequentially provisionally determines the addition signal.
  • a signal provisionally determined by the Viterbi decoder (hereinafter referred to as “provisional determination signal”) is input as a teacher signal to the linear waveform equalizer and the nonlinear waveform equalizer.
  • the linear waveform equalizer and the nonlinear waveform equalizer perform equalization processing by deriving tap coefficients based on the teacher signal. For example, since the neural network is used for the nonlinear equalizer, the neural network is learned without using the training signal.
  • the reproducing apparatus further performs the following operation. Execute the process.
  • the nonlinear waveform equalizer requires a longer period for tap coefficient convergence than the linear waveform equalizer.
  • the playback device monitors the learning convergence state of the neural network, and if convergence is not confirmed, the linear equalization signal is used instead of the addition signal. Is output to the Viterbi decoder.
  • a linear equalization signal, a non-linear equalization signal, and a temporary determination signal are used for derivation of tap coefficients in the linear waveform equalizer and the non-linear waveform equalizer.
  • the output timing is different from the provisional determination signal. Therefore, in order to match these timings, the playback device delays the linear equalized signal and the nonlinear equalized signal. Examples of the present invention will be described below.
  • FIG. 13 shows the configuration of the processing unit 24.
  • the processing unit 24 includes an equalization processing unit 1030, a first delay unit 1032, a second delay unit 1034, an addition unit 1036, a Viterbi decoding unit 1038, an equalization error generation unit 1040, and a determination unit 1042.
  • the equalization processing unit 1030 includes a linear equalization unit 1044 and a non-linear equalization unit 1046. Also, the equalization error generation unit 1040 and the non-linear equalization unit 1046 are grouped with the adaptive non-linear equalization unit 1048. Further, the signal includes a linear equalization error signal 1300, a non-linear equalization error signal 1302, and a provisional determination signal 1306.
  • the Viterbi decoding unit 1038 functions as a temporary determination unit of the present invention that sequentially determines the signals added by the addition unit 1036.
  • a temporary determination unit may be provided separately from the Viterbi decoding unit 1038. It doesn't matter.
  • the reproduction signal sampled for each bit clock in the A / D conversion unit 22 in FIG. 1 is sequentially input to the linear equalization unit 1044 and the non-linear equalization unit 1046.
  • the linear equalizer 1044 sequentially performs linear equalization on the input reproduction signal.
  • the linear equalization unit 1044 is configured by a transversal filter, delays the reproduction signal by a multistage tap, multiplies the output from the multistage tap and a plurality of tap coefficients, and adds the multiplication results.
  • the addition result corresponds to the linear equalization signal described above.
  • the linear equalization unit 1044 receives a linear equalization error signal 1300 from an equalization error generation unit 1040 described later, and derives a plurality of tap coefficients based on the linear equalization error signal 1300.
  • an adaptive algorithm such as a LMS (Least Mean Square) algorithm is used to derive a plurality of tap coefficients.
  • the linear equalization unit 1044 outputs the linear equalization signal to the equalization error generation unit 1040 via
  • the non-linear equalization unit 1046 sequentially performs non-linear equalization on the reproduction signal in parallel with the linear equalization in the linear equalization unit 1044.
  • the nonlinear equalization unit 1046 is configured by a neural network.
  • the result of nonlinear equalization in the nonlinear equalization unit 1046 corresponds to the aforementioned nonlinear equalization signal.
  • the nonlinear equalization unit 1046 receives a nonlinear equalization error signal 1302 from an equalization error generation unit 1040, which will be described later, and based on the nonlinear equalization error signal 1302, a plurality of signals used in the neural network.
  • a tap coefficient is derived.
  • the non-linear equalization error signal 1302 is generated by the difference between the sum of the delay signal from the first delay unit 1032 and the delay signal from the second delay unit 1034 and the provisional determination signal. It can be said that the nonlinear equalization unit 1046 derives a plurality of coefficients using the temporary determination signal as a teacher signal. The nonlinear equalization unit 1046 outputs the nonlinear equalization signal to the equalization error generation unit 1040 via the second delay unit 1034 and directly to the addition unit 1036 without passing through the second delay unit 1034.
  • the first delay unit 1032 receives the linear equalization signal from the linear equalization unit 1044.
  • the first delay unit 1032 delays the linear equalization signal and then outputs the delayed linear equalization signal (hereinafter referred to as “linear equalization signal” or “delayed signal”) to the equalization error generation unit 1040.
  • the first delay unit 1032 executes a delay over a period corresponding to the processing delay for provisional determination in the Viterbi decoding unit 1038. That is, the first delay unit 1032 uses the provisional determination signal 1306 output from the Viterbi decoding unit 1038 after reaching the Viterbi decoding unit 1038 from the linear equalization unit 1044 and the linear equalization signal from the linear equalization unit 1044. Timing is adjusted.
  • the first delay unit 1032 is constituted by, for example, a latch circuit driven by a bit clock.
  • the second delay unit 1034 receives the nonlinear equalization signal from the nonlinear equalization unit 1046.
  • the second delay unit 1034 delays the nonlinear equalization signal, and then outputs the delayed nonlinear equalization signal (hereinafter referred to as “nonlinear equalization signal” or “delayed signal”) to the equalization error generation unit 1040.
  • the second delay unit 1034 executes a delay over a period corresponding to the processing delay for provisional determination in the Viterbi decoding unit 1038.
  • the addition unit 1036 receives the linear equalization signal from the linear equalization unit 1044 and the non-linear equalization signal from the non-linear equalization unit 1046.
  • the adder 1036 generates an addition signal by adding the linear equalization signal and the nonlinear equalization signal.
  • Adder 1036 outputs the addition signal to Viterbi decoder 1038.
  • the Viterbi decoding unit 1038 receives the addition signal from the addition unit 1036 and performs Viterbi decoding on the addition signal.
  • the Viterbi decoding unit 1038 includes a branch metric calculation circuit that calculates a branch metric from the addition signal, a path metric calculation circuit that calculates a path metric by accumulating the branch metrics for each clock, and a data sequence that minimizes the path metric And a path memory that selects and stores them as the most probable candidate series.
  • the path memory stores a plurality of candidate sequences, and selects candidate sequences according to a selection signal from the path metric calculation circuit. In addition, the selected candidate series is output as a data series.
  • the Viterbi decoding unit 1038 sequentially provisionally determines the addition signal by executing provisional determination on the data series stored in the path memory according to the partial response rule. That is, the Viterbi decoding unit 1038 performs a partial response provisional determination operation using a predetermined number of bits for one of the candidate sequences stored in the path memory. Specifically, the Viterbi decoding unit 1038 tentatively determines the output level for a predetermined input bit when the partial response equalization is normally performed, and the tentative determination signal indicates the level temporarily determined for the input bit. The result is output to the equalization error generation unit 1040 as 1306.
  • the provisional determination is not limited to the final result of the path memory, but may be made for a candidate series in the middle of the path memory. For example, if the path memory length is 64 bits, provisional determination may be made for one of the 24th and 32nd bit candidate series.
  • the equalization error generation unit 1040 receives the linear equalization signal from the first delay unit 1032, the nonlinear equalization signal from the second delay unit 1034, and the provisional determination signal 1306 from the Viterbi decoding unit 1038. As mentioned above, the timing of these signals is correct.
  • the equalization error generation unit 1040 generates a linear equalization error signal 1300 based on the difference between the sum of the linear equalization error and the nonlinear equalization signal and the provisional determination signal 1306. For example, after the sum of the linear equalization error and the non-linear equalization signal is calculated, the provisional determination signal 1306 is subtracted from the sum to derive the linear equalization error signal 1300.
  • the equalization error generation unit 1040 generates a non-linear equalization error signal 1302 based on the difference between the sum of the linear equalization error and the non-linear equalization signal and the provisional determination signal 1306. For example, after the sum of the linear equalization error and the nonlinear equalization signal is calculated, the provisional determination signal 1306 is subtracted from the sum to derive the nonlinear equalization error signal 1302.
  • the equalization error generation unit 1040 outputs the linear equalization error signal 1300 to the linear equalization unit 1044, and outputs the nonlinear equalization error signal 1302 to the nonlinear equalization unit 1046.
  • the nonlinear equalizer 1046 updates the tap coefficients of the neural network based on the error signal 1302 for nonlinear equalization, but the operation of the nonlinear equalizer 1046 is unstable until the tap coefficients converge. become. As a result, there is a high possibility that the data series output from the Viterbi decoding unit 1038 is erroneous. Therefore, it is not preferable to input the addition signal to the Viterbi decoding unit 1038 before convergence. In order to cope with this, the determination unit 1042 determines convergence of a plurality of tap coefficients in the nonlinear equalization unit 1046. More specifically, the determination unit 1042 adds up the square values of the nonlinear equalization error signal 1302 by adding them over a predetermined period.
  • the determination unit 1042 compares the integrated value with a threshold value, and determines that the integrated value is converged if the integrated value is equal to or less than the threshold value. On the other hand, if the integrated value is larger than the threshold value, the determination unit 1042 determines that the convergence has not occurred.
  • the determination unit 1042 outputs the determination result to the addition unit 1036 and the equalization error generation unit 1040. The determination result indicates whether or not it has converged.
  • the adding unit 1036 does not output the above addition signal to the Viterbi decoding unit 1038 until the determination unit 1042 determines convergence, that is, when it is determined that convergence has not occurred, the linear equalization unit 1044
  • the linear equalized signal is output to the Viterbi decoding unit 1038.
  • the adder 1036 outputs the added signal to the Viterbi decoder 1038 as described above after the convergence is determined by the determination unit 1042.
  • the linear equalized signal is output to the Viterbi decoding unit 1038 until the neural network converges, and the added signal is output to the Viterbi decoding unit 1038 after the neural network converges.
  • adverse effects due to the nonlinear equalized signal are reduced.
  • the equalization error generation unit 1040 does not use the nonlinear equalization signal until the convergence is determined by the determination unit 1042, and performs linear equalization based on the difference between the linear equalization signal and the provisional determination signal 1306. An error signal 1300 is generated. Also, the equalization error generation unit 1040 causes the linear equalization unit 1044 to derive a plurality of tap coefficients based on the linear equalization error signal 1300. On the other hand, the equalization error generation unit 1040 operates as described above after the determination unit 1042 determines convergence. Further, the equalization error generation unit 1040 performs non-linear equalization based on the difference between the sum of the linear equalization signal and the non-linear equalization signal and the provisional determination signal 1306 regardless of the determination result from the determination unit 1042.
  • the determination unit 1042 determines that the integrated value that is the sum of the square values of the nonlinear equalization error signal 1302 becomes larger than the threshold value again. It is determined that the tap coefficient is divergent. At this time, the determination unit 1042 causes the nonlinear equalization unit 1046 to newly derive a plurality of tap coefficients. Note that the non-linear equalization unit 1046 may only stop updating the plurality of tap coefficients when the value becomes equal to or less than the predetermined convergence value.
  • an addition signal of the linear equalization signal and the non-linear equalization signal may be output to the Viterbi decoding unit 1038 from the beginning, or the non-linear equalization error signal may be output.
  • the convergence of 1302 is equal to or less than the threshold value, it may be determined that the convergence has occurred after a predetermined time.
  • FIG. 14 shows the configuration of the linear equalization unit 1044.
  • the linear equalization unit 1044 includes a multistage tap 1050 and a linear processing unit 1052.
  • the multistage tap 1050 includes a first delay tap 1054a, a second delay tap 1054b, a third delay tap 1054c, and an Nth delay tap 1054n, which are collectively referred to as a delay tap 1054.
  • the linear processing unit 1052 includes a first multiplication unit 1056a, a second multiplication unit 1056b, a third multiplication unit 1056c, an (N + 1) th multiplication unit 1056n + 1, a tap coefficient derivation unit 1058, and an accumulation unit 1060, which are collectively referred to as a multiplication unit 1056.
  • the multi-stage tap 1050 is formed by connecting a plurality of delay taps 1054 serially. More specifically, the first delay tap 1054a inputs the reproduction signal, and outputs the reproduction signal after delay. The second delay tap 1054b receives the reproduction signal from the first delay tap 1054a, and outputs the reproduction signal after delay. The third delay tap 1054c to the Nth delay tap 1054n perform the same processing. An input portion and an output portion to the delay tap 1054 are output signals from the multistage tap 1050. For example, when four delay taps 1054 are arranged, there are five output signals. These output signals are output to the multiplication unit 1056.
  • Multiplier 1056 receives the output signal from delay tap 1054 and also receives the tap coefficient from tap coefficient deriving section 1058. Here, the tap coefficient is derived in association with each output signal. Multiplier 1056 multiplies the output signal and the tap coefficient. Multiplication unit 1056 outputs each multiplication result to integration unit 1060. The accumulating unit 1060 adds the multiplication results from the multiplying unit 1056 one after another to obtain an accumulated value as an addition result. The integrated value as the addition result corresponds to the above-described linear equalization signal. The integrating unit 1060 outputs a linear equalization signal. The tap coefficient deriving unit 1058 receives the linear equalization error signal 1300.
  • the tap coefficient deriving unit 1058 controls the plurality of tap coefficients using the linear equalization error signal 1300 and the multiplication result of the multiplication unit 1056 so that the reproduction signal matches the partial response characteristic.
  • the tap coefficient is derived by using an adaptive algorithm such as the LMS algorithm so that the linear equalization error signal 1300 is controlled to be small. Since the LMS algorithm is a known technique, the description thereof is omitted here.
  • FIG. 15 shows the configuration of the nonlinear equalization unit 1046.
  • the non-linear equalization unit 1046 includes a multistage tap 1070 and a non-linear processing unit 1072.
  • the multistage tap 1070 includes a first delay tap 1074a, a second delay tap 1074b, and an Nth delay tap 1074n, which are collectively referred to as a delay tap 1074.
  • the nonlinear processing unit 1072 includes an eleventh multiplication unit 1076aa, a twelfth multiplication unit 1076ab, a first M multiplication unit 1076am, a twenty-first multiplication unit 1076ba, a twenty-second multiplication unit 1076bb, a second M multiplication unit 1076bm, (N + 1) 1 multiplier 1076 (n + 1) a, (N + 1) 2 multiplier 1076 (n + 1) b, (N + 1) M multiplier 1076 (n + 1) m, and first integrator 1078a collectively referred to as integrator 1078 , A second summation unit 1078b, an Mth summation unit 1078m, a first function computation unit 1080a collectively referred to as a function computation unit 1080, a second function computation unit 1080b, an Mth function computation unit 1080m, and a multiplication unit 1082. 1 multiplier 1082a, 2nd multiplier 1082b, Mth multiplier 1082m, integrator 1084, function calculator
  • the non-linear equalization unit 1046 is configured by a three-layer perceptron type neural network as illustrated.
  • the input layer corresponds to the multistage tap 1070
  • the hidden layer corresponds to the function calculation unit 1080
  • the output layer corresponds to the function calculation unit 1086.
  • the multistage tap 1070 is formed by serially connecting a plurality of delay taps 1074. More specifically, the first delay tap 1074a inputs the reproduction signal, and outputs the reproduction signal after delay.
  • the second delay tap 1074b receives the reproduction signal from the first delay tap 1074a, and outputs the reproduction signal after delay.
  • the Nth delay tap 1074n performs the same process.
  • An input part and an output part to the delay tap 1074 are output signals from the multistage tap 1070. These output signals are output to the multiplier 1076.
  • Multiplier 1076 multiplies the output signal from multistage tap 1070 and the tap coefficient from tap coefficient deriving section 1088. Specifically, the IJ multiplication unit 1076ij multiplies the i-th output signal S (i) from the top of the multistage tap 1070 by the tap coefficient W1 (i, j), thereby obtaining a multiplication result U (i , J).
  • the accumulating unit 1078 performs an accumulation by sequentially adding the multiplication results in the multiplying unit 1076. More specifically, the J-th integrating unit 1078j adds the multiplication results U (1, j), U (2, j), U (3, j),..., U (n + 1, j). By integrating, an integration result V (j) is generated.
  • the function calculation unit 1080 calculates a sigmoid function on the integration result V (j) in the integration unit 1078.
  • the integration result V (j) is input to x in Expression 7.
  • the calculation result in the J-th function calculation unit 1080j is denoted as X (j), and the calculation result corresponds to the output from the hidden layer.
  • the multiplication unit 1082 multiplies the calculation result in the function calculation unit 1080 and the tap coefficient from the tap coefficient derivation unit 1088. Specifically, the J-th multiplication unit 1082j generates a multiplication result Y (j) by multiplying the calculation result X (j) in the J-th function calculation unit 1080j by the tap coefficient W2 (j). .
  • the accumulating unit 1084 performs accumulation in which the multiplication results in the multiplying unit 1082 are added one after another. Here, the multiplication results in all the multiplication units 1082 are added and integrated, and an integration result Z is generated.
  • the function calculation unit 1086 calculates a sigmoid function on the integration result in the integration unit 1084. Here, the integration result Z is input to x in Expression 7.
  • the calculation result of the function calculation unit 1086 corresponds to the output from the output layer, and corresponds to the above-described nonlinear equalization signal.
  • A corresponds to the sum of the linear equalization signal and the nonlinear equalization signal
  • D corresponds to the provisional determination signal 1306. That is, AD corresponds to the error signal 1302 for nonlinear equalization.
  • the tap coefficient deriving unit 1088 controls W1 (i, j) and W2 (j) so that E is minimized.
  • the result of back propagation at the output layer is shown as follows.
  • ( ⁇ E) / ( ⁇ Y (j)) f ′ (Y (j)) ⁇ 2 (AD) (Equation 9)
  • the tap coefficient deriving unit 1088 updates the tap coefficient W2 (j) as follows.
  • W2 (j) W2 (j) old ⁇ ⁇ ( ⁇ E) / ( ⁇ W2 (j)) (Equation 10)
  • W2 (j) old indicates the tap coefficient W2 (j) at the previous timing.
  • the tap coefficient deriving unit 1088 updates the tap coefficient W1 (i, j) as follows.
  • W1 (i, j) W1 (i, j) old ⁇ ⁇ ( ⁇ E) / ( ⁇ W1 (i, j)) (Equation 12)
  • W1 (i, j) old indicates the tap coefficient W1 (i, j) at the previous timing.
  • FIG. 16 shows the configuration of the Viterbi decoding unit 1038.
  • the Viterbi decoding unit 1038 includes a branch metric calculation unit 1090, a path memory unit 1092, a majority decision unit 1094, and a specification unit 1096.
  • the signal includes a selection signal SEL and a bit signal 1304.
  • the branch metric calculation unit 1090 executes branch metric calculation and path metric calculation based on a linear equalized signal or an addition signal (hereinafter collectively referred to as “addition signal”) from an addition unit 1036 (not shown). Therefore, the branch metric calculation unit 1090 includes the aforementioned branch metric calculation circuit and path metric calculation circuit.
  • the partial response method is applied in the present embodiment, but before the configuration of the Viterbi decoding unit 1038 is described, state transition in the partial response method will be described here.
  • FIG. 17 shows a state transition when the Viterbi decoding unit 1038 corresponds to a partial response (1, 2, 2, 2, 1).
  • the partial response (1, 2, 2, 2, 1) the amplitude falls within a range of ⁇ 4.
  • 10 states from S0 to S9 are defined according to the values included in the combination.
  • the state transitions as shown in the figure according to the next input bit value. For example, when a bit value “1” is input to the state S0, a transition to the state S1 is made.
  • a value such as “x / y” is shown in the arrows connecting the states, where x indicates an input bit value and y indicates a new bit value added to the original state.
  • the temporary decision value for 5 bits is shown.
  • FIG. 18 shows a state transition when the Viterbi decoding unit 1038 corresponds to a partial response (1, 2, 2, 2, 1).
  • FIG. 18 shows a state at two consecutive timings, and each state is the same as FIG.
  • FIG. 19 shows the configuration of the branch metric calculation unit 1090.
  • the branch metric calculation unit 1090 includes a first addition unit 1110a, a second addition unit 1110b, a third addition unit 1110c, a fourth addition unit 1110d, a fifth addition unit 1110e, a sixth addition unit 1110f, which are collectively referred to as an addition unit 1110.
  • SEL0 a 0th selection signal SEL0, a first selection signal SEL1, a second selection signal SEL2, a seventh selection signal SEL7, an eighth selection signal SEL8, and a ninth selection signal SEL9, which are collectively referred to as a selection signal SEL.
  • the addition unit 1110 subtracts a predetermined target value from the addition signal.
  • the square circuit 1112 calculates the square value of the subtraction result in the adder 1110.
  • the ACS circuit 1114 performs a metric operation by addition, comparison, and selection on the square from the square circuit 1112. Further, the ACS circuit 1114 outputs the 0th selection signal SEL0, the first selection signal SEL1, the second selection signal SEL2, the seventh selection signal SEL7, the eighth selection signal SEL8, and the ninth selection signal SEL9 as the result of the metric calculation. To do. There is also a square value that is not input to the ACS circuit 1114 due to the partial response characteristic. An addition unit 1116 adds the square value.
  • FIG. 20 shows the configuration of the path memory unit 1092.
  • the path memory unit 1092 receives the selection signal SEL from the branch metric calculation unit 1090 and stores a path corresponding to the selection signal SEL.
  • FIG. 20 shows the configuration of the path memory unit 1092.
  • the path memory unit 1092 includes an eleventh memory 1120aa, a twelfth memory 1120ab, a thirteenth memory 1120ac, a fourteenth memory 1120ad, a fifteenth memory 1120ae, a sixteenth memory 1120af, a seventeenth memory 1120ag, and an eighteenth memory.
  • one path is stored in the L + 1 memory 1120, and ten types of paths are stored so as to correspond to each of the ten types shown in FIGS.
  • the selection unit 1122 selects one of the paths according to the selection signal SEL.
  • the selected path corresponds to the survival path.
  • the majority decision unit 1124 inputs bit values stored in the (L + 1) th memory 1120 (l + 1) a to the (L + 1) th memory 1120 (l + 1) j, and executes majority decision.
  • the bit value selected by the majority vote corresponds to the decoding result.
  • the majority decision unit 1124 outputs the decryption result. Note that the bit value stored in the memory 1120 during the pass is output as a bit signal 1304.
  • the bit signal 1304 includes 10 bit values corresponding to the same timing among the 10 types of paths.
  • the majority decision unit 1094 receives the bit signal 1304 and executes a majority decision on the 10 bit values included in the bit signal 1304.
  • the majority decision unit 1094 outputs the bit value selected by the majority decision (hereinafter referred to as “selected value”) to the identification unit 1096.
  • the specifying unit 1096 receives the selection value from the majority decision unit 1094 and holds the selection value in a latch. Here, the specifying unit 1096 selects one combination from selection values corresponding to five timings including past selection values. When a new selection value is input to the specifying unit 1096, the combination is updated by removing the oldest selection value from the combination.
  • FIG. 21 shows the data structure of the table stored in the specifying unit 1096.
  • memory value column 1200 b (k) column 1202, b (k-1) column 1204, b (k-2) column 1206, b (k-3) column 1208, b (k-4) column. 1210 and a provisional determination output column 1212 are included.
  • b (k) corresponds to the most recently input selection value
  • b (k ⁇ 1) corresponds to the selection value input at the previous timing
  • b (k ⁇ 4) This corresponds to the selection value input at the previous four timings. As described above, these are held by the latch.
  • combinations of values that the selection value held in the latch can take are shown.
  • the memory value column 1200 shows memory values corresponding to possible values
  • the temporary determination output column 1212 shows temporary determination values corresponding to possible values. For example, if the content of the path memory is “00000”, the temporary determination value “ ⁇ 4” is associated, and if “00001”, the temporary determination value “ ⁇ 3” is associated.
  • the specifying unit 1096 specifies the provisional determination value corresponding to the combination while referring to the table shown in FIG.
  • the identifying unit 1096 outputs the temporary determination value as the temporary determination signal 1306.
  • FIG. 22 is a flowchart showing the adding procedure in the adding unit 1036.
  • the determination unit 1042 receives the nonlinear equalization error signal 1302 and derives the magnitude of the nonlinear equalization error signal 1302. If the magnitude of the non-linear equalization error signal 1302 does not converge within the threshold (N in S1010), the determination unit 1042 causes the addition unit 1036 to output a linear equalization signal (S1012). On the other hand, if the magnitude of the nonlinear equalization error signal 1302 has converged within the threshold value (Y in S1010), the determination unit 1042 causes the addition unit 1036 to output an addition signal (S1014).
  • FIG. 23 is a flowchart showing a generation procedure in the equalization error generation unit 1040.
  • the determination unit 1042 receives the nonlinear equalization error signal 1302 and derives the magnitude of the nonlinear equalization error signal 1302. If the magnitude of the non-linear equalization error signal 1302 does not converge within the threshold value (N in S20), the equalization error generation unit 1040 determines the difference between the delay signal from the first delay unit 1032 and the provisional determination signal 1306. Is output as an error signal 1300 for linear equalization to the linear equalization unit 1044 (S22).
  • the equalization error generation unit 1040 uses the difference between the sum of the delay signal from the first delay unit 1032 and the delay signal from the second delay unit 1034 and the provisional determination signal 1306 as a non-linear equalization error signal 1302. It outputs to the nonlinear equalization part 1046 (S24).
  • the equalization error generation unit 1040 and the delay signal from the first delay unit 1032 and the second delay unit 1034 The difference between the sum of the delay signal and the provisional determination signal 1306 is output to the linear equalization unit 1044 as the linear equalization error signal 1300 (S26).
  • the equalization error generation unit 1040 uses the difference between the sum of the delay signal from the first delay unit 1032 and the delay signal from the second delay unit 1034 and the provisional determination signal 1306 as a non-linear equalization error signal 1302. It outputs to the nonlinear equalization part 1046 (S28).
  • FIG. 24 is a flowchart showing a coefficient derivation procedure in the nonlinear equalization unit 1046.
  • the determination unit 1042 continuously derives the magnitude of the nonlinear equalization error signal 1302 even after the magnitude of the nonlinear equalization error signal 1302 has converged.
  • the linear equalization unit 1044 causes the nonlinear equalization unit 1046 to derive a new tap coefficient (S42). If the size does not become larger than the threshold value (N in S40), the process is terminated.
  • linear equalization and nonlinear equalization are performed in parallel, and the linear equalization signal and the nonlinear equalization signal are added, and the result of provisional determination of the addition signal is used as a teacher signal.
  • the temporary determination signal can be used as the teacher signal instead of the training signal.
  • a coefficient for nonlinear equalization can be derived without using the training signal.
  • nonlinear equalization can be performed without using the training signal.
  • a coefficient for nonlinear equalization is derived without using a training signal, nonlinear distortion of a reproduction signal caused by improvement in recording density, recording power fluctuation, or the like can be reduced without a training signal.
  • provisional judgment is executed according to the partial response rule, so it can support partial response processing.
  • the nonlinear equalization signal is not output to the Viterbi decoding unit until the nonlinear equalization coefficient is converged, deterioration of the accuracy of the provisional determination signal can be suppressed.
  • the deterioration of the accuracy of the provisional determination signal until the nonlinear equalization coefficient is converged is suppressed, the deterioration of the accuracy of the equalization processing can be suppressed even at the initial stage of the processing.
  • the nonlinear equalization signal is not used to derive the coefficient for linear equalization until the nonlinear equalization coefficient is converged, deterioration of the derivation accuracy of the coefficient can be suppressed.
  • deterioration in the derivation accuracy of the coefficient for linear equalization until the nonlinear equalization coefficient is converged is suppressed, deterioration in the accuracy of the equalization process can be suppressed even at the initial stage of the process. .
  • a plurality of coefficients are newly derived when the divergence of the non-linear equalization error signal is detected, deterioration of the equalization characteristics can be suppressed.
  • the third embodiment of the present invention relates to a reproducing apparatus in which a linear waveform equalizer and a nonlinear waveform equalizer are arranged in parallel. Both the linear waveform equalizer and the nonlinear waveform equalizer have multi-stage taps.
  • the multistage tap is shared by the linear waveform equalizer and the nonlinear waveform equalizer.
  • the playback apparatus 100 and the processing unit 24 according to the third embodiment are the same type as those in FIGS. Here, the difference from the second embodiment will be mainly described.
  • FIG. 25 shows a configuration of the equalization processing unit 1030 according to the third embodiment of the present invention.
  • the equalization processing unit 1030 includes a linear processing unit 1052, a non-linear processing unit 1072, and a multistage tap 1130.
  • the multistage tap 1130 includes a first delay tap 1132a, a second delay tap 1132b, a third delay tap 1132c, and an Nth delay tap 1132n, which are collectively referred to as a delay tap 1132.
  • the multistage tap 1130 is configured in the same manner as the multistage tap 1050 included in the linear equalization unit 1044 in FIG. 14 and the multistage tap 1070 included in the nonlinear equalization unit 1046 in FIG. 15, and a plurality of delay taps 1132 are serially connected. It is connected. That is, the multistage tap 1130 is shared by the linear equalization unit and the nonlinear equalization unit.
  • the output from the multistage tap 1130 is output to the linear processing unit 1052 and also to the nonlinear processing unit 1072.
  • the linear processing unit 1052 is the same as the configuration shown in FIG. 14, and the nonlinear processing unit 1072 is the same as the configuration shown in FIG. Here, these descriptions are omitted.
  • the multistage tap is shared in linear equalization and nonlinear equalization, an increase in circuit scale can be suppressed.
  • multistage taps performing the same processing are shared in linear equalization and nonlinear equalization, it is possible to suppress deterioration in processing accuracy with respect to linear equalization and nonlinear equalization.
  • the circuit scale can be reduced while suppressing the accuracy of equalization processing.
  • both the linear equalization unit 44 and the non-linear equalization unit 46 use a plurality of tap coefficients, and derive a plurality of tap coefficients adaptively.
  • the present invention is not limited to this.
  • the nonlinear equalizer 46 may adaptively derive a plurality of tap coefficients, and the linear equalizer 44 may use a fixed value as the plurality of tap coefficients. That is, the linear equalization unit 44 does not have to adaptively derive a plurality of tap coefficients.
  • the first addition unit 40 does not output the linear equalization error signal 300 to the linear equalization unit 44. According to this modification, since the coefficient for linear equalization is a fixed value, the processing can be simplified.
  • each of the linear equalization unit 1044 and the nonlinear equalization unit 1046 uses a plurality of tap coefficients and adaptively derives the plurality of tap coefficients.
  • the present invention is not limited to this.
  • the nonlinear equalizer 1046 may adaptively derive a plurality of tap coefficients, and the linear equalizer 1044 may use a fixed value as the plurality of tap coefficients. That is, the linear equalization unit 1044 may not derive a plurality of tap coefficients adaptively.
  • FIG. 26 shows the configuration of the processing unit 24 according to a modification of the present invention.
  • the processing unit 24 includes a delay unit 1140 as compared with FIG. Here, it demonstrates centering on the difference with FIG.
  • the adder 1036 adds the linear equalized signal from the linear equalizer 1044 and the nonlinear equalized signal from the nonlinear equalizer 1046.
  • the delay unit 1140 delays the signal output from the equalization error generation unit 1040.
  • the delay unit 1140 performs a delay over a period corresponding to the processing delay in the Viterbi decoding unit 1038.
  • the equalization error generation unit 1040 generates a non-linear equalization error signal 1302 based on the difference between the signal delayed in the delay unit 1140 and the provisional determination signal 1306 from the Viterbi decoding unit 1038, and performs non-linear equalization
  • the unit 1046 derives a plurality of coefficients based on the non-linear equalization error signal 1302. According to this modification, since the coefficient for linear equalization is a fixed value, the stability of the equalization process can be improved. In addition, since the coefficient for linear equalization is a fixed value, the processing can be simplified.
  • optical discs 10 optical discs, 12 optical disc drive units, 14 optical pickups, 16 preamplifier units, 18 AGC units, 20 PLL units, 22 A / D conversion units, 24 processing units, 26 control units, 30 temporary determination units, 32 first delay units, 34 second delay unit, 38 Viterbi decoding unit, 40 first addition unit, 42 second addition unit, 44 linear equalization unit, 46 non-linear equalization unit, 100 playback device.
  • the present invention it is possible to reduce the non-linear distortion of the reproduction signal caused by the improvement of the recording density, the recording power fluctuation or the like without the training signal.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Signal Processing For Digital Recording And Reproducing (AREA)

Abstract

A linear equalization unit (44) sequentially linearly equalizes signals to be processed. A temporary determination unit (30) sequentially temporarily determines the signals linearly equalized by the linear equalization unit (44). A nonlinear equalization unit (46) derives a plurality of coefficients using the temporarily determined signals as teacher signals, and, on the basis of the plurality of coefficients, sequentially nonlinearly equalizes the signals linearly equalized by the linear equalization unit (44).

Description

等化器および等化方法Equalizer and equalization method
 本発明は、等化器に関し、特に非線形ひずみを等化する等化器および等化方法に関する。 The present invention relates to an equalizer, and more particularly to an equalizer and an equalization method for equalizing nonlinear distortion.
 光ディスクから検出した再生信号を良好に復号するために、パーシャルレスポンス方式の線形波形等化回路とビタビ復号との併用が有効である。これは、例えば、DVDフォーラムより公開されているHD DVD(High Definition DVD)-ROM Part1(物理)規格書などに示されている。これに対応した光ディスク再生装置において、ディスク制御回路は、光ディスクを所定の回転速度で回転させ、光ピックアップは、光ディスクに記録された再生信号を読み取る。再生信号は、プリアンプで増幅された後、AGC回路等で所定の振幅に増幅される。 In order to satisfactorily decode the reproduced signal detected from the optical disc, it is effective to use a partial response type linear waveform equalization circuit and Viterbi decoding. This is shown, for example, in the HD DVD (High Definition DVD) -ROM Part 1 (physical) standard published by the DVD Forum. In an optical disk reproducing apparatus corresponding to this, the disk control circuit rotates the optical disk at a predetermined rotation speed, and the optical pickup reads a reproduction signal recorded on the optical disk. The reproduction signal is amplified by a preamplifier and then amplified to a predetermined amplitude by an AGC circuit or the like.
 さらに、再生信号は、A/D変換され、線形波形等化回路で波形等化された後、ビタビ復号で復号される。その結果、光ディスクに記録された画像データや音楽データが再現される。一方、光ディスクがさらに高密度化されると、再生波形の非線形歪が大きくなるので、線形波形等化回路だけでは不十分になる。そのため、非線形歪を低減するために、非線形波形等化回路が使用される。また、非線形波形等化回路を実現するために、ニューラルネットワークが使用される(例えば、特許文献1参照)。 Further, the reproduction signal is A / D converted, waveform equalized by a linear waveform equalization circuit, and then decoded by Viterbi decoding. As a result, image data and music data recorded on the optical disc are reproduced. On the other hand, when the density of the optical disk is further increased, the nonlinear distortion of the reproduced waveform becomes large, so that only the linear waveform equalization circuit is insufficient. Therefore, a nonlinear waveform equalization circuit is used to reduce nonlinear distortion. Further, a neural network is used to realize a nonlinear waveform equalization circuit (see, for example, Patent Document 1).
特開平10-106158号公報Japanese Patent Laid-Open No. 10-106158
 一般的に、ニューラルネットワークに所望の動作を実行させるためには、既知のトレーニング信号を使用して、学習動作を予め実行させることが必要である。例えば、光ディスクの所定箇所にトレーニング信号を記録しておき、トレーニング信号に対応した出力を教師信号として、ニューラルネットワークにおける係数が決定される。そのため、光ディスクにトレーニング信号が予め記録されるので、光ディスクの利用効率が低減する。また、学習動作の終了後は係数が固定されるので、光ディスクの面内において再生波形特性が変動することへの追従が困難になる。さらに、光ディスクにデータを記録した記録機によってパワー変動等があるが、それへの追従も困難になる。そのため、線形歪を適応的に低減するとともに、記録密度の向上や記録パワー変動等によって生じる再生信号の非線形歪をトレーニング信号なしで適応的に低減することが要求される。 Generally, in order to cause a neural network to execute a desired operation, it is necessary to execute a learning operation in advance using a known training signal. For example, a training signal is recorded at a predetermined location on the optical disc, and a coefficient in the neural network is determined using an output corresponding to the training signal as a teacher signal. Therefore, since the training signal is recorded in advance on the optical disc, the utilization efficiency of the optical disc is reduced. In addition, since the coefficient is fixed after the learning operation is finished, it is difficult to follow the fluctuation of the reproduction waveform characteristic in the plane of the optical disc. Furthermore, there are power fluctuations and the like depending on the recorder that records data on the optical disc, but it is also difficult to follow them. For this reason, it is required to adaptively reduce the linear distortion without using a training signal as well as adaptively reducing the non-linear distortion of the reproduction signal caused by the improvement of the recording density and the recording power fluctuation.
 本発明はこうした状況に鑑みてなされたものであり、その目的は、記録密度の向上や記録パワー変動等によって生じる再生信号の非線形歪をトレーニング信号なしで低減する技術を提供することにある。 The present invention has been made in view of such circumstances, and an object of the present invention is to provide a technique for reducing non-linear distortion of a reproduction signal caused by improvement in recording density, recording power fluctuation or the like without a training signal.
 上記課題を解決するために、本発明のある態様の等化器は、処理対象の信号を順次線形等化する線形等化部と、線形等化部において線形等化した信号を順次仮判定する仮判定部と、仮判定部において仮判定した信号を教師信号として複数の係数を導出するとともに、複数の係数をもとに、線形等化部において線形等化した信号を順次非線形等化する非線形等化部と、を備える。 In order to solve the above problems, an equalizer according to an aspect of the present invention sequentially linearly equalizes a signal to be processed and linearly equalizes a signal that is linearly equalized in the linear equalization unit. Nonlinear for deriving a plurality of coefficients using a provisional determination unit and a signal temporarily determined by the temporary determination unit as a teacher signal, and sequentially linearly equalizing a signal linearly equalized by a linear equalization unit based on the plurality of coefficients An equalization unit.
 この態様によると、線形等化信号を仮判定した結果を教師信号とするので、トレーニング信号を使用せずに、非線形等化のための係数を導出できる。 According to this aspect, since the result of provisional determination of the linear equalization signal is used as a teacher signal, a coefficient for nonlinear equalization can be derived without using a training signal.
 仮判定部は、パーシャルレスポンス規則にしたがって仮判定を実行してもよい。この場合、パーシャルレスポンス規則にしたがった仮判定が実行されるので、パーシャルレスポンス処理に対応できる。 The temporary determination unit may execute a temporary determination according to the partial response rule. In this case, the provisional determination according to the partial response rule is executed, so that the partial response process can be handled.
 非線形等化部において非線形等化した信号を遅延させる遅延部をさらに備えてもよい。遅延部は、仮判定部における処理遅延と非線形等化部における処理遅延との差異に応じた期間にわたって遅延を実行し、非線形等化部は、遅延部において遅延した信号と仮判定部において仮判定した信号との差異をもとに、複数の係数を導出してもよい。この場合、仮判定部における処理遅延と非線形等化部における処理遅延との差異に応じた期間にわたって遅延を実行するので、遅延した信号と仮判定した信号とのタイミングを合わせることができる。 A delay unit that delays the signal that is nonlinearly equalized in the nonlinear equalizer may be further provided. The delay unit executes delay over a period corresponding to the difference between the processing delay in the temporary determination unit and the processing delay in the nonlinear equalization unit, and the nonlinear equalization unit performs the temporary determination in the signal delayed in the delay unit and the temporary determination unit. A plurality of coefficients may be derived based on the difference from the signal. In this case, since the delay is executed over a period corresponding to the difference between the processing delay in the provisional determination unit and the processing delay in the nonlinear equalization unit, the timing of the delayed signal and the provisionally determined signal can be matched.
 非線形等化部は、差異がしきい値よりも大きくなった場合に、複数の係数を新たに導出してもよい。この場合、発散を検出した場合に複数の係数を新たに導出するので、等化特性の悪化を抑制できる。 The non-linear equalization unit may newly derive a plurality of coefficients when the difference becomes larger than the threshold value. In this case, since a plurality of coefficients are newly derived when divergence is detected, deterioration of equalization characteristics can be suppressed.
 本発明の別の態様は、等化方法である。この方法は、入力した信号を順次線形等化するステップと、線形等化した信号を順次仮判定するステップと、仮判定した信号を教師信号として複数の係数を導出するとともに、複数の係数をもとに、線形等化した信号を順次非線形等化するステップと、を備える。 Another aspect of the present invention is an equalization method. In this method, an input signal is sequentially linearly equalized, a linearly equalized signal is sequentially provisionally determined, a plurality of coefficients are derived using the provisionally determined signal as a teacher signal, and a plurality of coefficients are included. And sequentially performing nonlinear equalization on the linearly equalized signal.
 本発明のさらに別の態様は、等化器である。この等化器は、処理対象の信号を順次入力する入力部と、入力部において入力した信号を順次線形等化する線形等化部と、線形等化部における線形等化に並行して、入力部において入力した信号を順次非線形等化する適応非線形等化部と、適応非線形等化部において非線形等化した信号と、線形等化部において線形等化した信号とを加算する加算部と、加算部において加算した信号を順次仮判定する仮判定部とを備える。適応非線形等化部は、仮判定部において仮判定した信号を教師信号として複数の係数を導出するとともに、複数の係数をもとに、非線形等化を実行する。 Still another aspect of the present invention is an equalizer. The equalizer includes an input unit that sequentially inputs a signal to be processed, a linear equalization unit that sequentially linearly equalizes the signal input at the input unit, and an input that is performed in parallel with the linear equalization in the linear equalization unit. An adaptive nonlinear equalization unit that sequentially performs nonlinear equalization on the input signal, an addition unit that adds the nonlinear equalization signal in the adaptive nonlinear equalization unit, and the linear equalization signal in the linear equalization unit, and addition A provisional determination unit that sequentially provisionally determines signals added in the unit. The adaptive nonlinear equalization unit derives a plurality of coefficients using the signal provisionally determined by the provisional determination unit as a teacher signal, and performs nonlinear equalization based on the plurality of coefficients.
 この態様によると、線形等化と非線形等化とを並列に実行するとともに、両方からの等化信号を加算し、加算信号を仮判定した結果を教師信号とするので、トレーニング信号を使用せずに、非線形等化のための係数を導出できる。 According to this aspect, linear equalization and non-linear equalization are performed in parallel, the equalization signals from both are added, and the result of provisional determination of the addition signal is used as a teacher signal, so that no training signal is used. In addition, a coefficient for nonlinear equalization can be derived.
 線形等化部に含まれた多段タップと、適応非線形等化部における多段タップとが、共通化されていてもよい。この場合、多段タップが共通化されるので、回路規模の増加を抑制できる。 The multi-stage tap included in the linear equalization unit and the multi-stage tap in the adaptive nonlinear equalization unit may be shared. In this case, since multistage taps are shared, an increase in circuit scale can be suppressed.
 仮判定部は、パーシャルレスポンス規則にしたがって仮判定を実行してもよい。この場合、パーシャルレスポンス規則にしたがった仮判定が実行されるので、パーシャルレスポンス処理に対応できる。 The temporary determination unit may execute a temporary determination according to the partial response rule. In this case, the provisional determination according to the partial response rule is executed, so that the partial response process can be handled.
 適応非線形等化部における複数の係数の収束を判定する判定部をさらに備えてもよい。加算部は、判定部において収束が判定されるまでの間、線形等化部において線形等化した信号を仮判定部へ出力し、判定部において収束が判定された後、加算した信号を仮判定部へ出力してもよい。この場合、非線形等化の係数が収束されるまで、非線形等化した信号を仮判定部へ出力しないので、仮判定の精度の悪化を抑制できる。 A determination unit that determines convergence of a plurality of coefficients in the adaptive nonlinear equalization unit may be further provided. The adder outputs the signal linearly equalized by the linear equalizer to the temporary determiner until convergence is determined by the determiner, and after the convergence is determined by the determiner, the added signal is temporarily determined. You may output to a part. In this case, since the nonlinear equalized signal is not output to the provisional determination unit until the nonlinear equalization coefficient is converged, deterioration of the accuracy of the provisional determination can be suppressed.
 線形等化部において線形等化した信号を遅延させる第1遅延部と、適応非線形等化部において非線形等化した信号を遅延させる第2遅延部とをさらに備えてもよい。第1遅延部は、仮判定部における処理遅延に応じた期間にわたって遅延を実行し、第2遅延部は、仮判定部における処理遅延に応じた期間にわたって遅延を実行し、適応非線形等化部は、第1遅延部において遅延した信号と第2遅延部において遅延した信号との和と、仮判定部において仮判定した信号との差異をもとに、複数の係数を導出し、線形等化部は、複数の係数を使用して線形等化を実行しており、判定部において収束が判定されるまでの間、第1遅延部において遅延した信号と仮判定部において仮判定した信号との差異をもとに、複数の係数を導出し、判定部において収束が判定された後、第1遅延部において遅延した信号と第2遅延部において遅延した信号との和と、仮判定部において仮判定した信号との差異をもとに、複数の係数を導出してもよい。この場合、非線形等化の係数が収束されるまで、線形等化のための係数を導出するために、非線形等化した信号を使用しないので、当該係数の導出精度の悪化を抑制できる。 A linear delay unit may further include a first delay unit that delays the linearly equalized signal, and a second delay unit that delays the nonlinear equalized signal in the adaptive nonlinear equalization unit. The first delay unit executes a delay over a period according to the processing delay in the temporary determination unit, the second delay unit executes a delay over a period according to the processing delay in the temporary determination unit, and the adaptive nonlinear equalization unit The linear equalization unit derives a plurality of coefficients based on the difference between the sum of the signal delayed in the first delay unit and the signal delayed in the second delay unit and the signal provisionally determined in the temporary determination unit. Is performing linear equalization using a plurality of coefficients, and the difference between the signal delayed in the first delay unit and the signal temporarily determined in the temporary determination unit until convergence is determined in the determination unit Based on the above, a plurality of coefficients are derived, and after the convergence is determined by the determination unit, the sum of the signal delayed in the first delay unit and the signal delayed in the second delay unit and the temporary determination in the temporary determination unit Based on the difference from the It may be derived the number. In this case, since the nonlinear equalized signal is not used to derive the coefficient for linear equalization until the coefficient for nonlinear equalization is converged, deterioration of the derivation accuracy of the coefficient can be suppressed.
 線形等化部において線形等化した信号と適応非線形等化部において非線形等化した信号とを加算する加算部と、加算部から出力した信号を遅延させる遅延部とをさらに備えてもよい。遅延部は、仮判定部における処理遅延に応じた期間にわたって遅延を実行し、適応非線形等化部は、遅延部において遅延した信号と、仮判定部において仮判定した信号との差異をもとに、複数の係数を導出し、線形等化部は、複数の係数を使用して線形等化を実行しており、かつ複数の係数として固定値を使用してもよい。この場合、線形等化のための係数を固定値とするので、等化処理の安定性を向上できる。 An addition unit that adds a signal that is linearly equalized in the linear equalization unit and a signal that is nonlinearly equalized in the adaptive nonlinear equalization unit, and a delay unit that delays the signal output from the addition unit may be further provided. The delay unit performs a delay over a period corresponding to the processing delay in the provisional determination unit, and the adaptive nonlinear equalization unit is based on a difference between the signal delayed in the delay unit and the signal provisionally determined in the provisional determination unit. The plurality of coefficients may be derived, and the linear equalization unit may perform the linear equalization using the plurality of coefficients, and may use a fixed value as the plurality of coefficients. In this case, since the coefficient for linear equalization is a fixed value, the stability of the equalization process can be improved.
 適応非線形等化部は、複数の係数の発散を検出した場合に、複数の係数を新たに導出してもよい。この場合、発散を検出した場合に複数の係数を新たに導出するので、等化特性の悪化を抑制できる。 The adaptive nonlinear equalization unit may newly derive a plurality of coefficients when detecting the divergence of the plurality of coefficients. In this case, since a plurality of coefficients are newly derived when divergence is detected, deterioration of equalization characteristics can be suppressed.
 本発明のさらに別の態様は、等化方法である。この方法は、入力した信号を順次線形等化するステップと、線形等化に並行して、入力した信号を順次非線形等化するステップと、非線形等化した信号と線形等化した信号とを加算するステップと、加算した信号を順次仮判定するステップとを備える。非線形等化するステップは、仮判定した信号を教師信号として複数の係数を導出するとともに、複数の係数をもとに、非線形等化を実行する。 Still another aspect of the present invention is an equalization method. In this method, the step of linearly equalizing the input signal, the step of sequentially nonlinearly equalizing the input signal in parallel with the linear equalization, and the nonlinear equalized signal and the linearly equalized signal are added. And a step of tentatively determining the added signals sequentially. In the nonlinear equalization step, a plurality of coefficients are derived using the temporarily determined signal as a teacher signal, and nonlinear equalization is executed based on the plurality of coefficients.
 なお、以上の構成要素の任意の組合せ、本発明の表現を方法、装置、システム、記録媒体、コンピュータプログラムなどの間で変換したものもまた、本発明の態様として有効である。 It should be noted that an arbitrary combination of the above-described components and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, etc. are also effective as an aspect of the present invention.
 本発明によれば、記録密度の向上や記録パワー変動等によって生じる再生信号の非線形歪をトレーニング信号なしで低減できる。 According to the present invention, it is possible to reduce the non-linear distortion of the reproduction signal caused by the improvement of the recording density, the recording power fluctuation or the like without the training signal.
本発明の実施例1に係る再生装置の構成を示す図である。It is a figure which shows the structure of the reproducing | regenerating apparatus which concerns on Example 1 of this invention. 図1の処理部の構成を示す図である。It is a figure which shows the structure of the process part of FIG. 図2の線形等化部の構成を示す図である。It is a figure which shows the structure of the linear equalization part of FIG. 図2の非線形等化部の構成を示す図である。It is a figure which shows the structure of the nonlinear equalization part of FIG. 図2の仮判定部の構成を示す図である。It is a figure which shows the structure of the temporary determination part of FIG. 図5の仮判定部がパーシャルレスポンス(1,2,2,2,1)に対応する場合の状態遷移を示す図である。It is a figure which shows a state transition in case the temporary determination part of FIG. 5 respond | corresponds to a partial response (1, 2, 2, 2, 1). 図5の仮判定部がパーシャルレスポンス(1,2,2,2,1)に対応する場合の状態遷移を示す別の図である。It is another figure which shows a state transition in case the temporary determination part of FIG. 5 respond | corresponds to a partial response (1, 2, 2, 2, 1). 図5のブランチメトリック演算部の構成を示す図である。It is a figure which shows the structure of the branch metric calculating part of FIG. 図5のパスメモリ部の構成を示す図である。FIG. 6 is a diagram illustrating a configuration of a path memory unit in FIG. 5. 図5の特定部に記憶されたテーブルのデータ構造を示す図である。It is a figure which shows the data structure of the table memorize | stored in the specific part of FIG. 図2の非線形等化部における係数の導出手順を示すフローチャートである。It is a flowchart which shows the derivation | leading-out procedure of the coefficient in the nonlinear equalization part of FIG. 図12(a)-(b)は、従来および図1の再生装置による出力信号のヒストグラムを示す図である。12A and 12B are diagrams showing histograms of output signals obtained by the conventional reproduction apparatus and the reproduction apparatus of FIG. 本発明の実施例2に係る処理部の構成を示す図である。It is a figure which shows the structure of the process part which concerns on Example 2 of this invention. 図13の線形等化部の構成を示す図である。It is a figure which shows the structure of the linear equalization part of FIG. 図13の非線形等化部の構成を示す図である。It is a figure which shows the structure of the nonlinear equalization part of FIG. 図13のビタビ復号部の構成を示す図である。It is a figure which shows the structure of the Viterbi decoding part of FIG. 図16のビタビ復号部がパーシャルレスポンス(1,2,2,2,1)に対応する場合の状態遷移を示す図である。It is a figure which shows a state transition in case the Viterbi decoding part of FIG. 16 respond | corresponds to a partial response (1, 2, 2, 2, 1). 図16のビタビ復号部がパーシャルレスポンス(1,2,2,2,1)に対応する場合の状態遷移を示す別の図である。It is another figure which shows a state transition in case the Viterbi decoding part of FIG. 16 respond | corresponds to a partial response (1, 2, 2, 2, 1). 図16のブランチメトリック演算部の構成を示す図である。It is a figure which shows the structure of the branch metric calculating part of FIG. 図16のパスメモリ部の構成を示す図である。It is a figure which shows the structure of the path memory part of FIG. 図16の特定部に記憶されたテーブルのデータ構造を示す図である。It is a figure which shows the data structure of the table memorize | stored in the specific | specification part of FIG. 図13の加算部における加算手順を示すフローチャートである。It is a flowchart which shows the addition procedure in the addition part of FIG. 図13の等化誤差生成部における生成手順を示すフローチャートである。It is a flowchart which shows the production | generation procedure in the equalization error production | generation part of FIG. 図13の非線形等化部における係数の導出手順を示すフローチャートである。It is a flowchart which shows the derivation | leading-out procedure of the coefficient in the nonlinear equalization part of FIG. 本発明の実施例3に係る等化処理部の構成を示す図である。It is a figure which shows the structure of the equalization process part which concerns on Example 3 of this invention. 本発明の変形例に係る処理部の構成を示す図である。It is a figure which shows the structure of the process part which concerns on the modification of this invention.
(実施例1)
 本発明を具体的に説明する前に、まず概要を述べる。本発明の実施例1は、光ディスク等の記録媒体に記録されている信号を再生し、再生した信号(以下、「再生信号」という)をパーシャルレスポンス方式にて等化するととともに、等化した信号(以下、「等化信号」という)を復号する再生装置に関する。前述のごとく、光ディスクの記録容量が高まるにつれ、線形波形等化器では除去しきれない非線形歪の影響が大きくなっている。非線形歪を除去するためには、非線形等化器としてのニューラルネットが有効であるが、トレーニング信号によって学習、収束させる必要がある。そこで、再生信号の非線形歪をトレーニング信号なしで低減するために、本実施例に係る再生装置は次の処理を実行する。
Example 1
Before describing the present invention specifically, an outline will be given first. Embodiment 1 of the present invention reproduces a signal recorded on a recording medium such as an optical disc, equalizes a reproduced signal (hereinafter referred to as “reproduced signal”) by a partial response method, and equalizes the signal. The present invention relates to a playback apparatus for decoding (hereinafter referred to as “equalized signal”). As described above, as the recording capacity of the optical disk increases, the influence of nonlinear distortion that cannot be removed by the linear waveform equalizer increases. In order to remove the nonlinear distortion, a neural network as a nonlinear equalizer is effective, but it is necessary to learn and converge with a training signal. Therefore, in order to reduce the non-linear distortion of the reproduction signal without the training signal, the reproduction apparatus according to the present embodiment executes the following processing.
 再生装置は、非線形波形等化器の前段に線形波形等化器を直列に配置する。また、再生装置は、線形波形等化器からの等化信号(以下、「線形等化信号」という)を非線形波形等化器に入力した後、非線形波形等化器からの等化信号(以下、「非線形等化信号」という)をビタビ復号器に入力する。線形等化信号は、仮判定部にも入力され、仮判定部において仮判定される。仮判定された信号(以下、「仮判定信号」という)は、教師信号として、線形波形等化器と非線形波形等化器とに入力される。線形波形等化器と非線形波形等化器とは、教師信号をもとにタップ係数を導出して等化処理を実行する。 The playback device arranges a linear waveform equalizer in series before the nonlinear waveform equalizer. The reproduction apparatus inputs an equalized signal from a linear waveform equalizer (hereinafter referred to as “linear equalized signal”) to a nonlinear waveform equalizer, and then receives an equalized signal (hereinafter referred to as a nonlinear waveform equalizer). , Referred to as “nonlinear equalized signal”) to the Viterbi decoder. The linear equalization signal is also input to the temporary determination unit, and is temporarily determined by the temporary determination unit. The provisionally determined signal (hereinafter referred to as “temporary determination signal”) is input to the linear waveform equalizer and the nonlinear waveform equalizer as a teacher signal. The linear waveform equalizer and the nonlinear waveform equalizer perform equalization processing by deriving tap coefficients based on the teacher signal.
 例えば、非線形等化器には、ニューラルネットワークが使用されているが、上記の構成によれば、トレーニング信号を使用せずにニューラルネットワークの学習がなされる。また、線形波形等化器と非線形波形等化器とにおけるタップ係数の導出には、線形等化信号、非線形等化信号、仮判定信号が使用されるが、線形等化信号や非線形信号と、仮判定信号とは、出力タイミングが異なる。そのため、これらのタイミングを合わせるために、再生装置は、タップ係数導出のために、線形等化信号と非線形等化信号とを遅延させる。 For example, a neural network is used for the nonlinear equalizer, but according to the above configuration, the neural network is learned without using a training signal. In addition, a linear equalization signal, a non-linear equalization signal, and a temporary determination signal are used for derivation of tap coefficients in the linear waveform equalizer and the non-linear waveform equalizer. The output timing is different from the provisional determination signal. Therefore, in order to match these timings, the playback device delays the linear equalization signal and the nonlinear equalization signal in order to derive the tap coefficient.
 図1は、本発明の実施例1に係る再生装置100の構成を示す。再生装置100は、光ディスク10、光ディスク駆動部12、光ピックアップ14、プリアンプ部16、AGC部18、PLL(Phase Locked Loop)部20、A/D変換部22、処理部24、制御部26を含む。 FIG. 1 shows the configuration of a playback apparatus 100 according to Embodiment 1 of the present invention. The playback device 100 includes an optical disc 10, an optical disc drive unit 12, an optical pickup 14, a preamplifier unit 16, an AGC unit 18, a PLL (Phase Locked Loop) unit 20, an A / D conversion unit 22, a processing unit 24, and a control unit 26. .
 光ディスク10は、再生装置100に着脱可能に構成された記録媒体である。光ディスク10は、CD、DVD、BD、HD DVDのようなさまざまな種類に対応する。ここでは、特に光ディスク10として、非線形歪が再生に影響を及ぼす程度に大きい場合を対象にする。光ディスク駆動部12は、所定の回転速度で光ディスク10を回転させるためのモータである。光ピックアップ14は、光ディスク10から処理対象となる信号を読み出すとともに、これに対して光電変換および増幅を実行する。その結果の信号が、前述の「再生信号」に相当する。光ピックアップ14は、再生信号をプリアンプ部16へ出力する。 The optical disc 10 is a recording medium configured to be detachable from the playback device 100. The optical disk 10 corresponds to various types such as CD, DVD, BD, and HD DVD. Here, the optical disk 10 is particularly targeted for a case where the nonlinear distortion is large enough to affect reproduction. The optical disk drive unit 12 is a motor for rotating the optical disk 10 at a predetermined rotation speed. The optical pickup 14 reads a signal to be processed from the optical disc 10 and performs photoelectric conversion and amplification on the signal. The resulting signal corresponds to the “reproduction signal” described above. The optical pickup 14 outputs a reproduction signal to the preamplifier unit 16.
 プリアンプ部16は、再生信号を増幅し、AGC部18は、プリアンプ部16からの再生信号を所定の振幅に増幅する。AGC部18は、増幅した再生信号をPLL部20へ出力し、PLL部20は、再生信号からクロックを検出する。A/D変換部22は、PLL部20によって検出されたクロックをもとに、再生信号をアナログ/デジタル変換する。処理部24は、A/D変換部22においてアナログ/デジタル変換された再生信号(以下、これもまた「再生信号」という)に対して、等化処理および復号処理を実行する。処理部24の詳細は、後述する。 The preamplifier unit 16 amplifies the reproduction signal, and the AGC unit 18 amplifies the reproduction signal from the preamplifier unit 16 to a predetermined amplitude. The AGC unit 18 outputs the amplified reproduction signal to the PLL unit 20, and the PLL unit 20 detects a clock from the reproduction signal. The A / D converter 22 performs analog / digital conversion of the reproduction signal based on the clock detected by the PLL unit 20. The processing unit 24 performs equalization processing and decoding processing on the reproduction signal (hereinafter also referred to as “reproduction signal”) analog / digital converted by the A / D conversion unit 22. Details of the processing unit 24 will be described later.
 この構成は、ハードウエア的には、任意のコンピュータのCPU、メモリ、その他のLSIで実現でき、ソフトウエア的にはメモリにロードされたプログラムなどによって実現されるが、ここではそれらの連携によって実現される機能ブロックを描いている。したがって、これらの機能ブロックがハードウエアのみ、ソフトウエアのみ、またはそれらの組合せによっていろいろな形で実現できることは、当業者には理解されるところである。 This configuration can be realized in terms of hardware by a CPU, memory, or other LSI of any computer, and in terms of software, it can be realized by a program loaded in the memory, but here it is realized by their cooperation. Draw functional blocks. Accordingly, those skilled in the art will understand that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof.
 図2は、処理部24の構成を示す。処理部24は、線形等化部44、第1遅延部32、仮判定部30、非線形等化部46、第2遅延部34、第1加算部40、第2加算部42、ビタビ復号部38を含む。また、信号として、線形等化用誤差信号300、非線形等化用誤差信号302、仮判定信号306を含む。 FIG. 2 shows the configuration of the processing unit 24. The processing unit 24 includes a linear equalization unit 44, a first delay unit 32, a provisional determination unit 30, a nonlinear equalization unit 46, a second delay unit 34, a first addition unit 40, a second addition unit 42, and a Viterbi decoding unit 38. including. Further, the signal includes a linear equalization error signal 300, a non-linear equalization error signal 302, and a provisional determination signal 306.
 図1のA/D変換部22においてビットクロック毎にサンプリングされた再生信号は、線形等化部44に順次入力される。線形等化部44は、入力した再生信号を順次線形等化する。線形等化部44は、トランスバーサルフィルタにて構成されており、多段タップで再生信号を遅延させるとともに、多段タップからの出力と複数のタップ係数とを乗算し、かつ乗算結果を加算する。ここで、加算結果が、前述の線形等化信号に相当する。また、線形等化部44は、後述の第1加算部40から線形等化用誤差信号300を入力し、線形等化用誤差信号300をもとに、複数のタップ係数を導出する。ここで、複数のタップ係数の導出には、LMS(Least Mean Square)アルゴリズムのような適応アルゴリズムが使用される。線形等化部44は、第1遅延部32、仮判定部30、非線形等化部46へ線形等化信号を出力する。 The reproduction signal sampled for each bit clock in the A / D conversion unit 22 in FIG. 1 is sequentially input to the linear equalization unit 44. The linear equalizer 44 sequentially performs linear equalization on the input reproduction signal. The linear equalization unit 44 is configured by a transversal filter, delays the reproduction signal by a multistage tap, multiplies the output from the multistage tap and a plurality of tap coefficients, and adds the multiplication results. Here, the addition result corresponds to the linear equalization signal described above. Further, the linear equalization unit 44 receives the linear equalization error signal 300 from the first addition unit 40 described later, and derives a plurality of tap coefficients based on the linear equalization error signal 300. Here, an adaptive algorithm such as a LMS (Least Mean Square) algorithm is used to derive a plurality of tap coefficients. The linear equalization unit 44 outputs a linear equalization signal to the first delay unit 32, the provisional determination unit 30, and the non-linear equalization unit 46.
 非線形等化部46は、線形等化部44からの線形等化信号を入力し、線形等化信号を順次非線形等化する。非線形等化部46は、ニューラルネットワークにて構成されている。非線形等化部46における非線形等化の結果が、前述の非線形等化信号に相当する。また、非線形等化部46は、後述の第2加算部42から、非線形等化用誤差信号302を入力し、非線形等化用誤差信号302をもとに、ニューラルネットワークにおいて使用される複数のタップ係数を導出する。ここで、非線形等化用誤差信号302は、第2遅延部34からの遅延信号と仮判定信号306との差異にて生成されているので、非線形等化部46は、仮判定信号を教師信号として複数の係数を導出するといえる。非線形等化部46は、非線形等化信号を第2遅延部34とビタビ復号部38へ出力する。 The non-linear equalization unit 46 receives the linear equalization signal from the linear equalization unit 44 and sequentially performs non-linear equalization on the linear equalization signal. The nonlinear equalization unit 46 is configured by a neural network. The result of nonlinear equalization in the nonlinear equalization unit 46 corresponds to the aforementioned nonlinear equalization signal. Further, the nonlinear equalization unit 46 receives the nonlinear equalization error signal 302 from the second addition unit 42 described later, and based on the nonlinear equalization error signal 302, a plurality of taps used in the neural network. Deriving coefficients. Here, since the non-linear equalization error signal 302 is generated by the difference between the delayed signal from the second delay unit 34 and the temporary determination signal 306, the non-linear equalization unit 46 uses the temporary determination signal as the teacher signal. It can be said that a plurality of coefficients are derived. The nonlinear equalization unit 46 outputs the nonlinear equalization signal to the second delay unit 34 and the Viterbi decoding unit 38.
 ビタビ復号部38は、非線形等化部46からの非線形等化信号を入力し、非線形等化信号に対してビタビ復号を実行する。ビタビ復号部38は、非線形等化信号からブランチメトリックを計算するブランチメトリック演算回路と、ブランチメトリックを1クロック毎に累積加算してパスメトリックを計算するパスメトリック演算回路と、パスメトリックが最小となるデータ系列を最も確からしい候補系列として選択して記憶するパスメモリとを含む。パスメモリは、複数の候補系列を格納しており、パスメトリック演算回路からの選択信号にしたがって候補系列を選択する。また、選択された候補系列がデータ系列として出力される。 The Viterbi decoding unit 38 receives the nonlinear equalization signal from the nonlinear equalization unit 46 and executes Viterbi decoding on the nonlinear equalization signal. The Viterbi decoding unit 38 has a branch metric calculation circuit that calculates a branch metric from a non-linear equalization signal, a path metric calculation circuit that calculates a path metric by accumulating the branch metrics every clock, and a path metric is minimized. And a path memory for selecting and storing the data series as the most probable candidate series. The path memory stores a plurality of candidate sequences, and selects candidate sequences according to a selection signal from the path metric calculation circuit. In addition, the selected candidate series is output as a data series.
 仮判定部30は、線形等化部44からの線形等化信号を入力し、線形等化信号に対してビタビ復号を実行することによって、線形等化信号を順次仮判定する。仮判定部30は、ビタビ復号部38と同様に構成されている。パスメモリは、複数の候補系列を格納しており、パスメトリック演算回路からの選択信号をもとに、パーシャルレスポンス規則にしたがって仮判定が実行される。具体的に説明すると、仮判定部30は、パーシャルレスポンス等化が正常になされた場合に、所定の入力ビットに対する出力のレベルを仮判定し、入力ビットに対して仮判定したレベルを仮判定信号306として出力する。ここで、仮判定部30とビタビ復号部38とは、パスメモリ長が異なるように構成されている。例えば、ビタビ復号部38のパスメモリ長が64ビットである場合、仮判定部30のパスメモリ長は24ビットや32ビットである。 The provisional determination unit 30 receives the linear equalization signal from the linear equalization unit 44 and executes Viterbi decoding on the linear equalization signal to sequentially provisionally determine the linear equalization signal. The provisional determination unit 30 is configured in the same manner as the Viterbi decoding unit 38. The path memory stores a plurality of candidate series, and temporary determination is executed according to the partial response rule based on the selection signal from the path metric calculation circuit. More specifically, the provisional determination unit 30 provisionally determines the output level for a predetermined input bit when the partial response equalization is normally performed, and the provisional determination signal indicates the level temporarily determined for the input bit. Output as 306. Here, the temporary determination unit 30 and the Viterbi decoding unit 38 are configured to have different path memory lengths. For example, when the path memory length of the Viterbi decoding unit 38 is 64 bits, the path memory length of the temporary determination unit 30 is 24 bits or 32 bits.
 第1遅延部32は、線形等化部44からの線形等化信号を入力する。第1遅延部32は、線形等化信号を遅延させた後、遅延した線形等化信号(以下、「線形等化信号」あるいは「遅延信号」という)を第1加算部40へ出力する。ここで、第1遅延部32は、仮判定部30での処理遅延に応じた期間にわたって遅延を実行する。つまり、仮判定部30から出力された仮判定信号306と、線形等化部44からの線形等化信号とのタイミングが、第1加算部40において合わされる。第1遅延部32は、例えば、ビットクロックで駆動されるラッチ回路にて構成される。第1加算部40は、第1遅延部32からの線形等化信号、仮判定信号306を入力する。第1加算部40は、線形等化誤差と仮判定信号306との差異をもとに、線形等化用誤差信号300を生成する。例えば、線形等化誤差から仮判定信号306を減算することによって、線形等化用誤差信号300が導出される。第1加算部40は、線形等化用誤差信号300を線形等化部44へ出力する。 The first delay unit 32 receives the linear equalization signal from the linear equalization unit 44. The first delay unit 32 delays the linear equalization signal and then outputs the delayed linear equalization signal (hereinafter referred to as “linear equalization signal” or “delay signal”) to the first addition unit 40. Here, the first delay unit 32 executes the delay over a period corresponding to the processing delay in the provisional determination unit 30. That is, the timing of the provisional determination signal 306 output from the provisional determination unit 30 and the linear equalization signal from the linear equalization unit 44 are matched in the first addition unit 40. The first delay unit 32 is configured by, for example, a latch circuit driven by a bit clock. The first addition unit 40 receives the linear equalization signal and the temporary determination signal 306 from the first delay unit 32. The first addition unit 40 generates a linear equalization error signal 300 based on the difference between the linear equalization error and the provisional determination signal 306. For example, the linear equalization error signal 300 is derived by subtracting the provisional determination signal 306 from the linear equalization error. The first addition unit 40 outputs the linear equalization error signal 300 to the linear equalization unit 44.
 第2遅延部34は、非線形等化部46からの非線形等化信号を入力する。第2遅延部34は、非線形等化信号を遅延させた後、遅延した非線形等化信号(以下、「非線形等化信号」あるいは「遅延信号」という)を第2加算部42へ出力する。ここで、第2遅延部34は、仮判定部30における処理遅延と非線形等化部46における処理遅延との差異に応じた期間にわたって遅延を実行する。第2加算部42は、第2遅延部34からの非線形等化信号と、仮判定信号306との差異をもとに、非線形等化用誤差信号302を生成する。例えば、非線形等化信号和から仮判定信号306を減算することによって、非線形等化用誤差信号302が導出される。第2加算部42は、非線形等化用誤差信号302を非線形等化部46へ出力する。 The second delay unit 34 inputs the nonlinear equalization signal from the nonlinear equalization unit 46. The second delay unit 34 delays the nonlinear equalized signal and then outputs the delayed nonlinear equalized signal (hereinafter referred to as “nonlinear equalized signal” or “delayed signal”) to the second adder 42. Here, the second delay unit 34 executes the delay over a period corresponding to the difference between the processing delay in the provisional determination unit 30 and the processing delay in the nonlinear equalization unit 46. The second adder 42 generates a non-linear equalization error signal 302 based on the difference between the non-linear equalization signal from the second delay unit 34 and the provisional determination signal 306. For example, the non-linear equalization error signal 302 is derived by subtracting the provisional determination signal 306 from the non-linear equalization signal sum. The second addition unit 42 outputs the non-linear equalization error signal 302 to the non-linear equalization unit 46.
 ここで、非線形等化部46は、非線形等化用誤差信号302をもとに、複数の係数を導出する。つまり、非線形等化部46は、仮判定信号306を教師信号として使用する。また、非線形等化部46は、非線形等化用誤差信号302の二乗値を次々に加算していった総和である積算値を計算することによって、非線形等化部46における複数のタップ係数の収束を監視する。つまり、非線形等化部46は、その積算値がしきい値よりも大きい状態から小さい状態へ、非線形等化用誤差信号302が変化した場合に、複数のタップ係数の収束を判定する。また、収束を判定した後、非線形等化用誤差信号302の二乗値の総和である積算値がしきい値よりも再び大きくなった場合に、非線形等化部46の複数のタップ係数の発散と判定する。その際、非線形等化部46は、複数のタップ係数を新たに導出する。 Here, the non-linear equalization unit 46 derives a plurality of coefficients based on the non-linear equalization error signal 302. That is, the nonlinear equalizer 46 uses the temporary determination signal 306 as a teacher signal. Further, the non-linear equalization unit 46 calculates an integrated value that is a sum obtained by successively adding the square values of the non-linear equalization error signal 302, thereby converging a plurality of tap coefficients in the non-linear equalization unit 46. To monitor. That is, the non-linear equalization unit 46 determines the convergence of a plurality of tap coefficients when the non-linear equalization error signal 302 changes from a state where the integrated value is larger than a threshold value to a small state. Further, after the convergence is determined, when the integrated value that is the sum of the square values of the error signal 302 for nonlinear equalization becomes larger than the threshold value again, the divergence of the tap coefficients of the nonlinear equalization unit 46 judge. At this time, the nonlinear equalization unit 46 newly derives a plurality of tap coefficients.
 図3は、線形等化部44の構成を示す。線形等化部44は、多段タップ50、線形処理部52を含む。多段タップ50は、遅延タップ54と総称される第1遅延タップ54a、第2遅延タップ54b、第3遅延タップ54c、第N遅延タップ54nを含む。線形処理部52は、乗算部56と総称される第1乗算部56a、第2乗算部56b、第3乗算部56c、第N+1乗算部56n+1、タップ係数導出部58、積算部60を含む。 FIG. 3 shows the configuration of the linear equalization unit 44. The linear equalization unit 44 includes a multistage tap 50 and a linear processing unit 52. The multistage tap 50 includes a first delay tap 54a, a second delay tap 54b, a third delay tap 54c, and an Nth delay tap 54n, which are collectively referred to as a delay tap 54. The linear processing unit 52 includes a first multiplication unit 56a, a second multiplication unit 56b, a third multiplication unit 56c, an N + 1 multiplication unit 56n + 1, a tap coefficient derivation unit 58, and an integration unit 60, which are collectively referred to as a multiplication unit 56.
 多段タップ50は、複数の遅延タップ54がシリアルに接続されることによって形成される。具体的に説明すると、第1遅延タップ54aは、再生信号を入力し、遅延後、再生信号を出力する。第2遅延タップ54bは、第1遅延タップ54aからの再生信号を入力し、遅延後、再生信号を出力する。第3遅延タップ54cから第N遅延タップ54nも、同様の処理を実行する。遅延タップ54への入力部分と出力部分が多段タップ50からの出力信号であり、例えば、4つの遅延タップ54が配置される場合、5つの出力信号が存在する。これらの出力信号は、乗算部56へ出力されている。 The multistage tap 50 is formed by serially connecting a plurality of delay taps 54. More specifically, the first delay tap 54a inputs the reproduction signal, and outputs the reproduction signal after delay. The second delay tap 54b receives the reproduction signal from the first delay tap 54a, and outputs the reproduction signal after delay. The third delay tap 54c to the Nth delay tap 54n perform the same processing. An input portion and an output portion to the delay tap 54 are output signals from the multistage tap 50. For example, when four delay taps 54 are arranged, there are five output signals. These output signals are output to the multiplication unit 56.
 乗算部56は、遅延タップ54からの出力信号を入力するとともに、タップ係数導出部58からのタップ係数も入力する。ここで、タップ係数は、各出力信号に対応づけられて導出されている。乗算部56は、出力信号とタップ係数とを乗算する。乗算部56は、各乗算結果を積算部60へ出力する。積算部60は、乗算部56からの乗算結果を次々に加算して加算結果である積算値を求める。加算結果である積算値が、前述の線形等化信号に相当する。積算部60は、線形等化信号を出力する。タップ係数導出部58は、線形等化用誤差信号300を入力する。タップ係数導出部58は、再生信号がパーシャルレスポンス特性に適合するように、線形等化用誤差信号300、乗算部56での乗算結果を使用して、複数のタップ係数を制御する。なお、タップ係数の導出には、例えばLMSアルゴリズムのような適応アルゴリズムが使用されることによって、線形等化用誤差信号300が小さくなるように制御される。なお、LMSアルゴリズムは、公知の技術であるので、ここでは説明を省略する。 The multiplication unit 56 inputs the output signal from the delay tap 54 and also receives the tap coefficient from the tap coefficient deriving unit 58. Here, the tap coefficient is derived in association with each output signal. The multiplier 56 multiplies the output signal and the tap coefficient. The multiplication unit 56 outputs each multiplication result to the integration unit 60. The accumulating unit 60 adds the multiplication results from the multiplying unit 56 one after another to obtain an accumulated value as an addition result. The integrated value as the addition result corresponds to the above-described linear equalization signal. The integrating unit 60 outputs a linear equalization signal. The tap coefficient deriving unit 58 receives the linear equalization error signal 300. The tap coefficient deriving unit 58 controls a plurality of tap coefficients using the linear equalization error signal 300 and the multiplication result of the multiplication unit 56 so that the reproduction signal matches the partial response characteristic. The tap coefficient is derived by using an adaptive algorithm such as the LMS algorithm so that the linear equalization error signal 300 is controlled to be small. Since the LMS algorithm is a known technique, the description thereof is omitted here.
 図4は、非線形等化部46の構成を示す。非線形等化部46は、多段タップ70、非線形処理部72を含む。多段タップ70は、遅延タップ74と総称される第1遅延タップ74a、第2遅延タップ74b、第N遅延タップ74nを含む。非線形処理部72は、乗算部76と総称される第11乗算部76aa、第12乗算部76ab、第1M乗算部76am、第21乗算部76ba、第22乗算部76bb、第2M乗算部76bm、第(N+1)1乗算部76(n+1)a、第(N+1)2乗算部76(n+1)b、第(N+1)M乗算部76(n+1)m、積算部78と総称される第1積算部78a、第2積算部78b、第M積算部78m、関数演算部80と総称される第1関数演算部80a、第2関数演算部80b、第M関数演算部80m、乗算部82と総称される第1乗算部82a、第2乗算部82b、第M乗算部82m、積算部84、関数演算部86、タップ係数導出部88を含む。 FIG. 4 shows the configuration of the nonlinear equalization unit 46. The nonlinear equalization unit 46 includes a multistage tap 70 and a nonlinear processing unit 72. The multistage tap 70 includes a first delay tap 74a, a second delay tap 74b, and an Nth delay tap 74n, which are collectively referred to as a delay tap 74. The nonlinear processing unit 72 includes an eleventh multiplication unit 76aa, a twelfth multiplication unit 76ab, a first M multiplication unit 76am, a twenty-first multiplication unit 76ba, a twenty-second multiplication unit 76bb, a second M multiplication unit 76bm, (N + 1) 1 multiplier 76 (n + 1) a, (N + 1) 2 multiplier 76 (n + 1) b, (N + 1) M multiplier 76 (n + 1) m, and first integrator 78a collectively referred to as integrator 78. , A second integration unit 78b, an Mth integration unit 78m, a first function calculation unit 80a collectively referred to as a function calculation unit 80, a second function calculation unit 80b, an Mth function calculation unit 80m, and a multiplication unit 82. A 1-multiplier 82a, a second multiplier 82b, an M-th multiplier 82m, an integrator 84, a function calculator 86, and a tap coefficient derivation unit 88 are included.
 非線形等化部46は、図示のごとく、3層パーセプトロン型のニューラルネットワークにて構成される。ここで、入力層が多段タップ70に相当し、隠れ層が関数演算部80に相当し、出力層が関数演算部86に相当する。多段タップ70は、複数の遅延タップ74がシリアルに接続されることによって形成される。具体的に説明すると、第1遅延タップ74aは、線形等化信号を入力し、遅延後、線形等化信号を出力する。第2遅延タップ74bは、第1遅延タップ74aからの線形等化信号を入力し、遅延後、線形等化信号を出力する。第N遅延タップ74nも、同様の処理を実行する。遅延タップ74への入力部分と出力部分が多段タップ70からの出力信号である。これらの出力信号は、乗算部76へ出力されている。 The non-linear equalization unit 46 is configured by a three-layer perceptron type neural network as shown in the figure. Here, the input layer corresponds to the multistage tap 70, the hidden layer corresponds to the function calculation unit 80, and the output layer corresponds to the function calculation unit 86. The multistage tap 70 is formed by serially connecting a plurality of delay taps 74. More specifically, the first delay tap 74a receives a linear equalization signal and outputs a linear equalization signal after delaying. The second delay tap 74b receives the linear equalization signal from the first delay tap 74a, and outputs the linear equalization signal after the delay. The Nth delay tap 74n also performs the same process. An input portion and an output portion to the delay tap 74 are output signals from the multistage tap 70. These output signals are output to the multiplier 76.
 乗算部76は、多段タップ70からの出力信号と、タップ係数導出部88からのタップ係数とを乗算する。具体的に説明すると、第IJ乗算部76ijは、多段タップ70の先頭からi番目の出力信号S(i)と、タップ係数W1(i,j)とを乗算することによって、乗算結果U(i,j)を生成する。積算部78は、乗算部76における乗算結果を次々に加算する積算を行う。具体的に説明すると、第J積算部78jは、乗算結果U(1,j)、U(2,j)、U(3,j)、・・・、U(n+1,j)を加算する積算によって、積算結果V(j)を生成する。関数演算部80は、積算部78における積算結果V(j)にシグモイド関数を演算する。シグモイド関数は、次のように示される。
    f(x)=(1-exp(-αx))/(1+exp(-αx)) (式1)
 ここで、式1のxに積算結果V(j)が入力される。ここでは、第J関数演算部80jでの演算結果をX(j)と示し、当該演算結果が隠れ層からの出力に相当する。
The multiplication unit 76 multiplies the output signal from the multistage tap 70 and the tap coefficient from the tap coefficient deriving unit 88. More specifically, the IJ multiplication unit 76ij multiplies the i-th output signal S (i) from the top of the multistage tap 70 by the tap coefficient W1 (i, j), thereby obtaining a multiplication result U (i , J). The accumulating unit 78 performs accumulating by sequentially adding the multiplication results in the multiplying unit 76. More specifically, the J-th integrating unit 78j adds the multiplication results U (1, j), U (2, j), U (3, j),..., U (n + 1, j). To generate an integration result V (j). The function calculation unit 80 calculates a sigmoid function on the integration result V (j) in the integration unit 78. The sigmoid function is shown as follows.
f (x) = (1−exp (−αx)) / (1 + exp (−αx)) (Formula 1)
Here, the integration result V (j) is input to x in Expression 1. Here, the calculation result in the J-th function calculation unit 80j is denoted as X (j), and the calculation result corresponds to the output from the hidden layer.
 乗算部82は、関数演算部80における演算結果とタップ係数導出部88からのタップ係数とを乗算する。具体的に説明すると、第J乗算部82jは、第J関数演算部80jにおける演算結果X(j)と、タップ係数W2(j)とを乗算することによって、乗算結果Y(j)を生成する。積算部84は、乗算部82における乗算結果を次々に加算する積算を行う。ここでは、すべての乗算部82における乗算結果が積算され、積算結果Zが生成される。関数演算部86は、積算部84における積算結果にシグモイド関数を演算する。ここでは、式1のxに積算結果Zが入力される。関数演算部86の演算結果が、出力層からの出力に相当し、前述の非線形等化信号に相当する。 The multiplication unit 82 multiplies the calculation result in the function calculation unit 80 and the tap coefficient from the tap coefficient derivation unit 88. Specifically, the J-th multiplication unit 82j generates a multiplication result Y (j) by multiplying the calculation result X (j) in the J-th function calculation unit 80j by the tap coefficient W2 (j). . The accumulating unit 84 performs an accumulation by sequentially adding the multiplication results in the multiplying unit 82. Here, the multiplication results in all the multiplication units 82 are integrated, and an integration result Z is generated. The function calculation unit 86 calculates a sigmoid function on the integration result in the integration unit 84. Here, the integration result Z is input to x in Equation 1. The calculation result of the function calculation unit 86 corresponds to the output from the output layer, and corresponds to the above-described nonlinear equalization signal.
 タップ係数導出部88は、乗算部76および乗算部82において使用されるタップ係数W1(i,j)とW2(j)とを導出する。なお、W1(i,j)、W2(j)の初期値として、ランダムな値や収束後に近い値が設定される。また、タップ係数導出部88は、図3のタップ係数導出部58と同様にLMSアルゴリズムによって、W1(i,j)、W2(j)を更新する。ここで、W1(i,j)、W2(j)の学習は、バックプロパゲーションによってなされる。非線形等化用誤差信号302の二乗値は、次のように示される。
    E=(A-D) (式2)
The tap coefficient deriving unit 88 derives tap coefficients W1 (i, j) and W2 (j) used in the multiplying unit 76 and the multiplying unit 82. A random value or a value close to that after convergence is set as the initial value of W1 (i, j) and W2 (j). Further, the tap coefficient deriving unit 88 updates W1 (i, j) and W2 (j) by the LMS algorithm similarly to the tap coefficient deriving unit 58 of FIG. Here, learning of W1 (i, j) and W2 (j) is performed by back propagation. The square value of the non-linear equalization error signal 302 is expressed as follows.
E = (AD) 2 (Formula 2)
 ここで、Aは、線形等化信号に相当し、Dは、仮判定信号306に相当する。つまり、A-Dは、非線形等化用誤差信号302に相当する。タップ係数導出部88は、Eが最小となるように、W1(i,j)、W2(j)を制御する。出力層でのバックプロパゲーションの結果は次のように示される。
    (∂E)/(∂Y(j))=f’(Y(j))×2(A-D) (式3)
 タップ係数導出部88は、タップ係数W2(j)を次のように更新する。
    W2(j)=W2(j)old-ε×(∂E)/(∂W2(j)) (式4)
Here, A corresponds to the linear equalization signal, and D corresponds to the provisional determination signal 306. That is, AD corresponds to the error signal 302 for nonlinear equalization. The tap coefficient deriving unit 88 controls W1 (i, j) and W2 (j) so that E is minimized. The result of back propagation at the output layer is shown as follows.
(∂E) / (∂Y (j)) = f ′ (Y (j)) × 2 (AD) (Formula 3)
The tap coefficient deriving unit 88 updates the tap coefficient W2 (j) as follows.
W2 (j) = W2 (j) old −ε × (∂E) / (∂W2 (j)) (Formula 4)
 ここで、W2(j)oldは、ひとつ前のタイミングにおけるタップ係数W2(j)を示す。一方、隠れ層でのバックプロパゲーションは次のように示される。
    (∂E)/(∂U(i,j))=
      f’(U(i,j))×(∂E)/(∂Y(j))×W2(j) (式5)
 タップ係数導出部88は、タップ係数W1(i,j)を次のように更新する。
    W1(i,j)=
      W1(i,j)old-ε×(∂E)/(∂W1(i,j)) (式6)
 ここで、W1(i,j)oldは、ひとつ前のタイミングにおけるタップ係数W1(i,j)を示す。
Here, W2 (j) old indicates the tap coefficient W2 (j) at the previous timing. On the other hand, back propagation in the hidden layer is shown as follows.
(∂E) / (∂U (i, j)) =
f ′ (U (i, j)) × (∂E) / (∂Y (j)) × W2 (j) (Formula 5)
The tap coefficient deriving unit 88 updates the tap coefficient W1 (i, j) as follows.
W1 (i, j) =
W1 (i, j) old −ε × (∂E) / (∂W1 (i, j)) (Formula 6)
Here, W1 (i, j) old indicates the tap coefficient W1 (i, j) at the previous timing.
 図5は、仮判定部30の構成を示す。仮判定部30は、ブランチメトリック演算部90、パスメモリ部92、特定部96を含む。また、信号として、選択信号SELを含む。ブランチメトリック演算部90は、図示しない線形等化部44からの線形等化信号をもとに、ブランチメトリック演算およびパスメトリック演算を実行する。そのため、ブランチメトリック演算部90には、前述のブランチメトリック演算回路およびパスメトリック演算回路が含まれる。前述のごとく、本実施例では、パーシャルレスポンス方式が適用されているが、仮判定部30の構成を説明する前に、ここでは、パーシャルレスポンス方式における状態遷移を説明する。 FIG. 5 shows a configuration of the provisional determination unit 30. The provisional determination unit 30 includes a branch metric calculation unit 90, a path memory unit 92, and a specifying unit 96. Further, the selection signal SEL is included as a signal. The branch metric calculation unit 90 performs branch metric calculation and path metric calculation based on a linear equalization signal from a linear equalization unit 44 (not shown). Therefore, the branch metric calculation unit 90 includes the aforementioned branch metric calculation circuit and path metric calculation circuit. As described above, the partial response method is applied in the present embodiment, but before the configuration of the provisional determination unit 30 is described, here, state transition in the partial response method will be described.
 図6は、仮判定部30がパーシャルレスポンス(1,2,2,2,1)に対応する場合の状態遷移を示す。パーシャルレスポンス(1,2,2,2,1)では、振幅が±4の範囲に収まる。4ビットをひとつの組合せとすれば、組合せに含まれる値に応じて、S0からS9までの10状態が規定されている。また、次に入力されるビット値に応じて図示のごとく、状態が遷移する。例えば、状態S0にビット値「1」が入力されると、状態S1への遷移がなされる。ここで、状態間を結ぶ矢印に「x/y」のような値が示されているが、xは、入力されるビット値を示し、yは、もとの状態に新たなビット値が加わった5ビットに対する仮判定値を示す。図7は、仮判定部30がパーシャルレスポンス(1,2,2,2,1)に対応する場合の状態遷移を示す。図7は、連続したふたつのタイミングでの状態を示しており、各状態は、図6と同様である。 FIG. 6 shows a state transition when the temporary determination unit 30 corresponds to a partial response (1, 2, 2, 2, 1). In the partial response (1, 2, 2, 2, 1), the amplitude falls within a range of ± 4. If 4 bits are taken as one combination, 10 states from S0 to S9 are defined according to the values included in the combination. Further, the state transitions as shown in the figure according to the next input bit value. For example, when a bit value “1” is input to the state S0, a transition to the state S1 is made. Here, a value such as “x / y” is shown in the arrows connecting the states, where x indicates an input bit value and y indicates a new bit value added to the original state. The temporary decision value for 5 bits is shown. FIG. 7 shows a state transition when the provisional determination unit 30 corresponds to a partial response (1, 2, 2, 2, 1). FIG. 7 shows a state at two consecutive timings, and each state is the same as FIG.
 図8は、ブランチメトリック演算部90の構成を示す。ブランチメトリック演算部90は、加算部110と総称される第1加算部110a、第2加算部110b、第3加算部110c、第4加算部110d、第5加算部110e、第6加算部110f、第7加算部110g、第8加算部110h、第9加算部110i、第10加算部110j、第11加算部110k、第12加算部110l、第13加算部110m、第14加算部110n、第15加算部110o、第16加算部110p、二乗回路112と総称される第1二乗回路112a、第2二乗回路112b、第3二乗回路112c、第4二乗回路112d、第5二乗回路112e、第6二乗回路112f、第7二乗回路112g、第8二乗回路112h、第9二乗回路112i、第10二乗回路112j、第11二乗回路112k、第12二乗回路112l、第13二乗回路112m、第14二乗回路112n、第15二乗回路112o、第16二乗回路112p、ACS回路114と総称される第1ACS回路114a、第2ACS回路114b、第3ACS回路114c、第4ACS回路114d、第5ACS回路114e、第6ACS回路114f、加算部116と総称される第1加算部116a、第2加算部116b、第3加算部116c、第4加算部116dを含む。また、選択信号SELと総称される第0選択信号SEL0、第1選択信号SEL1、第2選択信号SEL2、第7選択信号SEL7、第8選択信号SEL8、第9選択信号SEL9を含む。 FIG. 8 shows the configuration of the branch metric calculation unit 90. The branch metric calculation unit 90 includes a first addition unit 110a, a second addition unit 110b, a third addition unit 110c, a fourth addition unit 110d, a fifth addition unit 110e, a sixth addition unit 110f, which are collectively referred to as an addition unit 110. The seventh adder 110g, the eighth adder 110h, the ninth adder 110i, the tenth adder 110j, the eleventh adder 110k, the twelfth adder 110l, the thirteenth adder 110m, the fourteenth adder 110n, the fifteenth Adder 110o, sixteenth adder 110p, first square circuit 112a, second square circuit 112b, third square circuit 112c, fourth square circuit 112d, fifth square circuit 112e, sixth square Circuit 112f, seventh square circuit 112g, eighth square circuit 112h, ninth square circuit 112i, tenth square circuit 112j, eleventh square circuit 112k, first A first ACS circuit 114a, a second ACS circuit 114b, a third ACS circuit 114c, collectively referred to as a square circuit 112l, a thirteenth square circuit 112m, a fourteenth square circuit 112n, a fifteenth square circuit 112o, a sixteenth square circuit 112p, an ACS circuit 114, It includes a fourth ACS circuit 114d, a fifth ACS circuit 114e, a sixth ACS circuit 114f, a first adder 116a, a second adder 116b, a third adder 116c, and a fourth adder 116d, which are collectively referred to as an adder 116. Further, it includes a 0th selection signal SEL0, a first selection signal SEL1, a second selection signal SEL2, a seventh selection signal SEL7, an eighth selection signal SEL8, and a ninth selection signal SEL9, which are collectively referred to as a selection signal SEL.
 加算部110は、線形等化信号から所定の目標値を減じる。二乗回路112は、加算部110における減算結果の二乗値を計算する。ACS回路114は、二乗回路112からの二乗に対して、加算、比較、選択によるメトリック演算を実行する。また、ACS回路114は、メトリック演算の結果として、第0選択信号SEL0、第1選択信号SEL1、第2選択信号SEL2、第7選択信号SEL7、第8選択信号SEL8、第9選択信号SEL9を出力する。また、パーシャルレスポンス特性からACS回路114へ入力されない二乗値も存在する。そのような二乗値に対して、加算部116において加算がなされる。図5に戻る。 The addition unit 110 subtracts a predetermined target value from the linear equalization signal. The square circuit 112 calculates the square value of the subtraction result in the addition unit 110. The ACS circuit 114 performs a metric operation on the square from the square circuit 112 by addition, comparison, and selection. Further, the ACS circuit 114 outputs a 0th selection signal SEL0, a first selection signal SEL1, a second selection signal SEL2, a seventh selection signal SEL7, an eighth selection signal SEL8, and a ninth selection signal SEL9 as a result of the metric calculation. To do. There is also a square value that is not input to the ACS circuit 114 due to the partial response characteristic. The adder 116 adds such a square value. Returning to FIG.
 パスメモリ部92は、ブランチメトリック演算部90からの選択信号SELを入力し、選択信号SELに応じたパスを記憶する。図9は、パスメモリ部92の構成を示す。パスメモリ部92は、メモリ120と総称される第11メモリ120aa、第12メモリ120ab、第13メモリ120ac、第14メモリ120ad、第15メモリ120ae、第16メモリ120af、第17メモリ120ag、第18メモリ120ah、第19メモリ120ai、第110メモリ120aj、第21メモリ120ba、第22メモリ120bb、第23メモリ120bc、第24メモリ120bd、第25メモリ120be、第26メモリ120bf、第27メモリ120bg、第28メモリ120bh、第29メモリ120bi、第210メモリ120bj、第(L+1)1メモリ120(l+1)a、第(L+1)2メモリ120(l+1)b、第(L+1)3メモリ120(l+1)c、第(L+1)4メモリ120(l+1)d、第(L+1)5メモリ120(l+1)e、第(L+1)6メモリ120(l+1)f、第(L+1)7メモリ120(l+1)g、第(L+1)8メモリ120(l+1)h、第(L+1)9メモリ120(l+1)i、第(L+1)10メモリ120(l+1)j、選択部122と総称される第11選択部122aa、第12選択部122ab、第13選択部122ac、第14選択部122ad、第15選択部122ae、第16選択部122af、第L1選択部122la、第L2選択部122lb、第L3選択部122lc、第L4選択部122ld、第L5選択部122le、第6L6選択部122lf、多数決部124を含む。 The path memory unit 92 receives the selection signal SEL from the branch metric calculation unit 90 and stores a path corresponding to the selection signal SEL. FIG. 9 shows the configuration of the path memory unit 92. The path memory unit 92 includes an eleventh memory 120aa, a twelfth memory 120ab, a thirteenth memory 120ac, a fourteenth memory 120ad, a fifteenth memory 120ae, a sixteenth memory 120af, a seventeenth memory 120ag, and an eighteenth memory. 120ah, 19th memory 120ai, 110th memory 120aj, 21st memory 120ba, 22nd memory 120bb, 23rd memory 120bc, 24th memory 120bd, 25th memory 120be, 26th memory 120bf, 27th memory 120bg, 28th memory 120bh, 29th memory 120bi, 210th memory 120bj, (L + 1) 1 memory 120 (l + 1) a, (L + 1) 2 memory 120 (l + 1) b, (L + 1) 3 memory 120 (l + 1) c, (( L + 1) 4 notes 120 (l + 1) d, (L + 1) 5th memory 120 (l + 1) e, (L + 1) 6th memory 120 (l + 1) f, (L + 1) 7th memory 120 (l + 1) g, (L + 1) 8th memory 120 ( l + 1) h, (L + 1) 9th memory 120 (l + 1) i, (L + 1) 10th memory 120 (l + 1) j, eleventh selection unit 122aa, twelfth selection unit 122ab, thirteenth selection collectively referred to as selection unit 122 122ac, 14th selector 122ad, 15th selector 122ae, 16th selector 122af, L1 selector 122la, L2 selector 122lb, L3 selector 122lc, L4 selector 122ld, L5 selector 122le , A sixth L6 selection unit 122lf and a majority decision unit 124.
 ここでは、L+1のメモリ120によってひとつのパスが記憶され、かつ図6、7に示した10種類の状態のそれぞれに対応するように、10種類のパスが記憶される。選択部122は、選択信号SELに応じて、いずれかのパスを選択する。選択されたパスが、生き残りパス相当する。多数決部124は、第(L+1)1メモリ120(l+1)aから第(L+1)10メモリ120(l+1)jのそれぞれに記憶されたビット値を入力し、多数決を実行する。多数決部124は、選択結果を出力する。図5に戻る。 Here, one path is stored in the L + 1 memory 120, and 10 types of paths are stored so as to correspond to each of the 10 types of states shown in FIGS. The selection unit 122 selects one of the paths according to the selection signal SEL. The selected path corresponds to the survival path. The majority decision unit 124 inputs bit values stored in the (L + 1) th memory 120 (l + 1) a to the (L + 1) th memory 120 (l + 1) j, respectively, and executes the majority decision. The majority decision unit 124 outputs the selection result. Returning to FIG.
 特定部96は、図示しない多数決部124からの選択値を入力し、ラッチにて選択値を保持する。ここで、特定部96は、過去の選択値を含めて、5つのタイミングに対応した選択値からひとつの組合せを選択する。なお、特定部96に新たな選択値が入力されると、組合せの中から最も過去の選択値が除外されることによって、組合せが更新される。 The specifying unit 96 inputs the selection value from the majority decision unit 124 (not shown) and holds the selection value in a latch. Here, the specifying unit 96 selects one combination from selection values corresponding to five timings including past selection values. When a new selection value is input to the specifying unit 96, the combination is updated by removing the oldest selection value from the combination.
 図10は、特定部96に記憶されたテーブルのデータ構造を示す。図示のごとく、メモリ値欄200、b(k)欄202、b(k-1)欄204、b(k-2)欄206、b(k-3)欄208、b(k-4)欄210、仮判定出力欄212が含まれる。ここで、b(k)は、最も新しく入力された選択値に相当し、b(k-1)は、ひとつ前のタイミングに入力された選択値に相当し、b(k-4)は、4つ前のタイミングに入力された選択値に相当する。前述のごとく、これらはラッチにて保持されている。b(k)欄202からb(k-4)欄210には、ラッチに保持された選択値が取り得る値の組合せが示されている。メモリ値欄200では、取り得る値に対応したメモリ値が示され、仮判定出力欄212では、取り得る値に対応した仮判定値が示されている。例えば、パスメモリの内容が「00000」であれば仮判定値「-4」、「00001」であれば仮判定値「-3」が対応づけられている。図5に戻る。特定部96は、図10に示したテーブルを参照しながら、組合せに対応した仮判定値を特定する。特定部96は、仮判定値を仮判定信号306として出力する。 FIG. 10 shows the data structure of the table stored in the specifying unit 96. As shown, memory value column 200, b (k) column 202, b (k-1) column 204, b (k-2) column 206, b (k-3) column 208, b (k-4) column. 210 and a provisional determination output column 212 are included. Here, b (k) corresponds to the most recently input selection value, b (k−1) corresponds to the selection value input at the previous timing, and b (k−4) This corresponds to the selection value input at the previous four timings. As described above, these are held by the latch. In the b (k) column 202 to the b (k-4) column 210, combinations of values that the selection values held in the latch can take are shown. In the memory value column 200, memory values corresponding to possible values are shown, and in the temporary determination output column 212, temporary determination values corresponding to possible values are shown. For example, if the content of the path memory is “00000”, the temporary determination value “−4” is associated, and if “00001”, the temporary determination value “−3” is associated. Returning to FIG. The specifying unit 96 specifies a provisional determination value corresponding to the combination while referring to the table shown in FIG. The specifying unit 96 outputs the temporary determination value as the temporary determination signal 306.
 以上の構成による再生装置100の動作を説明する。図11は、非線形等化部46における係数の導出手順を示すフローチャートである。非線形等化部46は、非線形等化用誤差信号302の大きさが収束した後も、継続して非線形等化用誤差信号302の大きさを導出する。大きさがしきい値よりも大きくなった場合(S40のY)、非線形等化部46は、新たにタップ係数を導出する(S42)。大きさがしきい値よりも大きくならなければ(S40のN)、処理は終了される。 The operation of the playback apparatus 100 configured as above will be described. FIG. 11 is a flowchart showing a procedure for deriving coefficients in the nonlinear equalization unit 46. The nonlinear equalization unit 46 continues to derive the magnitude of the nonlinear equalization error signal 302 even after the magnitude of the nonlinear equalization error signal 302 has converged. When the magnitude is larger than the threshold value (Y in S40), the nonlinear equalization unit 46 newly derives a tap coefficient (S42). If the size does not become larger than the threshold value (N in S40), the process is terminated.
 図12(a)-(b)は、従来および図1の再生装置100による出力信号のヒストグラムを示す。図12(a)は、従来の線形波形等化器で等化した信号のヒストグラムを示す。このときのビットエラーレートは、1.1×10-2である。一方、図12(b)は、再生装置100で等化した信号のヒストグラムを示す。このときの目標値は、前述のごとく、パーシャルレスポンス(1,2,2,2,1)の9値としている。また、このときのビットエラーレートは、1.5×10-4である。従来の線形波形等化器における特性悪化は、波形に非線形成分が含まれているので、ビタビ目標値に収束がなされないためであると推定される。 12 (a)-(b) show histograms of output signals obtained by the conventional and the reproducing apparatus 100 of FIG. FIG. 12A shows a histogram of signals equalized by a conventional linear waveform equalizer. The bit error rate at this time is 1.1 × 10 −2 . On the other hand, FIG. 12B shows a histogram of signals equalized by the playback apparatus 100. The target value at this time is 9 values of partial response (1, 2, 2, 2, 1) as described above. Further, the bit error rate at this time is 1.5 × 10 −4 . It is estimated that the characteristic deterioration in the conventional linear waveform equalizer is due to the fact that the waveform does not converge to the Viterbi target value because the waveform includes a nonlinear component.
 本発明の実施例によれば、線形等化信号を仮判定した結果を教師信号とするので、トレーニング信号の代わりに仮判定信号を教師信号に使用できる。また、トレーニング信号の代わりに仮判定信号を教師信号に使用するので、トレーニング信号を使用せずに、非線形等化のための係数を導出できる。また、トレーニング信号を使用せずに、非線形等化のための係数が導出されるので、トレーニング信号を使用せずに非線形等化を実行できる。また、トレーニング信号を使用せずに、非線形等化のための係数が導出されるので、記録密度の向上や記録パワー変動等によって生じる再生信号の非線形歪をトレーニング信号なしで低減できる。 According to the embodiment of the present invention, the result of provisional determination of the linear equalization signal is used as the teacher signal, so that the provisional determination signal can be used as the teacher signal instead of the training signal. Further, since the temporary determination signal is used as the teacher signal instead of the training signal, a coefficient for nonlinear equalization can be derived without using the training signal. Further, since the coefficient for nonlinear equalization is derived without using the training signal, nonlinear equalization can be performed without using the training signal. In addition, since a coefficient for nonlinear equalization is derived without using a training signal, nonlinear distortion of a reproduction signal caused by improvement in recording density, recording power fluctuation, or the like can be reduced without a training signal.
 また、パーシャルレスポンス規則にしたがった仮判定が実行されるので、パーシャルレスポンス処理に対応できる。また、仮判定部における処理遅延と非線形等化部における処理遅延との差異に応じた期間にわたって遅延を実行するので、非線形等化信号と仮判定信号とのタイミングを合わせることができる。また、非線形等化信号と仮判定信号とのタイミングが合わされるので、非線形等化のためのタップ係数の推定精度を向上できる。また、非線形等化用誤差信号の発散を検出した場合に複数の係数を新たに導出するので、等化特性の悪化を抑制できる。 Also, provisional judgment is executed according to the partial response rule, so it can support partial response processing. In addition, since the delay is executed over a period corresponding to the difference between the processing delay in the provisional determination unit and the processing delay in the nonlinear equalization unit, the timing of the nonlinear equalization signal and the provisional determination signal can be matched. In addition, since the timing of the nonlinear equalization signal and the provisional determination signal are matched, it is possible to improve the estimation accuracy of tap coefficients for nonlinear equalization. Further, since a plurality of coefficients are newly derived when the divergence of the non-linear equalization error signal is detected, deterioration of the equalization characteristics can be suppressed.
(実施例2)
 本発明を具体的に説明する前に、まず概要を述べる。本発明の実施例2は、光ディスク等の記録媒体に記録されている信号を再生し、再生した信号(以下、「再生信号」という)をパーシャルレスポンス方式にて等化するととともに、等化した信号(以下、「等化信号」という)を復号する再生装置に関する。前述のごとく、光ディスクの記録容量が高まるにつれ、線形波形等化器では除去しきれない非線形歪の影響が大きくなっている。非線形歪を除去するためには、非線形等化器としてのニューラルネットが有効であるが、トレーニング信号によって学習、収束させる必要がある。そこで、再生信号の非線形歪をトレーニング信号なしで低減するために、本実施例に係る再生装置は次の処理を実行する。
(Example 2)
Before describing the present invention specifically, an outline will be given first. The second embodiment of the present invention reproduces a signal recorded on a recording medium such as an optical disk, equalizes the reproduced signal (hereinafter referred to as “reproduced signal”) using a partial response method, and equalizes the signal. The present invention relates to a playback apparatus for decoding (hereinafter referred to as “equalized signal”). As described above, as the recording capacity of the optical disk increases, the influence of nonlinear distortion that cannot be removed by the linear waveform equalizer increases. In order to remove the nonlinear distortion, a neural network as a nonlinear equalizer is effective, but it is necessary to learn and converge with a training signal. Therefore, in order to reduce the non-linear distortion of the reproduction signal without the training signal, the reproduction apparatus according to the present embodiment executes the following processing.
 再生装置は、線形波形等化器と非線形波形等化器とを並列に配置し、いずれにも再生信号を入力する。また、再生装置は、線形波形等化器からの等化信号(以下、「線形等化信号」という)と非線形波形等化器からの等化信号(以下、「非線形等化信号」という)とを合成し、合成した信号(以下、「加算信号」という)をビタビ復号器に入力する。ここで、ビタビ復号器は、加算信号を順次仮判定する本発明の仮判定部として機能する。ビタビ復号器において仮判定された信号(以下、「仮判定信号」という)は、教師信号として、線形波形等化器と非線形波形等化器とに入力される。線形波形等化器と非線形波形等化器とは、教師信号をもとにタップ係数を導出して等化処理を実行する。例えば、非線形等化器には、ニューラルネットワークが使用されているので、トレーニング信号を使用せずにニューラルネットワークの学習がなされる。 The playback device has a linear waveform equalizer and a nonlinear waveform equalizer arranged in parallel, and inputs a playback signal to both. In addition, the reproducing apparatus includes an equalized signal from a linear waveform equalizer (hereinafter referred to as “linear equalized signal”) and an equalized signal from a nonlinear waveform equalizer (hereinafter referred to as “nonlinear equalized signal”). And the synthesized signal (hereinafter referred to as “addition signal”) is input to the Viterbi decoder. Here, the Viterbi decoder functions as a provisional determination unit of the present invention that sequentially provisionally determines the addition signal. A signal provisionally determined by the Viterbi decoder (hereinafter referred to as “provisional determination signal”) is input as a teacher signal to the linear waveform equalizer and the nonlinear waveform equalizer. The linear waveform equalizer and the nonlinear waveform equalizer perform equalization processing by deriving tap coefficients based on the teacher signal. For example, since the neural network is used for the nonlinear equalizer, the neural network is learned without using the training signal.
 ここで、トランスバーサル型フィルタで構成される線形波形等化器と、ニューラルネットワークで構成される非線形波形等化器の両方を組み合わせても適応動作を可能にするために、再生装置は、さらに次の処理を実行する。一般的に、線形波形等化器よりも非線形波形等化器の方がタップ係数の収束までに要する期間が長い。ニューラルネットワークが収束する前の不安点な動作を解消するために、再生装置は、ニューラルネットワークの学習の収束状況を監視しており、収束が確認されなければ、加算信号ではなく、線形等化信号をビタビ復号器へ出力する。また、線形波形等化器と非線形波形等化器とにおけるタップ係数の導出には、線形等化信号、非線形等化信号、仮判定信号が使用されるが、線形等化信号や非線形信号と、仮判定信号とは、出力タイミングが異なる。そのため、これらのタイミングを合わせるために、再生装置は、線形等化信号と非線形等化信号とを遅延させる。以下、本発明の実施例を説明する。 Here, in order to enable adaptive operation even when both a linear waveform equalizer configured with a transversal filter and a nonlinear waveform equalizer configured with a neural network are combined, the reproducing apparatus further performs the following operation. Execute the process. In general, the nonlinear waveform equalizer requires a longer period for tap coefficient convergence than the linear waveform equalizer. In order to eliminate the anxious operation before the neural network converges, the playback device monitors the learning convergence state of the neural network, and if convergence is not confirmed, the linear equalization signal is used instead of the addition signal. Is output to the Viterbi decoder. In addition, a linear equalization signal, a non-linear equalization signal, and a temporary determination signal are used for derivation of tap coefficients in the linear waveform equalizer and the non-linear waveform equalizer. The output timing is different from the provisional determination signal. Therefore, in order to match these timings, the playback device delays the linear equalized signal and the nonlinear equalized signal. Examples of the present invention will be described below.
 図13は、処理部24の構成を示す。処理部24は、等化処理部1030、第1遅延部1032、第2遅延部1034、加算部1036、ビタビ復号部1038、等化誤差生成部1040、判定部1042を含む。また、等化処理部1030は、線形等化部1044、非線形等化部1046を含む。また、等化誤差生成部1040、非線形等化部1046は、適応非線形等化部1048とグループ化される。さらに、信号として、線形等化用誤差信号1300、非線形等化用誤差信号1302、仮判定信号1306を含む。なお、本実施例では、ビタビ復号部1038が加算部1036によって加算された信号を順次仮判定する本発明の仮判定部として機能するが、ビタビ復号部1038とは別に仮判定部を設けてもかまわない。 FIG. 13 shows the configuration of the processing unit 24. The processing unit 24 includes an equalization processing unit 1030, a first delay unit 1032, a second delay unit 1034, an addition unit 1036, a Viterbi decoding unit 1038, an equalization error generation unit 1040, and a determination unit 1042. The equalization processing unit 1030 includes a linear equalization unit 1044 and a non-linear equalization unit 1046. Also, the equalization error generation unit 1040 and the non-linear equalization unit 1046 are grouped with the adaptive non-linear equalization unit 1048. Further, the signal includes a linear equalization error signal 1300, a non-linear equalization error signal 1302, and a provisional determination signal 1306. In this embodiment, the Viterbi decoding unit 1038 functions as a temporary determination unit of the present invention that sequentially determines the signals added by the addition unit 1036. However, a temporary determination unit may be provided separately from the Viterbi decoding unit 1038. It doesn't matter.
 図1のA/D変換部22においてビットクロック毎にサンプリングされた再生信号は、線形等化部1044と非線形等化部1046とに順次入力される。線形等化部1044は、入力した再生信号を順次線形等化する。線形等化部1044は、トランスバーサルフィルタにて構成されており、多段タップで再生信号を遅延させるとともに、多段タップからの出力と複数のタップ係数とを乗算し、かつ乗算結果を加算する。ここで、加算結果が、前述の線形等化信号に相当する。また、線形等化部1044は、後述の等化誤差生成部1040から線形等化用誤差信号1300を入力し、線形等化用誤差信号1300をもとに、複数のタップ係数を導出する。ここで、複数のタップ係数の導出には、LMS(Least Mean Square)アルゴリズムのような適応アルゴリズムが使用される。線形等化部1044は、線形等化信号を第1遅延部1032を介して等化誤差生成部1040へ出力する。 The reproduction signal sampled for each bit clock in the A / D conversion unit 22 in FIG. 1 is sequentially input to the linear equalization unit 1044 and the non-linear equalization unit 1046. The linear equalizer 1044 sequentially performs linear equalization on the input reproduction signal. The linear equalization unit 1044 is configured by a transversal filter, delays the reproduction signal by a multistage tap, multiplies the output from the multistage tap and a plurality of tap coefficients, and adds the multiplication results. Here, the addition result corresponds to the linear equalization signal described above. The linear equalization unit 1044 receives a linear equalization error signal 1300 from an equalization error generation unit 1040 described later, and derives a plurality of tap coefficients based on the linear equalization error signal 1300. Here, an adaptive algorithm such as a LMS (Least Mean Square) algorithm is used to derive a plurality of tap coefficients. The linear equalization unit 1044 outputs the linear equalization signal to the equalization error generation unit 1040 via the first delay unit 1032.
 非線形等化部1046は、線形等化部1044における線形等化に並行して、再生信号を順次非線形等化する。非線形等化部1046は、ニューラルネットワークにて構成されている。非線形等化部1046における非線形等化の結果が、前述の非線形等化信号に相当する。また、非線形等化部1046は、後述の等化誤差生成部1040から、非線形等化用誤差信号1302を入力し、非線形等化用誤差信号1302をもとに、ニューラルネットワークにおいて使用される複数のタップ係数を導出する。ここで、非線形等化用誤差信号1302は、第1遅延部1032からの遅延信号と第2遅延部1034からの遅延信号との和と、仮判定信号との差異にて生成されているので、非線形等化部1046は、仮判定信号を教師信号として複数の係数を導出するといえる。非線形等化部1046は、非線形等化信号を、第2遅延部1034を介して等化誤差生成部1040と、第2遅延部1034を介さずに直接に加算部1036へ出力する。 The non-linear equalization unit 1046 sequentially performs non-linear equalization on the reproduction signal in parallel with the linear equalization in the linear equalization unit 1044. The nonlinear equalization unit 1046 is configured by a neural network. The result of nonlinear equalization in the nonlinear equalization unit 1046 corresponds to the aforementioned nonlinear equalization signal. The nonlinear equalization unit 1046 receives a nonlinear equalization error signal 1302 from an equalization error generation unit 1040, which will be described later, and based on the nonlinear equalization error signal 1302, a plurality of signals used in the neural network. A tap coefficient is derived. Here, the non-linear equalization error signal 1302 is generated by the difference between the sum of the delay signal from the first delay unit 1032 and the delay signal from the second delay unit 1034 and the provisional determination signal. It can be said that the nonlinear equalization unit 1046 derives a plurality of coefficients using the temporary determination signal as a teacher signal. The nonlinear equalization unit 1046 outputs the nonlinear equalization signal to the equalization error generation unit 1040 via the second delay unit 1034 and directly to the addition unit 1036 without passing through the second delay unit 1034.
 第1遅延部1032は、線形等化部1044からの線形等化信号を入力する。第1遅延部1032は、線形等化信号を遅延させた後、遅延した線形等化信号(以下、「線形等化信号」あるいは「遅延信号」という)を等化誤差生成部1040へ出力する。ここで、第1遅延部1032は、ビタビ復号部1038における仮判定のための処理遅延に応じた期間にわたって遅延を実行する。つまり、第1遅延部1032によって、線形等化部1044からビタビ復号部1038へ至った後にビタビ復号部1038から出力された仮判定信号1306と、線形等化部1044からの線形等化信号とのタイミングが合わされる。第1遅延部1032は、例えば、ビットクロックで駆動されるラッチ回路にて構成される。第2遅延部1034は、非線形等化部1046からの非線形等化信号を入力する。第2遅延部1034は、非線形等化信号を遅延させた後、遅延した非線形等化信号(以下、「非線形等化信号」あるいは「遅延信号」という)を等化誤差生成部1040へ出力する。ここで、第2遅延部1034は、第1遅延部1032と同様に、ビタビ復号部1038における仮判定のための処理遅延に応じた期間にわたって遅延を実行する。 The first delay unit 1032 receives the linear equalization signal from the linear equalization unit 1044. The first delay unit 1032 delays the linear equalization signal and then outputs the delayed linear equalization signal (hereinafter referred to as “linear equalization signal” or “delayed signal”) to the equalization error generation unit 1040. Here, the first delay unit 1032 executes a delay over a period corresponding to the processing delay for provisional determination in the Viterbi decoding unit 1038. That is, the first delay unit 1032 uses the provisional determination signal 1306 output from the Viterbi decoding unit 1038 after reaching the Viterbi decoding unit 1038 from the linear equalization unit 1044 and the linear equalization signal from the linear equalization unit 1044. Timing is adjusted. The first delay unit 1032 is constituted by, for example, a latch circuit driven by a bit clock. The second delay unit 1034 receives the nonlinear equalization signal from the nonlinear equalization unit 1046. The second delay unit 1034 delays the nonlinear equalization signal, and then outputs the delayed nonlinear equalization signal (hereinafter referred to as “nonlinear equalization signal” or “delayed signal”) to the equalization error generation unit 1040. Here, like the first delay unit 1032, the second delay unit 1034 executes a delay over a period corresponding to the processing delay for provisional determination in the Viterbi decoding unit 1038.
 加算部1036は、線形等化部1044からの線形等化信号と、非線形等化部1046からの非線形等化信号とを入力する。加算部1036は、線形等化信号と非線形等化信号とを加算することによって、加算信号を生成する。加算部1036は、加算信号をビタビ復号部1038へ出力する。ビタビ復号部1038は、加算部1036からの加算信号を入力し、加算信号に対してビタビ復号を実行する。ビタビ復号部1038は、加算信号からブランチメトリックを計算するブランチメトリック演算回路と、ブランチメトリックを1クロック毎に累積加算してパスメトリックを計算するパスメトリック演算回路と、パスメトリックが最小となるデータ系列を最も確からしい候補系列として選択して記憶するパスメモリとを含む。パスメモリは、複数の候補系列を格納しており、パスメトリック演算回路からの選択信号にしたがって候補系列を選択する。また、選択された候補系列がデータ系列として出力される。 The addition unit 1036 receives the linear equalization signal from the linear equalization unit 1044 and the non-linear equalization signal from the non-linear equalization unit 1046. The adder 1036 generates an addition signal by adding the linear equalization signal and the nonlinear equalization signal. Adder 1036 outputs the addition signal to Viterbi decoder 1038. The Viterbi decoding unit 1038 receives the addition signal from the addition unit 1036 and performs Viterbi decoding on the addition signal. The Viterbi decoding unit 1038 includes a branch metric calculation circuit that calculates a branch metric from the addition signal, a path metric calculation circuit that calculates a path metric by accumulating the branch metrics for each clock, and a data sequence that minimizes the path metric And a path memory that selects and stores them as the most probable candidate series. The path memory stores a plurality of candidate sequences, and selects candidate sequences according to a selection signal from the path metric calculation circuit. In addition, the selected candidate series is output as a data series.
 また、ビタビ復号部1038は、パスメモリに記憶されているデータ系列に対して、パーシャルレスポンス規則にしたがって仮判定を実行することによって、加算信号を順次仮判定する。つまり、ビタビ復号部1038は、パスメモリに記憶されている候補系列のうちのひとつに対して、所定ビット数を使ってパーシャルレスポンスの仮判定動作を実行する。具体的に説明すると、ビタビ復号部1038は、パーシャルレスポンス等化が正常になされた場合に、所定の入力ビットに対する出力のレベルを仮判定し、入力ビットに対して仮判定したレベルを仮判定信号1306として等化誤差生成部1040へ出力する。仮判定はパスメモリの最終結果に限らず、パスメモリの途中の候補系列に対してなされてもよい。例えば、パスメモリ長が64ビットだとした場合、24ビット目や32ビット目の候補系列のひとつに対して仮判定がなされてもよい。 Further, the Viterbi decoding unit 1038 sequentially provisionally determines the addition signal by executing provisional determination on the data series stored in the path memory according to the partial response rule. That is, the Viterbi decoding unit 1038 performs a partial response provisional determination operation using a predetermined number of bits for one of the candidate sequences stored in the path memory. Specifically, the Viterbi decoding unit 1038 tentatively determines the output level for a predetermined input bit when the partial response equalization is normally performed, and the tentative determination signal indicates the level temporarily determined for the input bit. The result is output to the equalization error generation unit 1040 as 1306. The provisional determination is not limited to the final result of the path memory, but may be made for a candidate series in the middle of the path memory. For example, if the path memory length is 64 bits, provisional determination may be made for one of the 24th and 32nd bit candidate series.
 等化誤差生成部1040は、第1遅延部1032からの線形等化信号、第2遅延部1034からの非線形等化信号、ビタビ復号部1038からの仮判定信号1306を入力する。前述のごとく、これらの信号のタイミングは合っている。等化誤差生成部1040は、線形等化誤差と非線形等化信号との和と、仮判定信号1306との差異をもとに、線形等化用誤差信号1300を生成する。例えば、線形等化誤差と非線形等化信号との和が計算された後、和から仮判定信号1306を減算することによって、線形等化用誤差信号1300が導出される。また、等化誤差生成部1040は、線形等化誤差と非線形等化信号との和と、仮判定信号1306との差異をもとに、非線形等化用誤差信号1302を生成する。例えば、線形等化誤差と非線形等化信号との和が計算された後、和から仮判定信号1306を減算することによって、非線形等化用誤差信号1302が導出される。等化誤差生成部1040は、線形等化用誤差信号1300を線形等化部1044へ出力し、非線形等化用誤差信号1302を非線形等化部1046へ出力する。 The equalization error generation unit 1040 receives the linear equalization signal from the first delay unit 1032, the nonlinear equalization signal from the second delay unit 1034, and the provisional determination signal 1306 from the Viterbi decoding unit 1038. As mentioned above, the timing of these signals is correct. The equalization error generation unit 1040 generates a linear equalization error signal 1300 based on the difference between the sum of the linear equalization error and the nonlinear equalization signal and the provisional determination signal 1306. For example, after the sum of the linear equalization error and the non-linear equalization signal is calculated, the provisional determination signal 1306 is subtracted from the sum to derive the linear equalization error signal 1300. Further, the equalization error generation unit 1040 generates a non-linear equalization error signal 1302 based on the difference between the sum of the linear equalization error and the non-linear equalization signal and the provisional determination signal 1306. For example, after the sum of the linear equalization error and the nonlinear equalization signal is calculated, the provisional determination signal 1306 is subtracted from the sum to derive the nonlinear equalization error signal 1302. The equalization error generation unit 1040 outputs the linear equalization error signal 1300 to the linear equalization unit 1044, and outputs the nonlinear equalization error signal 1302 to the nonlinear equalization unit 1046.
 非線形等化部1046は、非線形等化用誤差信号1302をもとに、ニューラルネットワークのタップ係数を更新しているが、タップ係数が収束するまでの間、非線形等化部1046の動作が不安定になる。その結果、ビタビ復号部1038から出力されるデータ系列が誤る可能性が高くなる。そのため、収束前から、加算信号をビタビ復号部1038へ入力することは、好ましくない。これに対応するために、判定部1042は、非線形等化部1046における複数のタップ係数の収束を判定する。具体的に説明すると、判定部1042は、非線形等化用誤差信号1302の二乗値を所定期間にわたって加算することにより積算する。また、判定部1042は、積算値としきい値とを比較し、積算値がしきい値以下になれば収束と判定する。一方、積算値がしきい値より大きければ、判定部1042は、収束していないと判定する。判定部1042は、加算部1036、等化誤差生成部1040へ判定結果を出力する。判定結果では、収束しているか否かが示されている。 The nonlinear equalizer 1046 updates the tap coefficients of the neural network based on the error signal 1302 for nonlinear equalization, but the operation of the nonlinear equalizer 1046 is unstable until the tap coefficients converge. become. As a result, there is a high possibility that the data series output from the Viterbi decoding unit 1038 is erroneous. Therefore, it is not preferable to input the addition signal to the Viterbi decoding unit 1038 before convergence. In order to cope with this, the determination unit 1042 determines convergence of a plurality of tap coefficients in the nonlinear equalization unit 1046. More specifically, the determination unit 1042 adds up the square values of the nonlinear equalization error signal 1302 by adding them over a predetermined period. In addition, the determination unit 1042 compares the integrated value with a threshold value, and determines that the integrated value is converged if the integrated value is equal to or less than the threshold value. On the other hand, if the integrated value is larger than the threshold value, the determination unit 1042 determines that the convergence has not occurred. The determination unit 1042 outputs the determination result to the addition unit 1036 and the equalization error generation unit 1040. The determination result indicates whether or not it has converged.
 加算部1036は、判定部1042において収束が判定されるまでの間、つまり収束していないと判定された場合、前述の加算信号をビタビ復号部1038へ出力せずに、線形等化部1044からの線形等化信号をビタビ復号部1038へ出力する。一方、加算部1036は、判定部1042において収束が判定された後、前述のごとく、加算信号をビタビ復号部1038へ出力する。つまり、ニューラルネットが収束するまでの間は、線形等化信号がビタビ復号部1038へ出力され、ニューラルネットが収束してからは、加算信号がビタビ復号部1038へ出力される。これにより、非線形等化信号による悪影響が低減される。 The adding unit 1036 does not output the above addition signal to the Viterbi decoding unit 1038 until the determination unit 1042 determines convergence, that is, when it is determined that convergence has not occurred, the linear equalization unit 1044 The linear equalized signal is output to the Viterbi decoding unit 1038. On the other hand, the adder 1036 outputs the added signal to the Viterbi decoder 1038 as described above after the convergence is determined by the determination unit 1042. In other words, the linear equalized signal is output to the Viterbi decoding unit 1038 until the neural network converges, and the added signal is output to the Viterbi decoding unit 1038 after the neural network converges. As a result, adverse effects due to the nonlinear equalized signal are reduced.
 等化誤差生成部1040は、判定部1042において収束が判定されるまでの間、非線形等化信号を使用せず、線形等化信号と仮判定信号1306との差異をもとに、線形等化用誤差信号1300を生成する。また、等化誤差生成部1040は、線形等化部1044に対して、線形等化用誤差信号1300をもとに複数のタップ係数を導出させる。一方、等化誤差生成部1040は、判定部1042において収束が判定された後、前述のごとく動作する。さらに、等化誤差生成部1040は、判定部1042からの判定結果に関係なく、線形等化信号と非線形等化信号との和と、仮判定信号1306との差異をもとに、非線形等化用誤差信号1302を生成する。これによっても、非線形等化信号による悪影響が低減される。また、判定部1042は、収束を判定した後、非線形等化用誤差信号1302の二乗値の総和である積算値がしきい値よりも再び大きくなった場合に、非線形等化部1046の複数のタップ係数の発散と判定する。その際、判定部1042は、非線形等化部1046に対して、複数のタップ係数を新たに導出させる。なお、非線形等化部1046は、所定の収束値以下となった場合には複数のタップ係数の更新を停止するだけでもよい。 The equalization error generation unit 1040 does not use the nonlinear equalization signal until the convergence is determined by the determination unit 1042, and performs linear equalization based on the difference between the linear equalization signal and the provisional determination signal 1306. An error signal 1300 is generated. Also, the equalization error generation unit 1040 causes the linear equalization unit 1044 to derive a plurality of tap coefficients based on the linear equalization error signal 1300. On the other hand, the equalization error generation unit 1040 operates as described above after the determination unit 1042 determines convergence. Further, the equalization error generation unit 1040 performs non-linear equalization based on the difference between the sum of the linear equalization signal and the non-linear equalization signal and the provisional determination signal 1306 regardless of the determination result from the determination unit 1042. Error signal 1302 is generated. This also reduces the adverse effects caused by the non-linear equalization signal. Further, after determining the convergence, the determination unit 1042 determines that the integrated value that is the sum of the square values of the nonlinear equalization error signal 1302 becomes larger than the threshold value again. It is determined that the tap coefficient is divergent. At this time, the determination unit 1042 causes the nonlinear equalization unit 1046 to newly derive a plurality of tap coefficients. Note that the non-linear equalization unit 1046 may only stop updating the plurality of tap coefficients when the value becomes equal to or less than the predetermined convergence value.
 なお、ニューラルネットワークが最初から収束値に近いタップ係数で設定できれば当初から線形等化信号と非線形等化信号との加算信号をビタビ復号部1038へ出力しても構わないし、非線形等化用誤差信号1302の収束をしきい値以下となった時に収束したと判定するのではなく、所定の時間経過をもって収束していると判断しても構わない。 If the neural network can be set with a tap coefficient close to the convergence value from the beginning, an addition signal of the linear equalization signal and the non-linear equalization signal may be output to the Viterbi decoding unit 1038 from the beginning, or the non-linear equalization error signal may be output. Rather than determining that the convergence of 1302 is equal to or less than the threshold value, it may be determined that the convergence has occurred after a predetermined time.
 図14は、線形等化部1044の構成を示す。線形等化部1044は、多段タップ1050、線形処理部1052を含む。多段タップ1050は、遅延タップ1054と総称される第1遅延タップ1054a、第2遅延タップ1054b、第3遅延タップ1054c、第N遅延タップ1054nを含む。線形処理部1052は、乗算部1056と総称される第1乗算部1056a、第2乗算部1056b、第3乗算部1056c、第N+1乗算部1056n+1、タップ係数導出部1058、積算部1060を含む。 FIG. 14 shows the configuration of the linear equalization unit 1044. The linear equalization unit 1044 includes a multistage tap 1050 and a linear processing unit 1052. The multistage tap 1050 includes a first delay tap 1054a, a second delay tap 1054b, a third delay tap 1054c, and an Nth delay tap 1054n, which are collectively referred to as a delay tap 1054. The linear processing unit 1052 includes a first multiplication unit 1056a, a second multiplication unit 1056b, a third multiplication unit 1056c, an (N + 1) th multiplication unit 1056n + 1, a tap coefficient derivation unit 1058, and an accumulation unit 1060, which are collectively referred to as a multiplication unit 1056.
 多段タップ1050は、複数の遅延タップ1054がシリアルに接続されることによって形成される。具体的に説明すると、第1遅延タップ1054aは、再生信号を入力し、遅延後、再生信号を出力する。第2遅延タップ1054bは、第1遅延タップ1054aからの再生信号を入力し、遅延後、再生信号を出力する。第3遅延タップ1054cから第N遅延タップ1054nも、同様の処理を実行する。遅延タップ1054への入力部分と出力部分が多段タップ1050からの出力信号であり、例えば、4つの遅延タップ1054が配置される場合、5つの出力信号が存在する。これらの出力信号は、乗算部1056へ出力されている。 The multi-stage tap 1050 is formed by connecting a plurality of delay taps 1054 serially. More specifically, the first delay tap 1054a inputs the reproduction signal, and outputs the reproduction signal after delay. The second delay tap 1054b receives the reproduction signal from the first delay tap 1054a, and outputs the reproduction signal after delay. The third delay tap 1054c to the Nth delay tap 1054n perform the same processing. An input portion and an output portion to the delay tap 1054 are output signals from the multistage tap 1050. For example, when four delay taps 1054 are arranged, there are five output signals. These output signals are output to the multiplication unit 1056.
 乗算部1056は、遅延タップ1054からの出力信号を入力するとともに、タップ係数導出部1058からのタップ係数も入力する。ここで、タップ係数は、各出力信号に対応づけられて導出されている。乗算部1056は、出力信号とタップ係数とを乗算する。乗算部1056は、各乗算結果を積算部1060へ出力する。積算部1060は、乗算部1056からの乗算結果を次々に加算して加算結果である積算値を求める。加算結果である積算値が、前述の線形等化信号に相当する。積算部1060は、線形等化信号を出力する。タップ係数導出部1058は、線形等化用誤差信号1300を入力する。タップ係数導出部1058は、再生信号がパーシャルレスポンス特性に適合するように、線形等化用誤差信号1300、乗算部1056での乗算結果を使用して、複数のタップ係数を制御する。なお、タップ係数の導出には、例えばLMSアルゴリズムのような適応アルゴリズムが使用されることによって、線形等化用誤差信号1300が小さくなるように制御される。なお、LMSアルゴリズムは、公知の技術であるので、ここでは説明を省略する。 Multiplier 1056 receives the output signal from delay tap 1054 and also receives the tap coefficient from tap coefficient deriving section 1058. Here, the tap coefficient is derived in association with each output signal. Multiplier 1056 multiplies the output signal and the tap coefficient. Multiplication unit 1056 outputs each multiplication result to integration unit 1060. The accumulating unit 1060 adds the multiplication results from the multiplying unit 1056 one after another to obtain an accumulated value as an addition result. The integrated value as the addition result corresponds to the above-described linear equalization signal. The integrating unit 1060 outputs a linear equalization signal. The tap coefficient deriving unit 1058 receives the linear equalization error signal 1300. The tap coefficient deriving unit 1058 controls the plurality of tap coefficients using the linear equalization error signal 1300 and the multiplication result of the multiplication unit 1056 so that the reproduction signal matches the partial response characteristic. The tap coefficient is derived by using an adaptive algorithm such as the LMS algorithm so that the linear equalization error signal 1300 is controlled to be small. Since the LMS algorithm is a known technique, the description thereof is omitted here.
 図15は、非線形等化部1046の構成を示す。非線形等化部1046は、多段タップ1070、非線形処理部1072を含む。多段タップ1070は、遅延タップ1074と総称される第1遅延タップ1074a、第2遅延タップ1074b、第N遅延タップ1074nを含む。非線形処理部1072は、乗算部1076と総称される第11乗算部1076aa、第12乗算部1076ab、第1M乗算部1076am、第21乗算部1076ba、第22乗算部1076bb、第2M乗算部1076bm、第(N+1)1乗算部1076(n+1)a、第(N+1)2乗算部1076(n+1)b、第(N+1)M乗算部1076(n+1)m、積算部1078と総称される第1積算部1078a、第2積算部1078b、第M積算部1078m、関数演算部1080と総称される第1関数演算部1080a、第2関数演算部1080b、第M関数演算部1080m、乗算部1082と総称される第1乗算部1082a、第2乗算部1082b、第M乗算部1082m、積算部1084、関数演算部1086、タップ係数導出部1088を含む。 FIG. 15 shows the configuration of the nonlinear equalization unit 1046. The non-linear equalization unit 1046 includes a multistage tap 1070 and a non-linear processing unit 1072. The multistage tap 1070 includes a first delay tap 1074a, a second delay tap 1074b, and an Nth delay tap 1074n, which are collectively referred to as a delay tap 1074. The nonlinear processing unit 1072 includes an eleventh multiplication unit 1076aa, a twelfth multiplication unit 1076ab, a first M multiplication unit 1076am, a twenty-first multiplication unit 1076ba, a twenty-second multiplication unit 1076bb, a second M multiplication unit 1076bm, (N + 1) 1 multiplier 1076 (n + 1) a, (N + 1) 2 multiplier 1076 (n + 1) b, (N + 1) M multiplier 1076 (n + 1) m, and first integrator 1078a collectively referred to as integrator 1078 , A second summation unit 1078b, an Mth summation unit 1078m, a first function computation unit 1080a collectively referred to as a function computation unit 1080, a second function computation unit 1080b, an Mth function computation unit 1080m, and a multiplication unit 1082. 1 multiplier 1082a, 2nd multiplier 1082b, Mth multiplier 1082m, integrator 1084, function calculator 1086, tap Including the number derivation unit 1088.
 非線形等化部1046は、図示のごとく、3層パーセプトロン型のニューラルネットワークにて構成される。ここで、入力層が多段タップ1070に相当し、隠れ層が関数演算部1080に相当し、出力層が関数演算部1086に相当する。多段タップ1070は、複数の遅延タップ1074がシリアルに接続されることによって形成される。具体的に説明すると、第1遅延タップ1074aは、再生信号を入力し、遅延後、再生信号を出力する。第2遅延タップ1074bは、第1遅延タップ1074aからの再生信号を入力し、遅延後、再生信号を出力する。第N遅延タップ1074nも、同様の処理を実行する。遅延タップ1074への入力部分と出力部分が多段タップ1070からの出力信号である。これらの出力信号は、乗算部1076へ出力されている。 The non-linear equalization unit 1046 is configured by a three-layer perceptron type neural network as illustrated. Here, the input layer corresponds to the multistage tap 1070, the hidden layer corresponds to the function calculation unit 1080, and the output layer corresponds to the function calculation unit 1086. The multistage tap 1070 is formed by serially connecting a plurality of delay taps 1074. More specifically, the first delay tap 1074a inputs the reproduction signal, and outputs the reproduction signal after delay. The second delay tap 1074b receives the reproduction signal from the first delay tap 1074a, and outputs the reproduction signal after delay. The Nth delay tap 1074n performs the same process. An input part and an output part to the delay tap 1074 are output signals from the multistage tap 1070. These output signals are output to the multiplier 1076.
 乗算部1076は、多段タップ1070からの出力信号と、タップ係数導出部1088からのタップ係数とを乗算する。具体的に説明すると、第IJ乗算部1076ijは、多段タップ1070の先頭からi番目の出力信号S(i)と、タップ係数W1(i,j)とを乗算することによって、乗算結果U(i,j)を生成する。積算部1078は、乗算部1076における乗算結果を次々に加算する積算を行う。具体的に説明すると、第J積算部1078jは、乗算結果U(1,j)、U(2,j)、U(3,j)、・・・、U(n+1,j)を加算して積算することによって、積算結果V(j)を生成する。関数演算部1080は、積算部1078における積算結果V(j)にシグモイド関数を演算する。シグモイド関数は、次のように示される。
    f(x)=(1-exp(-αx))/(1+exp(-αx))…(式7)
 ここで、式7のxに積算結果V(j)が入力される。ここでは、第J関数演算部1080jでの演算結果をX(j)と示し、当該演算結果が隠れ層からの出力に相当する。
Multiplier 1076 multiplies the output signal from multistage tap 1070 and the tap coefficient from tap coefficient deriving section 1088. Specifically, the IJ multiplication unit 1076ij multiplies the i-th output signal S (i) from the top of the multistage tap 1070 by the tap coefficient W1 (i, j), thereby obtaining a multiplication result U (i , J). The accumulating unit 1078 performs an accumulation by sequentially adding the multiplication results in the multiplying unit 1076. More specifically, the J-th integrating unit 1078j adds the multiplication results U (1, j), U (2, j), U (3, j),..., U (n + 1, j). By integrating, an integration result V (j) is generated. The function calculation unit 1080 calculates a sigmoid function on the integration result V (j) in the integration unit 1078. The sigmoid function is shown as follows.
f (x) = (1−exp (−αx)) / (1 + exp (−αx)) (Expression 7)
Here, the integration result V (j) is input to x in Expression 7. Here, the calculation result in the J-th function calculation unit 1080j is denoted as X (j), and the calculation result corresponds to the output from the hidden layer.
 乗算部1082は、関数演算部1080における演算結果とタップ係数導出部1088からのタップ係数とを乗算する。具体的に説明すると、第J乗算部1082jは、第J関数演算部1080jにおける演算結果X(j)と、タップ係数W2(j)とを乗算することによって、乗算結果Y(j)を生成する。積算部1084は、乗算部1082における乗算結果を次々に加算する積算を行う。ここでは、すべての乗算部1082における乗算結果が加算されて積算され、積算結果Zが生成される。関数演算部1086は、積算部1084における積算結果にシグモイド関数を演算する。ここでは、式7のxに積算結果Zが入力される。関数演算部1086の演算結果が、出力層からの出力に相当し、前述の非線形等化信号に相当する。 The multiplication unit 1082 multiplies the calculation result in the function calculation unit 1080 and the tap coefficient from the tap coefficient derivation unit 1088. Specifically, the J-th multiplication unit 1082j generates a multiplication result Y (j) by multiplying the calculation result X (j) in the J-th function calculation unit 1080j by the tap coefficient W2 (j). . The accumulating unit 1084 performs accumulation in which the multiplication results in the multiplying unit 1082 are added one after another. Here, the multiplication results in all the multiplication units 1082 are added and integrated, and an integration result Z is generated. The function calculation unit 1086 calculates a sigmoid function on the integration result in the integration unit 1084. Here, the integration result Z is input to x in Expression 7. The calculation result of the function calculation unit 1086 corresponds to the output from the output layer, and corresponds to the above-described nonlinear equalization signal.
 タップ係数導出部1088は、乗算部1076および乗算部1082において使用されるタップ係数W1(i,j)とW2(j)とを導出する。なお、W1(i,j)、W2(j)の初期値として、ランダムな値や収束後に近い値が設定される。また、タップ係数導出部1088は、図14のタップ係数導出部1058と同様にLMSアルゴリズムによって、W1(i,j)、W2(j)を更新する。ここで、W1(i,j)、W2(j)の学習は、バックプロパゲーションによってなされる。非線形等化用誤差信号1302の二乗値は、次のように示される。
    E=(A-D)…(式8)
Tap coefficient derivation unit 1088 derives tap coefficients W1 (i, j) and W2 (j) used in multiplication unit 1076 and multiplication unit 1082. A random value or a value close to that after convergence is set as the initial value of W1 (i, j) and W2 (j). Further, the tap coefficient deriving unit 1088 updates W1 (i, j) and W2 (j) by the LMS algorithm similarly to the tap coefficient deriving unit 1058 of FIG. Here, learning of W1 (i, j) and W2 (j) is performed by back propagation. The square value of the non-linear equalization error signal 1302 is expressed as follows.
E = (AD) 2 (Formula 8)
 ここで、Aは、線形等化信号と非線形等化信号との和に相当し、Dは、仮判定信号1306に相当する。つまり、A-Dは、非線形等化用誤差信号1302に相当する。タップ係数導出部1088は、Eが最小となるように、W1(i,j)、W2(j)を制御する。出力層でのバックプロパゲーションの結果は次のように示される。
    (∂E)/(∂Y(j))=f’(Y(j))×2(A-D)…(式9)
 タップ係数導出部1088は、タップ係数W2(j)を次のように更新する。
    W2(j)=W2(j)old-ε×(∂E)/(∂W2(j))…(式10)
Here, A corresponds to the sum of the linear equalization signal and the nonlinear equalization signal, and D corresponds to the provisional determination signal 1306. That is, AD corresponds to the error signal 1302 for nonlinear equalization. The tap coefficient deriving unit 1088 controls W1 (i, j) and W2 (j) so that E is minimized. The result of back propagation at the output layer is shown as follows.
(∂E) / (∂Y (j)) = f ′ (Y (j)) × 2 (AD) (Equation 9)
The tap coefficient deriving unit 1088 updates the tap coefficient W2 (j) as follows.
W2 (j) = W2 (j) old −ε × (∂E) / (∂W2 (j)) (Equation 10)
 ここで、W2(j)oldは、ひとつ前のタイミングにおけるタップ係数W2(j)を示す。一方、隠れ層でのバックプロパゲーションは次のように示される。
    (∂E)/(∂U(i,j))=
      f’(U(i,j))×(∂E)/(∂Y(j))×W2(j)…(式11)
 タップ係数導出部1088は、タップ係数W1(i,j)を次のように更新する。
    W1(i,j)=
      W1(i,j)old-ε×(∂E)/(∂W1(i,j))…(式12)
 ここで、W1(i,j)oldは、ひとつ前のタイミングにおけるタップ係数W1(i,j)を示す。
Here, W2 (j) old indicates the tap coefficient W2 (j) at the previous timing. On the other hand, back propagation in the hidden layer is shown as follows.
(∂E) / (∂U (i, j)) =
f ′ (U (i, j)) × (∂E) / (∂Y (j)) × W2 (j) (Equation 11)
The tap coefficient deriving unit 1088 updates the tap coefficient W1 (i, j) as follows.
W1 (i, j) =
W1 (i, j) old −ε × (∂E) / (∂W1 (i, j)) (Equation 12)
Here, W1 (i, j) old indicates the tap coefficient W1 (i, j) at the previous timing.
 図16は、ビタビ復号部1038の構成を示す。ビタビ復号部1038は、ブランチメトリック演算部1090、パスメモリ部1092、多数決部1094、特定部1096を含む。また、信号として、選択信号SEL、ビット信号1304を含む。ブランチメトリック演算部1090は、図示しない加算部1036からの線形等化信号あるいは加算信号(以下、「加算信号」と総称する)をもとに、ブランチメトリック演算およびパスメトリック演算を実行する。そのため、ブランチメトリック演算部1090には、前述のブランチメトリック演算回路およびパスメトリック演算回路が含まれる。前述のごとく、本実施例では、パーシャルレスポンス方式が適用されているが、ビタビ復号部1038の構成を説明する前に、ここでは、パーシャルレスポンス方式における状態遷移を説明する。 FIG. 16 shows the configuration of the Viterbi decoding unit 1038. The Viterbi decoding unit 1038 includes a branch metric calculation unit 1090, a path memory unit 1092, a majority decision unit 1094, and a specification unit 1096. In addition, the signal includes a selection signal SEL and a bit signal 1304. The branch metric calculation unit 1090 executes branch metric calculation and path metric calculation based on a linear equalized signal or an addition signal (hereinafter collectively referred to as “addition signal”) from an addition unit 1036 (not shown). Therefore, the branch metric calculation unit 1090 includes the aforementioned branch metric calculation circuit and path metric calculation circuit. As described above, the partial response method is applied in the present embodiment, but before the configuration of the Viterbi decoding unit 1038 is described, state transition in the partial response method will be described here.
 図17は、ビタビ復号部1038がパーシャルレスポンス(1,2,2,2,1)に対応する場合の状態遷移を示す。パーシャルレスポンス(1,2,2,2,1)では、振幅が±4の範囲に収まる。4ビットをひとつの組合せとすれば、組合せに含まれる値に応じて、S0からS9までの10状態が規定されている。また、次に入力されるビット値に応じて図示のごとく、状態が遷移する。例えば、状態S0にビット値「1」が入力されると、状態S1への遷移がなされる。ここで、状態間を結ぶ矢印に「x/y」のような値が示されているが、xは、入力されるビット値を示し、yは、もとの状態に新たなビット値が加わった5ビットに対する仮判定値を示す。図18は、ビタビ復号部1038がパーシャルレスポンス(1,2,2,2,1)に対応する場合の状態遷移を示す。図18は、連続したふたつのタイミングでの状態を示しており、各状態は、図17と同様である。 FIG. 17 shows a state transition when the Viterbi decoding unit 1038 corresponds to a partial response (1, 2, 2, 2, 1). In the partial response (1, 2, 2, 2, 1), the amplitude falls within a range of ± 4. If 4 bits are taken as one combination, 10 states from S0 to S9 are defined according to the values included in the combination. Further, the state transitions as shown in the figure according to the next input bit value. For example, when a bit value “1” is input to the state S0, a transition to the state S1 is made. Here, a value such as “x / y” is shown in the arrows connecting the states, where x indicates an input bit value and y indicates a new bit value added to the original state. The temporary decision value for 5 bits is shown. FIG. 18 shows a state transition when the Viterbi decoding unit 1038 corresponds to a partial response (1, 2, 2, 2, 1). FIG. 18 shows a state at two consecutive timings, and each state is the same as FIG.
 図19は、ブランチメトリック演算部1090の構成を示す。ブランチメトリック演算部1090は、加算部1110と総称される第1加算部1110a、第2加算部1110b、第3加算部1110c、第4加算部1110d、第5加算部1110e、第6加算部1110f、第7加算部1110g、第8加算部1110h、第9加算部1110i、第10加算部1110j、第11加算部1110k、第12加算部1110l、第13加算部1110m、第14加算部1110n、第15加算部1110o、第16加算部1110p、二乗回路1112と総称される第1二乗回路1112a、第2二乗回路1112b、第3二乗回路1112c、第4二乗回路1112d、第5二乗回路1112e、第6二乗回路1112f、第7二乗回路1112g、第8二乗回路1112h、第9二乗回路1112i、第10二乗回路1112j、第11二乗回路1112k、第12二乗回路1112l、第13二乗回路1112m、第14二乗回路1112n、第15二乗回路1112o、第16二乗回路1112p、ACS回路1114と総称される第1ACS回路1114a、第2ACS回路1114b、第3ACS回路1114c、第4ACS回路1114d、第5ACS回路1114e、第6ACS回路1114f、加算部1116と総称される第1加算部1116a、第2加算部1116b、第3加算部1116c、第4加算部1116dを含む。また、選択信号SELと総称される第0選択信号SEL0、第1選択信号SEL1、第2選択信号SEL2、第7選択信号SEL7、第8選択信号SEL8、第9選択信号SEL9を含む。 FIG. 19 shows the configuration of the branch metric calculation unit 1090. The branch metric calculation unit 1090 includes a first addition unit 1110a, a second addition unit 1110b, a third addition unit 1110c, a fourth addition unit 1110d, a fifth addition unit 1110e, a sixth addition unit 1110f, which are collectively referred to as an addition unit 1110. 7th addition part 1110g, 8th addition part 1110h, 9th addition part 1110i, 10th addition part 1110j, 11th addition part 1110k, 12th addition part 1110l, 13th addition part 1110m, 14th addition part 1110n, 15th Adder 1110o, 16th adder 1110p, first square circuit 1112a, second square circuit 1112b, third square circuit 1112c, fourth square circuit 1112d, fifth square circuit 1112e, sixth square Circuit 1112f, seventh square circuit 1112g, eighth square circuit 1112h, ninth square circuit 111 i, 10th square circuit 1112j, 11th square circuit 1112k, 12th square circuit 1112l, 13th square circuit 1112m, 14th square circuit 1112n, 15th square circuit 1112o, 16th square circuit 1112p, ACS circuit 1114 First ACS circuit 1114a, second ACS circuit 1114b, third ACS circuit 1114c, fourth ACS circuit 1114d, fifth ACS circuit 1114e, sixth ACS circuit 1114f, first adder 1116a, second adder 1116b, which are collectively referred to as adder 1116, A third adder 1116c and a fourth adder 1116d are included. Further, it includes a 0th selection signal SEL0, a first selection signal SEL1, a second selection signal SEL2, a seventh selection signal SEL7, an eighth selection signal SEL8, and a ninth selection signal SEL9, which are collectively referred to as a selection signal SEL.
 加算部1110は、加算信号から所定の目標値を減じる。二乗回路1112は、加算部1110における減算結果の二乗値を計算する。ACS回路1114は、二乗回路1112からの二乗に対して、加算、比較、選択によるメトリック演算を実行する。また、ACS回路1114は、メトリック演算の結果として、第0選択信号SEL0、第1選択信号SEL1、第2選択信号SEL2、第7選択信号SEL7、第8選択信号SEL8、第9選択信号SEL9を出力する。また、パーシャルレスポンス特性からACS回路1114へ入力されない二乗値も存在する。そのような二乗値に対して、加算部1116において加算がなされる。 The addition unit 1110 subtracts a predetermined target value from the addition signal. The square circuit 1112 calculates the square value of the subtraction result in the adder 1110. The ACS circuit 1114 performs a metric operation by addition, comparison, and selection on the square from the square circuit 1112. Further, the ACS circuit 1114 outputs the 0th selection signal SEL0, the first selection signal SEL1, the second selection signal SEL2, the seventh selection signal SEL7, the eighth selection signal SEL8, and the ninth selection signal SEL9 as the result of the metric calculation. To do. There is also a square value that is not input to the ACS circuit 1114 due to the partial response characteristic. An addition unit 1116 adds the square value.
 図20は、パスメモリ部1092の構成を示す。パスメモリ部1092は、ブランチメトリック演算部1090からの選択信号SELを入力し、選択信号SELに応じたパスを記憶する。図20は、パスメモリ部1092の構成を示す。パスメモリ部1092は、メモリ1120と総称される第11メモリ1120aa、第12メモリ1120ab、第13メモリ1120ac、第14メモリ1120ad、第15メモリ1120ae、第16メモリ1120af、第17メモリ1120ag、第18メモリ1120ah、第19メモリ1120ai、第110メモリ1120aj、第21メモリ1120ba、第22メモリ1120bb、第23メモリ1120bc、第24メモリ1120bd、第25メモリ1120be、第26メモリ1120bf、第27メモリ1120bg、第28メモリ1120bh、第29メモリ1120bi、第210メモリ1120bj、第(L+1)1メモリ1120(l+1)a、第(L+1)2メモリ1120(l+1)b、第(L+1)3メモリ1120(l+1)c、第(L+1)4メモリ1120(l+1)d、第(L+1)5メモリ1120(l+1)e、第(L+1)6メモリ1120(l+1)f、第(L+1)7メモリ1120(l+1)g、第(L+1)8メモリ1120(l+1)h、第(L+1)9メモリ1120(l+1)i、第(L+1)10メモリ1120(l+1)j、選択部1122と総称される第11選択部1122aa、第12選択部1122ab、第13選択部1122ac、第14選択部1122ad、第15選択部1122ae、第16選択部1122af、第L1選択部1122la、第L2選択部1122lb、第L3選択部1122lc、第L4選択部1122ld、第L5選択部1122le、第6L6選択部1122lf、多数決部1124を含む。 FIG. 20 shows the configuration of the path memory unit 1092. The path memory unit 1092 receives the selection signal SEL from the branch metric calculation unit 1090 and stores a path corresponding to the selection signal SEL. FIG. 20 shows the configuration of the path memory unit 1092. The path memory unit 1092 includes an eleventh memory 1120aa, a twelfth memory 1120ab, a thirteenth memory 1120ac, a fourteenth memory 1120ad, a fifteenth memory 1120ae, a sixteenth memory 1120af, a seventeenth memory 1120ag, and an eighteenth memory. 1120ah, 19th memory 1120ai, 110th memory 1120aj, 21st memory 1120ba, 22nd memory 1120bb, 23rd memory 1120bc, 24th memory 1120bd, 25th memory 1120be, 26th memory 1120bf, 27th memory 1120bg, 28th memory 1120bh, 29th memory 1120bi, 210th memory 1120bj, (L + 1) 1 memory 1120 (l + 1) a, (L + 1) 2 memory 1120 (l + 1) b, (L +) ) 3 memory 1120 (l + 1) c, (L + 1) 4 memory 1120 (l + 1) d, (L + 1) 5 memory 1120 (l + 1) e, (L + 1) 6 memory 1120 (l + 1) f, (L + 1) 7 The memory 1120 (l + 1) g, the (L + 1) 8th memory 1120 (l + 1) h, the (L + 1) 9th memory 1120 (l + 1) i, the (L + 1) 10th memory 1120 (l + 1) j, and the selector 1122 11th selection unit 1122aa, 12th selection unit 1122ab, 13th selection unit 1122ac, 14th selection unit 1122ad, 15th selection unit 1122ae, 16th selection unit 1122af, L1 selection unit 1122la, L2 selection unit 1122lb, L3 Selector 1122lc, L4th selector 1122ld, L5th selector 1122le, 6thL6 selector 1122l , Including a majority unit 1124.
 ここでは、L+1のメモリ1120によってひとつのパスが記憶され、かつ図17、18に示した10種類の状態のそれぞれに対応するように、10種類のパスが記憶される。選択部1122は、選択信号SELに応じて、いずれかのパスを選択する。選択されたパスが、生き残りパス相当する。多数決部1124は、第(L+1)1メモリ1120(l+1)aから第(L+1)10メモリ1120(l+1)jのそれぞれに記憶されたビット値を入力し、多数決を実行する。多数決によって選択されたビット値が、復号結果に相当する。多数決部1124は、復号結果を出力する。なお、パスの途中のメモリ1120に記憶されたビット値がビット信号1304と出力される。ビット信号1304には、10種類のパスのうち、同一のタイミングに対応した10のビット値が含まれる。図16に戻る。 Here, one path is stored in the L + 1 memory 1120, and ten types of paths are stored so as to correspond to each of the ten types shown in FIGS. The selection unit 1122 selects one of the paths according to the selection signal SEL. The selected path corresponds to the survival path. The majority decision unit 1124 inputs bit values stored in the (L + 1) th memory 1120 (l + 1) a to the (L + 1) th memory 1120 (l + 1) j, and executes majority decision. The bit value selected by the majority vote corresponds to the decoding result. The majority decision unit 1124 outputs the decryption result. Note that the bit value stored in the memory 1120 during the pass is output as a bit signal 1304. The bit signal 1304 includes 10 bit values corresponding to the same timing among the 10 types of paths. Returning to FIG.
 多数決部1094は、ビット信号1304を入力し、ビット信号1304に含まれた10のビット値に対して多数決を実行する。多数決部1094は、多数決によって選択したビット値(以下、「選択値」という)を特定部1096へ出力する。特定部1096は、多数決部1094からの選択値を入力し、ラッチにて選択値を保持する。ここで、特定部1096は、過去の選択値を含めて、5つのタイミングに対応した選択値からひとつの組合せを選択する。なお、特定部1096に新たな選択値が入力されると、組合せの中から最も過去の選択値が除外されることによって、組合せが更新される。 The majority decision unit 1094 receives the bit signal 1304 and executes a majority decision on the 10 bit values included in the bit signal 1304. The majority decision unit 1094 outputs the bit value selected by the majority decision (hereinafter referred to as “selected value”) to the identification unit 1096. The specifying unit 1096 receives the selection value from the majority decision unit 1094 and holds the selection value in a latch. Here, the specifying unit 1096 selects one combination from selection values corresponding to five timings including past selection values. When a new selection value is input to the specifying unit 1096, the combination is updated by removing the oldest selection value from the combination.
 図21は、特定部1096に記憶されたテーブルのデータ構造を示す。図示のごとく、メモリ値欄1200、b(k)欄1202、b(k-1)欄1204、b(k-2)欄1206、b(k-3)欄1208、b(k-4)欄1210、仮判定出力欄1212が含まれる。ここで、b(k)は、最も新しく入力された選択値に相当し、b(k-1)は、ひとつ前のタイミングに入力された選択値に相当し、b(k-4)は、4つ前のタイミングに入力された選択値に相当する。前述のごとく、これらはラッチにて保持されている。b(k)欄1202からb(k-4)欄1210には、ラッチに保持された選択値が取り得る値の組合せが示されている。メモリ値欄1200では、取り得る値に対応したメモリ値が示され、仮判定出力欄1212では、取り得る値に対応した仮判定値が示されている。例えば、パスメモリの内容が「00000」であれば仮判定値「-4」、「00001」であれば仮判定値「-3」が対応づけられている。図16に戻る。特定部1096は、図21に示したテーブルを参照しながら、組合せに対応した仮判定値を特定する。特定部1096は、仮判定値を仮判定信号1306として出力する。 FIG. 21 shows the data structure of the table stored in the specifying unit 1096. As shown, memory value column 1200, b (k) column 1202, b (k-1) column 1204, b (k-2) column 1206, b (k-3) column 1208, b (k-4) column. 1210 and a provisional determination output column 1212 are included. Here, b (k) corresponds to the most recently input selection value, b (k−1) corresponds to the selection value input at the previous timing, and b (k−4) This corresponds to the selection value input at the previous four timings. As described above, these are held by the latch. In the b (k) column 1202 to the b (k-4) column 1210, combinations of values that the selection value held in the latch can take are shown. The memory value column 1200 shows memory values corresponding to possible values, and the temporary determination output column 1212 shows temporary determination values corresponding to possible values. For example, if the content of the path memory is “00000”, the temporary determination value “−4” is associated, and if “00001”, the temporary determination value “−3” is associated. Returning to FIG. The specifying unit 1096 specifies the provisional determination value corresponding to the combination while referring to the table shown in FIG. The identifying unit 1096 outputs the temporary determination value as the temporary determination signal 1306.
 以上の構成による再生装置100の動作を説明する。図22は、加算部1036における加算手順を示すフローチャートである。判定部1042は、非線形等化用誤差信号1302を入力し、非線形等化用誤差信号1302の大きさを導出する。非線形等化用誤差信号1302の大きさがしきい値以内に収束していなければ(S1010のN)、判定部1042は、加算部1036に線形等化信号を出力させる(S1012)。一方、非線形等化用誤差信号1302の大きさがしきい値以内に収束していれば(S1010のY)、判定部1042は、加算部1036に加算信号を出力させる(S1014)。 The operation of the playback apparatus 100 configured as above will be described. FIG. 22 is a flowchart showing the adding procedure in the adding unit 1036. The determination unit 1042 receives the nonlinear equalization error signal 1302 and derives the magnitude of the nonlinear equalization error signal 1302. If the magnitude of the non-linear equalization error signal 1302 does not converge within the threshold (N in S1010), the determination unit 1042 causes the addition unit 1036 to output a linear equalization signal (S1012). On the other hand, if the magnitude of the nonlinear equalization error signal 1302 has converged within the threshold value (Y in S1010), the determination unit 1042 causes the addition unit 1036 to output an addition signal (S1014).
 図23は、等化誤差生成部1040における生成手順を示すフローチャートである。判定部1042は、非線形等化用誤差信号1302を入力し、非線形等化用誤差信号1302の大きさを導出する。非線形等化用誤差信号1302の大きさがしきい値以内に収束していなければ(S20のN)、等化誤差生成部1040は、第1遅延部1032からの遅延信号と仮判定信号1306との差を線形等化用誤差信号1300として線形等化部1044へ出力する(S22)。また、等化誤差生成部1040は、第1遅延部1032からの遅延信号と第2遅延部1034からの遅延信号との和と、仮判定信号1306との差を非線形等化用誤差信号1302として非線形等化部1046へ出力する(S24)。一方、非線形等化用誤差信号1302の大きさがしきい値以内に収束していれば(S20のY)、等化誤差生成部1040は、第1遅延部1032からの遅延信号と第2遅延部1034からの遅延信号との和と、仮判定信号1306との差を線形等化用誤差信号1300として線形等化部1044へ出力する(S26)。また、等化誤差生成部1040は、第1遅延部1032からの遅延信号と第2遅延部1034からの遅延信号との和と、仮判定信号1306との差を非線形等化用誤差信号1302として非線形等化部1046へ出力する(S28)。 FIG. 23 is a flowchart showing a generation procedure in the equalization error generation unit 1040. The determination unit 1042 receives the nonlinear equalization error signal 1302 and derives the magnitude of the nonlinear equalization error signal 1302. If the magnitude of the non-linear equalization error signal 1302 does not converge within the threshold value (N in S20), the equalization error generation unit 1040 determines the difference between the delay signal from the first delay unit 1032 and the provisional determination signal 1306. Is output as an error signal 1300 for linear equalization to the linear equalization unit 1044 (S22). Further, the equalization error generation unit 1040 uses the difference between the sum of the delay signal from the first delay unit 1032 and the delay signal from the second delay unit 1034 and the provisional determination signal 1306 as a non-linear equalization error signal 1302. It outputs to the nonlinear equalization part 1046 (S24). On the other hand, if the magnitude of the non-linear equalization error signal 1302 has converged within the threshold value (Y in S20), the equalization error generation unit 1040 and the delay signal from the first delay unit 1032 and the second delay unit 1034 The difference between the sum of the delay signal and the provisional determination signal 1306 is output to the linear equalization unit 1044 as the linear equalization error signal 1300 (S26). Further, the equalization error generation unit 1040 uses the difference between the sum of the delay signal from the first delay unit 1032 and the delay signal from the second delay unit 1034 and the provisional determination signal 1306 as a non-linear equalization error signal 1302. It outputs to the nonlinear equalization part 1046 (S28).
 図24は、非線形等化部1046における係数の導出手順を示すフローチャートである。判定部1042は、非線形等化用誤差信号1302の大きさが収束した後も、継続して非線形等化用誤差信号1302の大きさを導出する。大きさがしきい値よりも大きくなった場合(S40のY)、線形等化部1044は、非線形等化部1046へ新たにタップ係数を導出させる(S42)。大きさがしきい値よりも大きくならなければ(S40のN)、処理は終了される。 FIG. 24 is a flowchart showing a coefficient derivation procedure in the nonlinear equalization unit 1046. The determination unit 1042 continuously derives the magnitude of the nonlinear equalization error signal 1302 even after the magnitude of the nonlinear equalization error signal 1302 has converged. When the magnitude is larger than the threshold value (Y in S40), the linear equalization unit 1044 causes the nonlinear equalization unit 1046 to derive a new tap coefficient (S42). If the size does not become larger than the threshold value (N in S40), the process is terminated.
 本発明の実施例によれば、線形等化と非線形等化とを並列に実行するとともに、線形等化信号と非線形等化信号を加算し、加算信号を仮判定した結果を教師信号とするので、トレーニング信号の代わりに仮判定信号を教師信号に使用できる。また、トレーニング信号の代わりに仮判定信号を教師信号に使用するので、トレーニング信号を使用せずに、非線形等化のための係数を導出できる。また、トレーニング信号を使用せずに、非線形等化のための係数が導出されるので、トレーニング信号を使用せずに非線形等化を実行できる。また、トレーニング信号を使用せずに、非線形等化のための係数が導出されるので、記録密度の向上や記録パワー変動等によって生じる再生信号の非線形歪をトレーニング信号なしで低減できる。 According to the embodiment of the present invention, linear equalization and nonlinear equalization are performed in parallel, and the linear equalization signal and the nonlinear equalization signal are added, and the result of provisional determination of the addition signal is used as a teacher signal. The temporary determination signal can be used as the teacher signal instead of the training signal. Further, since the temporary determination signal is used as the teacher signal instead of the training signal, a coefficient for nonlinear equalization can be derived without using the training signal. Further, since the coefficient for nonlinear equalization is derived without using the training signal, nonlinear equalization can be performed without using the training signal. In addition, since a coefficient for nonlinear equalization is derived without using a training signal, nonlinear distortion of a reproduction signal caused by improvement in recording density, recording power fluctuation, or the like can be reduced without a training signal.
 また、パーシャルレスポンス規則にしたがった仮判定が実行されるので、パーシャルレスポンス処理に対応できる。また、非線形等化の係数が収束されるまで、非線形等化信号をビタビ復号部へ出力しないので、仮判定信号の精度の悪化を抑制できる。また、非線形等化の係数が収束されるまでの仮判定信号の精度の悪化が抑制されるので、処理の初期段階であっても、等化処理の精度の悪化を抑制できる。また、非線形等化の係数が収束されるまで、線形等化のための係数を導出するために、非線形等化信号を使用しないので、当該係数の導出精度の悪化を抑制できる。また、非線形等化の係数が収束されるまでの線形等化のための係数の導出精度の悪化が抑制されるので、処理の初期段階であっても、等化処理の精度の悪化を抑制できる。また、非線形等化用誤差信号の発散を検出した場合に複数の係数を新たに導出するので、等化特性の悪化を抑制できる。 Also, provisional judgment is executed according to the partial response rule, so it can support partial response processing. In addition, since the nonlinear equalization signal is not output to the Viterbi decoding unit until the nonlinear equalization coefficient is converged, deterioration of the accuracy of the provisional determination signal can be suppressed. In addition, since the deterioration of the accuracy of the provisional determination signal until the nonlinear equalization coefficient is converged is suppressed, the deterioration of the accuracy of the equalization processing can be suppressed even at the initial stage of the processing. Further, since the nonlinear equalization signal is not used to derive the coefficient for linear equalization until the nonlinear equalization coefficient is converged, deterioration of the derivation accuracy of the coefficient can be suppressed. In addition, since deterioration in the derivation accuracy of the coefficient for linear equalization until the nonlinear equalization coefficient is converged is suppressed, deterioration in the accuracy of the equalization process can be suppressed even at the initial stage of the process. . Further, since a plurality of coefficients are newly derived when the divergence of the non-linear equalization error signal is detected, deterioration of the equalization characteristics can be suppressed.
(実施例3)
 本発明の実施例3は、実施例2と同様に、線形波形等化器と非線形波形等化器とを並列に配置した再生装置に関する。線形波形等化器と非線形波形等化器とは、いずれも多段タップを備える。実施例3に係る再生装置では、回路規模を減少させるために、線形波形等化器と非線形波形等化器とにおいて、多段タップが共通化される。実施例3に係る再生装置100、処理部24は、図1、図13と同様のタイプである。ここでは、実施例2との差異を中心に説明する。
(Example 3)
As in the second embodiment, the third embodiment of the present invention relates to a reproducing apparatus in which a linear waveform equalizer and a nonlinear waveform equalizer are arranged in parallel. Both the linear waveform equalizer and the nonlinear waveform equalizer have multi-stage taps. In the reproducing apparatus according to the third embodiment, in order to reduce the circuit scale, the multistage tap is shared by the linear waveform equalizer and the nonlinear waveform equalizer. The playback apparatus 100 and the processing unit 24 according to the third embodiment are the same type as those in FIGS. Here, the difference from the second embodiment will be mainly described.
 図25は、本発明の実施例3に係る等化処理部1030の構成を示す。等化処理部1030は、線形処理部1052、非線形処理部1072、多段タップ1130を含む。多段タップ1130は、遅延タップ1132と総称される第1遅延タップ1132a、第2遅延タップ1132b、第3遅延タップ1132c、第N遅延タップ1132nを含む。 FIG. 25 shows a configuration of the equalization processing unit 1030 according to the third embodiment of the present invention. The equalization processing unit 1030 includes a linear processing unit 1052, a non-linear processing unit 1072, and a multistage tap 1130. The multistage tap 1130 includes a first delay tap 1132a, a second delay tap 1132b, a third delay tap 1132c, and an Nth delay tap 1132n, which are collectively referred to as a delay tap 1132.
 多段タップ1130は、図14の線形等化部1044に含まれる多段タップ1050、図15の非線形等化部1046に含まれる多段タップ1070と同様に構成されており、複数の遅延タップ1132がシリアルに接続されている。つまり、多段タップ1130は、線形等化部と非線形等化部とにおいて共通化されている。多段タップ1130からの出力は、線形処理部1052へ出力されるとともに非線形処理部1072へも出力されている。線形処理部1052は、図14に示された構成と同様であり、非線形処理部1072は、図15に示された構成と同様である。ここでは、これらの説明を省略する。 The multistage tap 1130 is configured in the same manner as the multistage tap 1050 included in the linear equalization unit 1044 in FIG. 14 and the multistage tap 1070 included in the nonlinear equalization unit 1046 in FIG. 15, and a plurality of delay taps 1132 are serially connected. It is connected. That is, the multistage tap 1130 is shared by the linear equalization unit and the nonlinear equalization unit. The output from the multistage tap 1130 is output to the linear processing unit 1052 and also to the nonlinear processing unit 1072. The linear processing unit 1052 is the same as the configuration shown in FIG. 14, and the nonlinear processing unit 1072 is the same as the configuration shown in FIG. Here, these descriptions are omitted.
 本発明の実施例によれば、線形等化と非線形等化とにおいて、多段タップが共通化されるので、回路規模の増加を抑制できる。また、線形等化と非線形等化とにおいて同一の処理を実行している多段タップを共通化しているので、線形等化と非線形等化とに対する処理精度の悪化を抑制できる。また、等化処理の精度を抑制させながら、回路規模を低減できる。 According to the embodiment of the present invention, since the multistage tap is shared in linear equalization and nonlinear equalization, an increase in circuit scale can be suppressed. In addition, since multistage taps performing the same processing are shared in linear equalization and nonlinear equalization, it is possible to suppress deterioration in processing accuracy with respect to linear equalization and nonlinear equalization. In addition, the circuit scale can be reduced while suppressing the accuracy of equalization processing.
 以上、本発明を実施例をもとに説明した。この実施例は例示であり、それらの各構成要素や各処理プロセスの組合せにいろいろな変形例が可能なこと、またそうした変形例も本発明の範囲にあることは当業者に理解されるところである。 The present invention has been described based on the embodiments. This embodiment is an exemplification, and it will be understood by those skilled in the art that various modifications can be made to the combination of each component and each processing process, and such modifications are also within the scope of the present invention. .
 本発明の実施例1において、線形等化部44および非線形等化部46は、いずれも複数のタップ係数を使用しており、複数のタップ係数を適応的に導出している。しかしながらこれに限らず例えば、非線形等化部46は、複数のタップ係数を適応的に導出し、線形等化部44は、複数のタップ係数として固定値を使用してもよい。つまり、線形等化部44は、複数のタップ係数を適応的に導出しなくてもよい。その際、第1加算部40は、線形等化部44へ線形等化用誤差信号300を出力しない。本変形例によれば、線形等化のための係数を固定値とするので、処理を簡易にできる。 In the first embodiment of the present invention, both the linear equalization unit 44 and the non-linear equalization unit 46 use a plurality of tap coefficients, and derive a plurality of tap coefficients adaptively. However, the present invention is not limited to this. For example, the nonlinear equalizer 46 may adaptively derive a plurality of tap coefficients, and the linear equalizer 44 may use a fixed value as the plurality of tap coefficients. That is, the linear equalization unit 44 does not have to adaptively derive a plurality of tap coefficients. At this time, the first addition unit 40 does not output the linear equalization error signal 300 to the linear equalization unit 44. According to this modification, since the coefficient for linear equalization is a fixed value, the processing can be simplified.
 本発明の実施例2と3において、線形等化部1044および非線形等化部1046は、いずれも複数のタップ係数を使用しており、複数のタップ係数を適応的に導出している。しかしながらこれに限らず例えば、非線形等化部1046は、複数のタップ係数を適応的に導出し、線形等化部1044は、複数のタップ係数として固定値を使用してもよい。つまり、線形等化部1044は、複数のタップ係数を適応的に導出しなくてもよい。図26は、本発明の変形例に係る処理部24の構成を示す。処理部24は、図13と比較して遅延部1140を備える。ここでは、図13との差異を中心に説明する。加算部1036は、線形等化部1044からの線形等化信号と非線形等化部1046からの非線形等化信号とを加算する。遅延部1140は、等化誤差生成部1040から出力した信号を遅延させる。ここで、遅延部1140は、ビタビ復号部1038における処理遅延に応じた期間にわたって遅延を実行する。また、等化誤差生成部1040は、遅延部1140において遅延した信号と、ビタビ復号部1038からの仮判定信号1306との差異をもとに非線形等化用誤差信号1302を生成し、非線形等化部1046は、非線形等化用誤差信号1302をもとに複数の係数を導出する。本変形例によれば、線形等化のための係数を固定値とするので、等化処理の安定性を向上できる。また、線形等化のための係数を固定値とするので、処理を簡易にできる。 In the second and third embodiments of the present invention, each of the linear equalization unit 1044 and the nonlinear equalization unit 1046 uses a plurality of tap coefficients and adaptively derives the plurality of tap coefficients. However, the present invention is not limited to this. For example, the nonlinear equalizer 1046 may adaptively derive a plurality of tap coefficients, and the linear equalizer 1044 may use a fixed value as the plurality of tap coefficients. That is, the linear equalization unit 1044 may not derive a plurality of tap coefficients adaptively. FIG. 26 shows the configuration of the processing unit 24 according to a modification of the present invention. The processing unit 24 includes a delay unit 1140 as compared with FIG. Here, it demonstrates centering on the difference with FIG. The adder 1036 adds the linear equalized signal from the linear equalizer 1044 and the nonlinear equalized signal from the nonlinear equalizer 1046. The delay unit 1140 delays the signal output from the equalization error generation unit 1040. Here, the delay unit 1140 performs a delay over a period corresponding to the processing delay in the Viterbi decoding unit 1038. Further, the equalization error generation unit 1040 generates a non-linear equalization error signal 1302 based on the difference between the signal delayed in the delay unit 1140 and the provisional determination signal 1306 from the Viterbi decoding unit 1038, and performs non-linear equalization The unit 1046 derives a plurality of coefficients based on the non-linear equalization error signal 1302. According to this modification, since the coefficient for linear equalization is a fixed value, the stability of the equalization process can be improved. In addition, since the coefficient for linear equalization is a fixed value, the processing can be simplified.
 10 光ディスク、 12 光ディスク駆動部、 14 光ピックアップ、 16 プリアンプ部、 18 AGC部、 20 PLL部、 22 A/D変換部、 24 処理部、 26 制御部、 30 仮判定部、 32 第1遅延部、 34 第2遅延部、 38 ビタビ復号部、 40 第1加算部、 42 第2加算部、 44 線形等化部、 46 非線形等化部、 100 再生装置。 10 optical discs, 12 optical disc drive units, 14 optical pickups, 16 preamplifier units, 18 AGC units, 20 PLL units, 22 A / D conversion units, 24 processing units, 26 control units, 30 temporary determination units, 32 first delay units, 34 second delay unit, 38 Viterbi decoding unit, 40 first addition unit, 42 second addition unit, 44 linear equalization unit, 46 non-linear equalization unit, 100 playback device.
 本発明によれば、記録密度の向上や記録パワー変動等によって生じる再生信号の非線形歪をトレーニング信号なしで低減できる。 According to the present invention, it is possible to reduce the non-linear distortion of the reproduction signal caused by the improvement of the recording density, the recording power fluctuation or the like without the training signal.

Claims (15)

  1.  処理対象の信号を順次線形等化する線形等化部と、
     前記線形等化部において線形等化した信号を順次仮判定する仮判定部と、
     前記仮判定部において仮判定した信号を教師信号として複数の係数を導出するとともに、複数の係数をもとに、前記線形等化部において線形等化した信号を順次非線形等化する非線形等化部と、
     を備えることを特徴とする等化器。
    A linear equalization unit that sequentially linearly equalizes the signal to be processed;
    A provisional determination unit that sequentially provisionally determines signals linearly equalized in the linear equalization unit;
    A non-linear equalization unit for deriving a plurality of coefficients using the signal provisionally determined by the temporary determination unit as a teacher signal and sequentially non-linear equalizing the signal linearly equalized by the linear equalization unit based on the plurality of coefficients When,
    An equalizer comprising:
  2.  前記仮判定部は、パーシャルレスポンス規則にしたがって仮判定を実行することを特徴とする請求項1に記載の等化器。 The equalizer according to claim 1, wherein the temporary determination unit performs a temporary determination according to a partial response rule.
  3.  前記非線形等化部において非線形等化した信号を遅延させる遅延部をさらに備え、
     前記遅延部は、前記仮判定部における処理遅延と前記非線形等化部における処理遅延との差異に応じた期間にわたって遅延を実行し、
     前記非線形等化部は、前記遅延部において遅延した信号と前記仮判定部において仮判定した信号との差異をもとに、複数の係数を導出することを特徴とする請求項1または2に記載の等化器。
    A delay unit that delays the signal that is nonlinearly equalized in the nonlinear equalizer;
    The delay unit executes a delay over a period according to a difference between a processing delay in the temporary determination unit and a processing delay in the nonlinear equalization unit,
    3. The non-linear equalization unit derives a plurality of coefficients based on a difference between a signal delayed by the delay unit and a signal provisionally determined by the temporary determination unit. Equalizer.
  4.  前記非線形等化部は、差異がしきい値よりも大きくなった場合に、複数の係数を新たに導出することを特徴とする請求項3に記載の等化器。 4. The equalizer according to claim 3, wherein the nonlinear equalizer newly derives a plurality of coefficients when the difference becomes larger than a threshold value.
  5.  入力した信号を順次線形等化するステップと、
     線形等化した信号を順次仮判定するステップと、
     仮判定した信号を教師信号として複数の係数を導出するとともに、複数の係数をもとに、線形等化した信号を順次非線形等化するステップと、
     を備えることを特徴とする等化方法。
    Sequentially equalizing the input signal, and
    Sequentially and temporarily determining linearly equalized signals;
    A step of deriving a plurality of coefficients using the provisionally determined signal as a teacher signal and sequentially nonlinearly equalizing the linearly equalized signal based on the plurality of coefficients;
    An equalization method comprising:
  6.  入力した信号を順次線形等化するステップと、
     線形等化した信号を順次仮判定するステップと、
     仮判定した信号を教師信号として複数の係数を導出するとともに、複数の係数をもとに、線形等化した信号を順次非線形等化するステップと、
     をコンピュータに実行させるためのプログラム。
    Sequentially equalizing the input signal, and
    Sequentially and temporarily determining linearly equalized signals;
    A step of deriving a plurality of coefficients using the provisionally determined signal as a teacher signal and sequentially nonlinearly equalizing the linearly equalized signal based on the plurality of coefficients;
    A program that causes a computer to execute.
  7.  処理対象の信号を順次入力する入力部と、
     前記入力部において入力した信号を順次線形等化する線形等化部と、
     前記線形等化部における線形等化に並行して、前記入力部において入力した信号を順次非線形等化する適応非線形等化部と、
     前記適応非線形等化部において非線形等化した信号と、前記線形等化部において線形等化した信号とを加算する加算部と、
     前記加算部において加算した信号を順次仮判定する仮判定部とを備え、
     前記適応非線形等化部は、前記仮判定部において仮判定した信号を教師信号として複数の係数を導出するとともに、複数の係数をもとに、非線形等化を実行することを特徴とする等化器。
    An input unit for sequentially inputting signals to be processed;
    A linear equalization unit that sequentially linearly equalizes the signal input in the input unit;
    In parallel with the linear equalization in the linear equalization unit, an adaptive nonlinear equalization unit that sequentially nonlinearly equalizes the signal input in the input unit;
    An adder that adds a signal that is nonlinearly equalized in the adaptive nonlinear equalizer and a signal that is linearly equalized in the linear equalizer;
    A temporary determination unit that sequentially determines the signals added in the addition unit sequentially,
    The adaptive nonlinear equalization unit derives a plurality of coefficients using the signal provisionally determined by the provisional determination unit as a teacher signal, and performs nonlinear equalization based on the plurality of coefficients vessel.
  8.  前記線形等化部に含まれた多段タップと、前記適応非線形等化部における多段タップとが、共通化されていることを特徴とする請求項7に記載の等化器。 The equalizer according to claim 7, wherein the multistage tap included in the linear equalization unit and the multistage tap in the adaptive nonlinear equalization unit are shared.
  9.  前記仮判定部は、パーシャルレスポンス規則にしたがって仮判定を実行することを特徴とする請求項7または8に記載の等化器。 The equalizer according to claim 7 or 8, wherein the temporary determination unit executes a temporary determination according to a partial response rule.
  10.  前記適応非線形等化部における複数の係数の収束を判定する判定部をさらに備え、
     前記加算部は、前記判定部において収束が判定されるまでの間、前記線形等化部において線形等化した信号を前記仮判定部へ出力し、前記判定部において収束が判定された後、加算した信号を前記仮判定部へ出力することを特徴とする請求項7から9のいずれかに記載の等化器。
    A determination unit for determining convergence of a plurality of coefficients in the adaptive nonlinear equalization unit;
    The addition unit outputs a signal linearly equalized by the linear equalization unit to the provisional determination unit until convergence is determined by the determination unit. After the convergence is determined by the determination unit, the addition is performed. The equalizer according to claim 7, wherein the received signal is output to the provisional determination unit.
  11.  前記線形等化部において線形等化した信号を遅延させる第1遅延部と、
     前記適応非線形等化部において非線形等化した信号を遅延させる第2遅延部とをさらに備え、
     前記第1遅延部は、前記仮判定部における処理遅延に応じた期間にわたって遅延を実行し、
     前記第2遅延部は、前記仮判定部における処理遅延に応じた期間にわたって遅延を実行し、
     前記適応非線形等化部は、前記第1遅延部において遅延した信号と前記第2遅延部において遅延した信号との和と、前記仮判定部において仮判定した信号との差異をもとに、複数の係数を導出し、
     前記線形等化部は、複数の係数を使用して線形等化を実行しており、前記判定部において収束が判定されるまでの間、前記第1遅延部において遅延した信号と前記仮判定部において仮判定した信号との差異をもとに、複数の係数を導出し、前記判定部において収束が判定された後、前記第1遅延部において遅延した信号と前記第2遅延部において遅延した信号との和と、前記仮判定部において仮判定した信号との差異をもとに、複数の係数を導出することを特徴とする請求項10に記載の等化器。
    A first delay unit that delays a signal linearly equalized in the linear equalization unit;
    A second delay unit that delays the signal that is nonlinearly equalized in the adaptive nonlinear equalization unit,
    The first delay unit executes a delay over a period according to the processing delay in the temporary determination unit,
    The second delay unit executes a delay over a period according to the processing delay in the temporary determination unit,
    The adaptive nonlinear equalization unit includes a plurality of signals based on a difference between a sum of a signal delayed in the first delay unit and a signal delayed in the second delay unit, and a signal temporarily determined in the temporary determination unit. The coefficient of
    The linear equalization unit performs linear equalization using a plurality of coefficients, and the signal delayed in the first delay unit and the provisional determination unit until convergence is determined by the determination unit. A plurality of coefficients are derived on the basis of the difference from the signal temporarily determined in step, and after the convergence is determined in the determination unit, the signal delayed in the first delay unit and the signal delayed in the second delay unit 11. The equalizer according to claim 10, wherein a plurality of coefficients are derived based on a difference between the sum of and a signal provisionally determined by the provisional determination unit.
  12.  前記線形等化部において線形等化した信号と前記適応非線形等化部において非線形等化した信号とを加算する加算部と、
     前記加算部から出力した信号を遅延させる遅延部とをさらに備え、
     前記遅延部は、前記仮判定部における処理遅延に応じた期間にわたって遅延を実行し、
     前記適応非線形等化部は、前記遅延部において遅延した信号と、前記仮判定部において仮判定した信号との差異をもとに、複数の係数を導出し、
     前記線形等化部は、複数の係数を使用して線形等化を実行しており、かつ複数の係数として固定値を使用することを特徴とする請求項7から10のいずれかに記載の等化器。
    An adder for adding the signal linearly equalized in the linear equalizer and the signal nonlinearly equalized in the adaptive nonlinear equalizer;
    A delay unit that delays the signal output from the adding unit;
    The delay unit executes a delay over a period according to the processing delay in the temporary determination unit,
    The adaptive nonlinear equalization unit derives a plurality of coefficients based on a difference between the signal delayed in the delay unit and the signal temporarily determined in the temporary determination unit,
    11. The linear equalization unit performs linear equalization using a plurality of coefficients, and uses a fixed value as the plurality of coefficients, etc. Generator.
  13.  前記適応非線形等化部は、複数の係数の発散を検出した場合に、複数の係数を新たに導出することを特徴とする請求項7から12のいずれかに記載の等化器。 13. The equalizer according to claim 7, wherein the adaptive nonlinear equalization unit newly derives a plurality of coefficients when detecting divergence of the plurality of coefficients.
  14.  入力した信号を順次線形等化するステップと、
     線形等化に並行して、入力した信号を順次非線形等化するステップと、
     非線形等化した信号と線形等化した信号とを加算するステップと、
     加算した信号を順次仮判定するステップとを備え、
     前記非線形等化するステップは、仮判定した信号を教師信号として複数の係数を導出するとともに、複数の係数をもとに、非線形等化を実行することを特徴とする等化方法。
    Sequentially equalizing the input signal, and
    In parallel with linear equalization, sequentially performing nonlinear equalization on the input signal;
    Adding a non-linear equalized signal and a linear equalized signal;
    The provisionally determining the added signal sequentially,
    The non-linear equalization step includes deriving a plurality of coefficients using the provisionally determined signal as a teacher signal and performing non-linear equalization based on the plurality of coefficients.
  15.  入力した信号を順次線形等化するステップと、
     線形等化に並行して、入力した信号を順次非線形等化するステップと、
     非線形等化した信号と線形等化した信号とを加算するステップと、
     加算した信号を順次仮判定するステップとを備え、
     前記非線形等化するステップは、仮判定した信号を教師信号として複数の係数を導出するとともに、複数の係数をもとに、非線形等化を実行することをコンピュータに実行させるためのプログラム。
    Sequentially equalizing the input signal; and
    In parallel with linear equalization, sequentially performing nonlinear equalization on the input signal;
    Adding a non-linear equalized signal and a linear equalized signal;
    The provisionally determining the added signals sequentially,
    The non-linear equalization step is a program for deriving a plurality of coefficients using the provisionally determined signal as a teacher signal and causing the computer to execute non-linear equalization based on the plurality of coefficients.
PCT/JP2010/001971 2009-03-19 2010-03-18 Equalizer and equalization method WO2010106810A1 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
US13/256,608 US8611411B2 (en) 2009-03-19 2010-03-18 Equalizer and equalization method
KR1020127033156A KR101495979B1 (en) 2009-03-19 2010-03-18 Equalizer and equalization method
CN201080012536.5A CN102356432B (en) 2009-03-19 2010-03-18 Equalizer and equalization method

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
JP2009-068949 2009-03-19
JP2009068949 2009-03-19
JP2009-072812 2009-03-24
JP2009072812A JP5136489B2 (en) 2009-03-24 2009-03-24 Equalizer and equalization method
JP2010-049358 2010-03-05
JP2010049358A JP5136577B2 (en) 2009-03-19 2010-03-05 Equalizer and equalization method

Publications (1)

Publication Number Publication Date
WO2010106810A1 true WO2010106810A1 (en) 2010-09-23

Family

ID=42739477

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2010/001971 WO2010106810A1 (en) 2009-03-19 2010-03-18 Equalizer and equalization method

Country Status (1)

Country Link
WO (1) WO2010106810A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113067786A (en) * 2020-01-02 2021-07-02 上海诺基亚贝尔股份有限公司 Apparatus, method, device and computer readable medium for equalizing a signal

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003085764A (en) * 2000-10-31 2003-03-20 Matsushita Electric Ind Co Ltd Waveform equalizer and prml detector
JP2005303361A (en) * 2004-04-06 2005-10-27 Sony Corp Signal processing apparatus and method, and signal decoding apparatus and method
JP2005302130A (en) * 2004-04-09 2005-10-27 Sony Corp Adaptive equalization apparatus and method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003085764A (en) * 2000-10-31 2003-03-20 Matsushita Electric Ind Co Ltd Waveform equalizer and prml detector
JP2005303361A (en) * 2004-04-06 2005-10-27 Sony Corp Signal processing apparatus and method, and signal decoding apparatus and method
JP2005302130A (en) * 2004-04-09 2005-10-27 Sony Corp Adaptive equalization apparatus and method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113067786A (en) * 2020-01-02 2021-07-02 上海诺基亚贝尔股份有限公司 Apparatus, method, device and computer readable medium for equalizing a signal
CN113067786B (en) * 2020-01-02 2023-08-29 上海诺基亚贝尔股份有限公司 Apparatus, method, device and computer readable medium for equalizing signals

Similar Documents

Publication Publication Date Title
KR101495979B1 (en) Equalizer and equalization method
JP5038495B2 (en) Feedforward DC regeneration method and circuit in perpendicular magnetic read channel
AU660227B2 (en) Signal processing system having intersymbol-interference cancelling means and method of same
JP2004080210A (en) Digital filter
JP2005135563A (en) Adaptive equalizer
JP2005302130A (en) Adaptive equalization apparatus and method
JP4072751B2 (en) Playback device
JP5445416B2 (en) Equalizer and equalization method
JP5136489B2 (en) Equalizer and equalization method
WO2010106810A1 (en) Equalizer and equalization method
JP5136577B2 (en) Equalizer and equalization method
JP5212260B2 (en) Equalizer, equalization method, and program
JP2011008836A (en) Equalizer and equalizing method
JP2008159138A (en) Reproduction device and reproducing method, signal processor and signal processing method, and program
JP4727310B2 (en) Waveform equalization apparatus, information reproducing apparatus, waveform equalization method, waveform equalization program, and recording medium
JP3471245B2 (en) Information reproducing device using feedback filter
JP4593312B2 (en) Waveform equalization apparatus, information reproducing apparatus, waveform equalization method, waveform equalization program, and recording medium
JP2000151425A (en) Prml(partial response most liklihood) detector and data error correction code generator
JP2005267840A (en) Adaptive equalizer, decoding device, and error detecting device
JP4318028B2 (en) Signal processing device
JP3875154B2 (en) Waveform equalization apparatus, information reproducing apparatus, communication apparatus, waveform equalization method, waveform equalization program, and computer-readable recording medium recording the waveform equalization program
JP5182170B2 (en) Equalizer
JP3994987B2 (en) Playback device
JP4009965B2 (en) Viterbi decoding method
JP2006121285A (en) Maximum likelihood decoding method and maximum likelihood decoding device

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 201080012536.5

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 10753305

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 13256608

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 20117024218

Country of ref document: KR

Kind code of ref document: A

122 Ep: pct application non-entry in european phase

Ref document number: 10753305

Country of ref document: EP

Kind code of ref document: A1