JPH08129833A - Regenerated data detecting method - Google Patents

Abstract

PURPOSE: To improve tone quality of a reproduced sound by reproducing its recording code for the recording code in which the same codes are continued by two bits or more as the ternary data by using partial response equalization, etc. CONSTITUTION: When the code of which the continuous code such as 1, 7 codes become two or more is reproduced, a signal a(t) after 1, 7 modulated is converted to the signal b(t) by a precoder 11 to be recorded on a recording medium 12. The signal c(t) read out of the recording medium 12 is converted to the ternary data like d(t) by a partial response (1.0, -1) equalizer 13 outputting '...0010-1000...' for an input signal '1' to be inputted to a viterbi decoder 14. At this time, since the frequency of appearance of '1' and '-1' become equal in the output of the PR (1.0, -1) equalizer, the signal c(t) is correctly ternarized.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a reproduced data detecting device, and more particularly to a reproduced data detecting device for error correction.

[0002]

2. Description of the Related Art Generally, in a digital VTR, a digital optical disk recording apparatus, etc., when the reproduced digital data is discriminated, "H" is given if the reproduction voltage level exceeds a predetermined threshold voltage, and "L" if not. Many devices are configured to determine
When the reproduction voltage of the bit originally "L" slightly exceeds the threshold voltage for some reason, the bit is naturally determined to be "H".

In order to prevent such an erroneous determination, in the technique described in Japanese Patent Laid-Open No. 4-307817,
A recording code having two or more consecutive codes, for example, 1,
Data is stored by 7 codes, the reproduction signal is partial response (1, 1) equalized, and the equalized ternary data is 4-state Viterbi-decoded. Error correction is performed.

[0004]

DISCLOSURE OF THE INVENTION Problems to be Solved by the Invention
The technique described in Japanese Patent Publication No. 7 has a drawback in that although bit error correction can be performed, partial response (1, 1) equalization is not DC-free and the bit error rate after equalization cannot be made sufficiently small. Here, DC-free means that the number of appearances of “1” and “−1” after equalization is equal, and for example, the recording signal “111111100” is used.
1111111000 ”is the partial response (1,
1) When equalized, "... 01111110-101111
110-1-1 ... ", and the number of appearances of" 1 "and" -1 "is completely different.

In such a state, the balance between the + side and the-side is poor, so the equalization characteristic is poor, and the bit error rate after equalization cannot be minimized.

Therefore, an object of the present invention is to provide a reproduction data detection method capable of sufficiently reducing the bit error rate for a recording code in which the number of continuous codes is always 2 or more.

[0007]

In the reproduced data detecting method according to the present invention, for a recording code in which the same code continues for 2 bits or more, the recording code is a partial response (1, 0, -1). The state S is reproduced as ternary data by using equalization, and transits to the state S0 when 0 is input and transits to the state S1 when 1 is input.
State S that transits to state S2 when 0 and 1 are input
A state S2 that transitions to state S2 when 1 and 0 are input, and a state S2 that transitions to state S3 when -1 is input, and -1
Is a value that corresponds to a path that is a transition history of states that are stored in association with four states, that is, a state S3 that transitions to the state S0 when is input, and the degree of conformity between each path and the input data history. The path metric is updated according to the probability that a value reproduced in each state is obtained, and when there is only one continuous path, the states S0 and S2 in the transition history corresponding to the path are updated. 0 for transition to state S
The data in which 1 is associated with 1 and the transition to S3 is set as reproduction data.

That is, in the reproduction data detecting method according to the first aspect, the fact that the partial response (1, 0, -1) equalization is DC free is utilized to improve the equalization characteristic, and further, Playback data is obtained by performing Viterbi decoding by making the most of the characteristic that consecutive codes are two or more.

In the reproduced data detecting method according to claim 2,
For a recording code in which the same code continues for 3 bits or more, the recording code is reproduced as ternary data by using partial response (1, 0, -1) equalization, and when 0 is input, state S0 is input. State S0 that changes to state S1 when 1 is input, state S1 that changes to state S4 when 1 is input, and state S4 that changes to state S2 when 0 is input. And a state S2 that transitions to state S2 when 0 is input and a transition to state S3 when -1 is input, and a state S5 when -1 is input.
State S3 transiting to, and state S0 when 0 is input
In each state, the path that is the transition history of the states stored in association with the six states of the state S5 that transits to and the path metric that is a value according to the degree of conformity between each path and the input data history are reproduced. The value is updated according to the probability that the obtained value is obtained, and when there is only one continuous path, the states S0 and S of the transition history corresponding to the path are updated.
Data in which 0 is assigned to the transitions to 2, S4, and S5 and 1 is assigned to the transitions to the states S1 and S3 is set as reproduction data.

That is, in the reproduction data detecting method according to the first aspect, the fact that the partial response (1, 0, -1) equalization is DC-free is utilized to improve the equalization characteristic. By utilizing the characteristic that the number of consecutive codes is 3 or more, the Viterbi decoding is performed to obtain the reproduction data.

In the reproduced data detecting method according to the third aspect,
For a recording code in which the same code continues for 2 bits or more, the recording code is reproduced as ternary data by using partial response (1, -1) equalization, and transitions to state S0 when 0 is input. Then, a state S0 that transitions to the state S1 when -1 is input, a state S1 that transitions to the state S2 when 0 is input, and a state S1 that transitions to the state S2 when 0 is input. A path, which is a transition history of states stored in association with four states, a state S2 that transitions to state S3 when input, and a state S3 that transitions to state S0 when 0 is input, and The path metric, which is a value corresponding to the goodness of fit between the path and the input data history, is updated according to the probability that the value reproduced in each state is obtained, and when there is only one continuous path. Corresponding to that path Of transfer history, the state S0 and S2
The data in which 0 is assigned to the transition to 1 and 1 is assigned to the transitions to the states S1 and S3 is the reproduction data.

That is, in the reproduced data detecting method according to the third aspect, the partial response (1, -1) equalization is D
By utilizing the fact that it is C-free, the equalization characteristic is improved, and further, the Viterbi decoding is performed by taking advantage of the characteristic that the number of consecutive codes is 2 or more, thereby obtaining the reproduction data.

In the reproduced data detecting method according to the fourth aspect,
For a recording code in which the same code continues for 3 bits or more, the recording code is reproduced as ternary data by using partial response (1, -1) equalization, and transits to state S0 when 0 is input. Then, a state S0 that transitions to the state S1 when 1 is input, a state S1 that transitions to the state S4 when 1 is input, and a state S4 that transitions to the state S2 when 0 is input, When 0 is input, the state transitions to state S2, when -1 is input, the state S2 transitions to state S3, when 0 is input, the state S3 transitions to state S5, and 0 is input. A path, which is a history of transitions stored in association with six states of state S5 that transitions to state S0 when the state changes, and a path metric that is a value according to the degree of conformity between each path and the input data history. , Play in each state The value is updated according to the probability that the obtained value is obtained, and when there is only one continuous path, 0 is set for the transition to states S0 and S2 in the transition history corresponding to that path. 1 for transition to S1, S3, S4 and S5
The data associated with is used as the reproduction data.

That is, in the reproduced data detecting method according to the fourth aspect, the partial response (1, -1) equalization is D
By utilizing the fact that it is C-free, the equalization characteristic is improved, and further, the Viterbi decoding is performed by taking advantage of the characteristic that the number of consecutive codes is 3 or more, thereby obtaining the reproduction data.

It should be noted that, in the same manner as the reproduction data detection method according to any one of claims 1 to 4, "1" for each state,
The NRZ data can be obtained by associating "0", but the NRZI data can also be obtained by making the association as in claim 5 or claim 6.

[0016]

EXAMPLES The present invention will be described in detail below with reference to examples.

First embodiment

FIG. 1 shows a block diagram of an apparatus for realizing the reproduction data detecting method of the first embodiment, and FIG. 2 shows
The simple timing chart of the signal state of each part is shown.
First, the outline of the operation of the reproduction data detecting method according to the first embodiment will be described with reference to these drawings.

The reproduction data detecting method of the first embodiment is a method for reproducing a code having two or more consecutive codes such as 1,7 codes, and is a signal after 1,7 modulation. a (t) is the signal b (t) generated by the precoder 11.
And is recorded on the recording medium 12.

The signal c read from the recording medium 12
(T) is "... 0010-1" for the input signal "1"
Partial response (1, 0,
-1) The equalizer 13 converts the data into ternary data like d (t) and inputs the data to the Viterbi decoder 14. At this time, the output of the PR (1,0, -1) equalizer has the same number of appearances of "1" and "-1", so that the signal c (t) can be accurately ternarized. ing.

In the Viterbi decoder used in the reproduction data detection method of the first embodiment, the signal d (t) (hereinafter referred to as y _n ) input from the PR (1, 0, -1) equalizer. )
Based on, the following processing is performed.

FIGS. 3 and 4 show a state transition diagram and a trellis diagram in the Viterbi decoder used in the reproduction data detecting method of the first embodiment, respectively. In the figure, among the numerical values expressed as X / Y (X, Y = -1, 0, 1), the numerical value written in the denominator is the input to the Viterbi decoder, and the numerator is the Viterbi decoder. Is the output from.

As is clear from these figures, in the Viterbi decoder of the first embodiment, since the codes to be reproduced are two or more continuous codes, "0" is input in the state S0. At this time, the output data is "0" and the state shifts to S0. When "1" is input in the state S0, the output data is "1" and the state shifts to S1. Also,
When "1" is input in the state S1, the output data is "0", the state shifts to S2, and "-" in the state S2.
When "1" is input, the output data is "1" and the state shifts to S3. When "0" is input in the state S2, the output data is "0" and the state shifts to S2. When "-1" is input in the state S3, the output data is "0" and the state changes to S0.

In the Viterbi decoder of the first embodiment, 0 is input.
Probability distribution in which -1 to 1 is input when it should be forced
Assuming that the normal distribution as shown in FIG.
Transition from state Si to state Sj (i, j = 0 to 3)
And the probability P of detecting Δy _ijIs defined as
ing.

[0025]

[Equation 1]

Therefore, the branch metric I _ij, which is a value obtained by multiplying the difference between the logarithmic value of the probability P _{ij and} the logarithmic value of the probability P _{00 by} a predetermined coefficient (2σ ² ), is the value of the input data y _n . Depending on the, the following values will be taken.

I ₀₀ = 0 I ₀₁ = -y _n +0.5 I ₁₂ = -y _n +0.5 I ₂₂ = 0 I ₂₃ = y _n +0.5 I ₃₀ = y _n +0.5

The path metric for each state is
The path metric before transitioning to the
Calculated as the sum of the launch metrics, shown in Figure 4.
As described above, only the transition from the state S2 is included in the state S3.
Signal y _nState S when is input
3 path metric m_n(S3) is the password for status S2.
Trick m_n-1(S2) and the transition from state S2 to state S3
Lunch Metric I_{twenty three}Is calculated as follows
You.

M _n (S3) = m _n-1 (S2) + I ₂₃ = m _n-1 (S2) + y _n +0.5

Further, since the transition to the state S2 includes the transition from the state S1 and the transition from the state S2, the state S2
The path metric m _n (S2) of the state S2 is the path metric m _n-1 (S1) of the state S2, the path metric calculated from the branch metric I ₁₂ from the state S1 to the state S2, and the path metric m _{n of the} state S2. _-1 (S2) and state S
Of the path metrics calculated from the branch metric I ₂₂ from 2 to the state S2, the smaller path metric is set. That is, the path metric m _n (S2)
Is calculated as follows.

M _n (S2) = min [m _n-1 (S2) + I ₂₂ , m _n-1 (S1) + I ₁₂ ] = min [m _n-1 (S2), m _n-1 (S1)- y _n +0.5]

Similarly, the path metrics for the states S1 and S0 are calculated according to the following equations.

M _n (S1) = m _n-1 (S0) + I ₀₁ = m _n-1 (S0) -y _n +0.5 m _n (S0) = min [m _n-1 (S0) + I ₀₀ , m _n-1 (S3) + I ₃₀ ] = min [m _n-1 (S0), m _n-1 (S3) + y _n +0.5]

As described above, the path metric calculation procedure has four calculation procedures depending on which relational expression is used when calculating m _n (S2) and m _n (S0). Calculation procedures and trellis diagrams in these four cases (denoted as merge 0 to merge 4) are shown in FIGS. 6 to 9, respectively.

The Viterbi decoder of the first embodiment uses the above relational expression to determine merge every time a signal is input and calculates a path metric based on the signal y. Then, when the three consecutive states are transited such as merge 0 or merge 1 → merge 1 → merge 0 or merge 1, the signal is reproduced by the path stored for the state S2. If there is a transition such as merging 0 or merging 2 → merging 2 → merging 0 or merging 2, it is assumed that the signal is reproduced by the path stored for state S0, and state S0 , S2, output 0, state S1,
The data corresponding to the output 1 is output to S3.

For example, as shown in FIG. 10, when the state has been changed, since the transitions by merging 1, merging 1, and merging 0 are performed after *, the state is changed to the position *. The path is merged with S2, and "10010"
Data will be obtained.

Second embodiment

The reproduction data detection method according to the second embodiment is a method for reproducing a code in which the number of consecutive codes such as 2, 7 codes is 3 or more, and is read from the recording medium. For the input signal "1",
Viterbi decoding is performed after conversion into ternary data by partial response (1, 0, -1) equalization that outputs "... 0010-100 ...".

11 and 12 show a state transition diagram and a trellis diagram in the Viterbi decoder used in the reproduction data detection method of the second embodiment, respectively. In addition,
In the figure, among the numerical values expressed as X / Y (X, Y = -1, 0, 1), the numerical value written in the denominator is the input to the Viterbi decoder, and the numerator is the Viterbi decoder. Is the output.

As is apparent from these figures, in the Viterbi decoder of the second embodiment, since the codes to be reproduced are three or more continuous codes, "0" is input in the state S0. At this time, the output data is "0" and the state shifts to S0. When "1" is input in the state S0, the output data is "1" and the state shifts to S1. Also,
When "1" is input in the state S1, the output data is "0", the state transits to S4, and "0" in the state S4.
When is input, the output data is "0", and the state shifts to S2. When "-1" is input in the state S2, the output data is "1" and the state is S3.
And when "0" is input in the state S2, the output data is "0" and the state shifts to S2. Also, when "-1" is input in the state S3, the output data is "0", the state shifts to S5, and when "0" is input in the state S5, the output data is "0". Yes, the state transitions to S0.

The branch metric I _ij in the second embodiment is obtained by the same derivation method as that described in the first embodiment.
Is calculated as follows according to the value of the input data y _n .

I ₀₀ = 0 I ₀₁ = -y _n +0.5 I ₁₄ = -y _n +0.5 I ₄₂ = 0 I ₂₂ = 0 I ₂₃ = y _n +0.5 I ₃₅ = y _n +0.5 I ₅₀ = 0

Therefore, the path metric for each state is calculated as follows.

M _n (S5) = m _n-1 (S3) + I ₃₅ = m _n-1 (S3) + y _n +0.5 m _n (S3) = m _n-1 (S2) + I ₂₃ = m _{n- 1} (S2) + y _n +0.5 m _n (S2) = min [m _n-1 (S2) + I ₂₂ , m _n-1 (S4) + I ₄₂ ] = min [m _n-1 (S2), m _{n -1 (S4)] m n (} S4) = m n-1 (S1) + I 14 = m n-1 (S1) -y n +0.5 m n (S1) = m n-1 (S0) + I 01 = M _n-1 (S0) -y _n +0.5 m _n (S0) = min [m _n-1 (S0) + I ₀₀ , m _n-1 (S5) + I ₅₀ ] = min [m _n-1 ( S0), m _n-1 (S5)]

Similar to the first embodiment, these relational expressions can be classified into merge 1 to merge 4, and the path metric calculation procedure and trellis diagram in each merge are shown in FIGS. 13 to 16, respectively. It becomes as shown.

The Viterbi decoder of the second embodiment uses the above relational expression to determine merge every time a signal is input, and calculates the path metric based on the signal y. Then, in the case where the continuous 5 states have changed such as merge 0 or merge 1 → merge 0 or merge 1 → merge 1 → merge 0 or merge 1 → merge 0 or merge 1, for state S2 Assuming that the signal is reproduced by the stored path, merge 0 or merge 2 → merge 0 or merge 2 → merge 2
→ merge 0 or merge 2 → merge 0 or merge 2,
If the signal is reproduced by the path stored for the state S0, the output 0 is changed to the states S0, S2, S4, S5 and the states S1, S3 are changed. On the other hand, the data corresponding to the output 1 is output.

For example, as shown in FIG. 17, when the state has changed, the path is merged with the state S0 at the position of *, and the data "10000" is obtained.

When the number of consecutive codes is 4 or more, it can be dealt with by inserting S4 and S5 for each (the number of consecutive codes-2).

Third Embodiment

The reproduction data detection method according to the third embodiment is a method for reproducing a code in which two or more consecutive codes such as 1,7 codes are read out from a recording medium. For the input signal "1",
Viterbi decoding is performed after conversion into ternary data by partial response (1, -1) equalization that outputs "... 001-100 ...".

FIG. 18 shows a state transition diagram of the Viterbi decoder in the reproduction data detection system of the third embodiment, and FIG.
Shows a simple timing chart. Note that FIG.
Among the numerical values written as X / Y (X, Y = -1, 0, 1), the numerical value written in the denominator is the input to the Viterbi decoder, and the numerator is the output from the Viterbi decoder. Is.

As is apparent from these figures, in the Viterbi decoder of the third embodiment, since the codes to be reproduced are two or more continuous codes, "0" is input in the state S0. At this time, the output data is "0" and the state shifts to S0. When "-1" is input in the state S0, the output data is "1" and the state shifts to S1. When "0" is input in the state S1, the output data is "0", the state transits to S2, and "1" is input in the state S2.
When is input, the output data is "1" and the state shifts to S3. When "0" is input in the state S2, the output data is "0" and the state shifts to S2. When "0" is input in the state S3,
The output data becomes "0" and the state changes to S0.

By the derivation method similar to the method described in the first embodiment, the branch metric I _ij in the third embodiment is obtained.
Is calculated as follows according to the value of the input data y _n .

I ₀₀ = 0 0 I ₀₁ = -y _n +0.5 I ₁₂ = 0 I ₂₂ = 0 I ₂₃ = y _n +0.5 I ₃₀ = 0

Therefore, the path metric for each state is calculated as follows.

M _n (S3) = m _n-1 (S2) + I ₂₃ = m _n-1 (S2) + y _n +0.5 m _n (S2) = min [m _n-1 (S2) + I ₂₂ , m _{n-1 (S1) + I} 12] = min [m n-1 (S2), m n-1 (S1)] m n (S1) = m n-1 (S0) + I 01 = m n-1 (S0 ) -Y _n +0.5 m _n (S0) = min [m _n-1 (S0) + I ₀₀ , m _n-1 (S3) + I ₃₀ ] = min [m _n-1 (S0), m _n-1 (S3)]

Similar to the first embodiment, these relational expressions can be classified into merge 1 to merge 4, and the path metric calculation procedure and trellis diagram in each merge are shown in FIGS. 20 to 23, respectively. It becomes as shown.

The Viterbi decoder of the third embodiment determines the merge by using the above relational expression each time a signal is input, and calculates the path metric based on the signal y. Then, when the three consecutive states are transited such as merge 0 or merge 1 → merge 1 → merge 0 or merge 1, the signal is reproduced by the path stored for the state S2. If there is a transition such as merging 0 or merging 2 → merging 2 → merging 0 or merging 2, it is assumed that the signal is reproduced by the path stored for state S0, and state S0 , S2, output 0, state S1,
The data corresponding to the output 1 is output to S3.

Fourth Embodiment

The reproduction data detection method according to the fourth embodiment is a method for reproducing a code in which the number of consecutive codes such as 2, 7 codes is 3 or more, and is read from the recording medium. For the input signal "1",
Viterbi decoding is performed after conversion into ternary data by partial response (1, -1) equalization that outputs "... 001-100 ...".

FIG. 24 shows a state transition diagram in the Viterbi decoder used in the reproduction data detecting method of the fourth embodiment. In the figure, among the numerical values expressed as X / Y (X, Y = -1, 0, 1), the numerical value written in the denominator is the input to the Viterbi decoder, and the numerator is the Viterbi decoder. Is the output from.

As is clear from this figure, in the Viterbi decoder of the fourth embodiment, since the codes to be reproduced are three or more continuous codes, when "0" is input in state S0. The output data is "0", the state changes to S0, and when "1" is input in the state S0, the output data is "1" and the state changes to S1. Further, when "1" is input in the state S1, the output data is "0", the state transits to S4, and when "0" is input in the state S4, the output data is "0". , The state transits to S2. When "-1" is input in the state S2, the output data is "1", the state shifts to S3, and when "0" is input in the state S2, the output data is "0". Yes, the state transitions to S2. When "0" is input in the state S3, the output data is "0", the state shifts to S5, and "0" is input in the state S5.
When is input, the output data becomes "0" and the state changes to S0.

The branch metric I _ij in the fourth embodiment is obtained by the derivation method similar to the method described in the first embodiment.
Is calculated as follows according to the value of the input data y _n .

[0064] _{I 00 = 0 I 01 = -y} n +0.5 I 14 = 0 I 42 = 0 I 22 = 0 I 23 = y n +0.5 I 35 = 0 I 50 = 0

Therefore, the path metric for each state is calculated as follows.

M _n (S5) = m _n-1 (S3) + I ₃₅ = m _n-1 (S3) m _n (S3) = m _n-1 (S2) + I ₂₃ = m _n-1 (S2) + y _n + 0.5 m _n (S2) = min [m _n-1 (S2) + I ₂₂ , m _n-1 (S4) + I ₄₂ ] = min [m _n-1 (S2), m _n-1 (S4) ] M _n (S4) = m _n-1 (S1) + I ₁₄ = m _n-1 (S1) m _n (S1) = m _n-1 (S0) + I ₀₁ = m _n-1 (S0) -y _n +0.5 m _n (S0) = min [m _n-1 (S0) + I ₀₀ , m _n-1 (S5) + I ₅₀ ] = min [m _n-1 (S0), m _n-1 (S5)]

Similar to the first embodiment, these relational expressions can be classified into merge 1 to merge 4, and the path metric calculation procedure and trellis diagram in each merge are shown in FIGS. 25 to 28, respectively. It becomes as shown.

The Viterbi decoder of the fourth embodiment uses the above relational expression each time a signal is input to determine the merge and calculates the path metric based on the signal y. Then, in the case where the continuous 5 states have changed such as merge 0 or merge 1 → merge 0 or merge 1 → merge 1 → merge 0 or merge 1 → merge 0 or merge 1, for state S2 Assuming that the signal is reproduced by the stored path, merge 0 or merge 2 → merge 0 or merge 2 → merge 2
→ merge 0 or merge 2 → merge 0 or merge 2,
If the signal is reproduced by the path stored for the state S0, the output 0 is changed to the states S0, S2, S4, S5 and the states S1, S3 are changed. On the other hand, the data corresponding to the output 1 is output.

In any of the embodiments described above, NR after 1,7 coding or 2,7 coding
Z data will be obtained, but if NRZI data is needed, in the first and third embodiments, state S0
The reproduction data may be data in which 0 is assigned to the transitions to S3 and S3 and 1 is assigned to the transitions to states S1 and S2. In the second and fourth embodiments, the states S0, S3, and S5 are set. The data in which 0 is associated with the transition of 1 and 1 is associated with the transitions to the states S1, S2, and S4 may be used as the reproduction data.

[0070]

As described above in detail, according to the reproduction data detecting method of the present invention, the equalization is performed in the DC-free state, and the determination utilizing the characteristics of the recording code is performed. Therefore, the bit error rate at the time can be sufficiently reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing an outline of an apparatus for realizing a reproduction data detection method according to the present invention.

FIG. 2 is a timing chart for explaining a reproduction data detection method according to the first embodiment.

FIG. 3 is a state transition diagram in Viterbi decoding used in the reproduction data detection method according to the first embodiment.

FIG. 4 is a trellis diagram in Viterbi decoding used in the reproduction data detection method according to the first embodiment.

FIG. 5 is a diagram showing a probability distribution function used for calculating a branch metric.

FIG. 6 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 0 in the reproduction data detection method according to the first embodiment.

FIG. 7 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 1 in the reproduction data detection method according to the first embodiment.

FIG. 8 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 2 in the reproduction data detection method according to the first embodiment.

FIG. 9 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 3 in the reproduction data detection method according to the first embodiment.

FIG. 10 is a trellis diagram showing an example of a detection result of the reproduction data detection method according to the first embodiment.

FIG. 11 is a state transition diagram in Viterbi decoding used in the reproduction data detection method according to the second embodiment.

FIG. 12 is a trellis diagram in Viterbi decoding used in the reproduction data detection method according to the second embodiment.

FIG. 13 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 0 in the reproduction data detection method according to the second embodiment.

FIG. 14 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 1 in the reproduction data detection method according to the second embodiment.

FIG. 15 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 2 in the reproduction data detection method according to the second embodiment.

FIG. 16 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 3 in the reproduction data detection method according to the second embodiment.

FIG. 17 is a trellis diagram showing an example of the detection result of the reproduction data detection method according to the second embodiment.

FIG. 18 is a state transition diagram in Viterbi decoding used in the reproduction data detection method according to the third embodiment.

FIG. 19 is a timing chart for explaining a reproduction data detection method according to the third embodiment.

FIG. 20 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 0 in the reproduction data detection method according to the third embodiment.

FIG. 21 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 1 in the reproduction data detection method according to the third embodiment.

FIG. 22 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 2 in the reproduction data detection method according to the third embodiment.

FIG. 23 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 3 in the reproduction data detection method according to the third embodiment.

FIG. 24 is a state transition diagram in Viterbi decoding used in the reproduction data detection method according to the fourth example.

FIG. 25 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 0 in the reproduction data detection method according to the fourth embodiment.

FIG. 26 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 1 in the reproduction data detection method according to the fourth embodiment.

FIG. 27 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 2 in the reproduction data detection method according to the fourth example.

FIG. 28 is an explanatory diagram showing a path metric calculation procedure and a trellis diagram corresponding to merge 3 in the reproduction data detection method according to the fourth example.

[Explanation of symbols]

 11 Precoder 12 Recording Medium 13 PR (1, 0, -1) Equalizer 14 Viterbi Decoder

Continuation of front page (51) Int.Cl. ⁶ Identification number Office reference number FI technical display location H03M 13/12 8730-5K H04L 1/00 F

Claims (6)

[Claims] 1. For a recording code in which the same code continues for 2 bits or more, the recording code is converted into a partial response (1,
0, -1) is reproduced as ternary data by using equalization, and when state 0 is input, state S0 transits to state S0, and when 1 is input state S0 transits to state S1 and 1 is input. State S1 that transitions to state S2 when input is performed, state S2 that transitions to state S2 when 0 is input, and state S2 that transitions to state S3 when -1 is input, and when -1 is input A path which is a transition history of states stored in association with four states of state S3 which transits to state S0, and a path metric which is a value corresponding to the compatibility between each path and the input data history are It is updated according to the probability that the value reproduced in the state is obtained, and when there is only one continuous path, the transition history to the states S0 and S2 in the transition history corresponding to that path becomes 0. Corresponds to the transition to states S1 and S3 Reproduced data detection method characterized in that the the reproduced data data. 2. With respect to a recording code in which the same code continues for 3 bits or more, the recording code is converted into a partial response (1,
0, -1) is reproduced as ternary data by using equalization, and when state 0 is input, state S0 transits to state S0, and when 1 is input state S0 transits to state S1 and 1 is input. State S1 that changes to state S4 when 0 is input, state S4 that changes to state S2 when 0 is input, and state S2 that changes to state S2 when 0 is input and -1 is input. It is stored in association with six states: a state S2 that transitions to state S3, a state S3 that transitions to state S5 when -1 is input, and a state S5 that transitions to state S0 when 0 is input. The path that is the transition history of the state and the path metric that is a value corresponding to the degree of conformity between each path and the input data history are updated according to the probability that the value reproduced in each state is obtained, When there is only one continuous pass In the transition history corresponding to the path, the reproduction data is data in which 0 is assigned to the transitions to the states S0, S2, S4, and S5 and 1 is assigned to the transitions to the states S1 and S3. Data detection method. 3. For a recording code in which the same code continues for 2 bits or more, the recording code is converted into a partial response (1,
-1) While reproducing as three-valued data by using equalization, the state S0 transits to the state S0 when 0 is input and the state S0 transits to the state S1 when -1 is input, and 0 is input. State S1 that changes to state S2 when 0 is input, state S2 that changes to state S2 when 0 is input, state S2 that changes to state S3 when 1 is input, and state S0 when 0 is input. In each state, the path, which is the transition history of the states stored in association with the four states of the state S3, and the path metric, which is a value according to the degree of conformity between each path and the input data history, are reproduced in each state. The updated value is updated according to the probability of being obtained, and when there is only one continuous path, the transition history corresponding to the path is set to 0 for the transition to states S0 and S2. Associate 1 with the transition to S1 and S3 Reproducing data detection method characterized in that the reproduction data data. 4. For a recording code in which the same code continues for 3 bits or more, the recording code is converted into a partial response (1,
-1) Playback as three-valued data using equalization, and transition to state S0 when 0 is input, and transition to state S1 when 1 is input, and state S0 and 1 are input. The state S1 sometimes changes to the state S4, the state S4 changes to the state S2 when 0 is input, the state S2 changes to the state S2 when 0 is input, and the state S3 changes when -1 is input. A state that is stored in association with six states, a state S2 that transitions to state S3, a state S3 that transitions to state S5 when 0 is input, and a state S5 that transitions to state S0 when 0 is input. The path metric, which is the transition history of, and the path metric, which is a value corresponding to the degree of conformity between each path and the input data history, are updated according to the probability that the value reproduced in each state is obtained, and continuous paths are updated. When there is only one Reproduction data characterized in that, in the transition history corresponding to the data, the reproduction data is data in which the transitions to the states S0 and S2 are associated with 0 and the transitions to the states S1, S3, S4, and S5 are associated with 1. Detection method. 5. The transition to states S0 and S3 is set to 0, and the transition to state S
4. The reproduction data detecting method according to claim 1 or 3, wherein the data in which 1 is associated with 1 and the transition to S2 is used as reproduction data. 6. The transition to states S0, S3 and S5 is set to 0,
5. The reproduction data detecting method according to claim 2, wherein the data in which 1 is associated with the transition to the states S1, S2 and S4 is used as the reproduction data.