JP2005346759A - Waveform equalizer, information reproduction device, communication device, waveform equalization program, computer readable recording medium on which waveform equalization program is recorded, and waveform equalization method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To realize a waveform equalizer which can stably reduce an error rate, while taking into account of the restrictions imposed on a tap coefficient value. <P>SOLUTION: The waveform equalizer 20a is provided with: an FIR filter 4 for generating a post-equalization signal string by performing the waveform equalization for an input signal string; a Viterbi decoding circuit 5 for detecting a path metric difference between a correct path and an incorrect path in a Viterbi decoding process for the post-equalization signal string; a target value register 9 for setting a target value of the path metric difference; and a tap coefficient updating circuit 10 for performing the adaptation based on the error of the detected path metric difference with respect to the target value. The target value register 9 sets the target value at a value smaller than the ideal path metric difference which is based on the ideal input signal string assumed by the Viterbi decoding. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a waveform equalization apparatus that adaptively equalizes a reproduced signal in a signal reproduction system, a waveform equalization program, a computer-readable recording medium that records the waveform equalization program, a waveform equalization method, and The present invention relates to an information reproducing apparatus and a communication apparatus provided with the waveform equalizer.

  In recent years, a PRML (Partial Response Maximum Likelihood) method has been adopted as a data detection method for realizing a higher recording density of an information recording medium. In PRML, it is necessary to perform waveform equalization in order to bring the reproduced signal waveform reproduced from the recording medium closer to the ideal frequency characteristic assumed in the PR class. Since there are reproduction characteristic fluctuations due to reproduction system characteristic fluctuations, such as servo offsets, adaptive equalization techniques for updating the waveform equalization characteristics adaptively with respect to reproduction characteristic fluctuations are used.

  Previously, the applicant has applied for an adaptive equalization technique that can favorably reduce the error rate in PRML (see Patent Document 1). In this method, equalization coefficients are adapted so that a mean square error of a path metric difference in a Viterbi decoding process with respect to a predetermined target value approaches a minimum value.

  The prior art will be briefly described with reference to FIG. FIG. 11 is a block diagram showing the configuration of the prior art waveform equalization apparatus 50. The waveform equalizer 50 includes an FIR (Finite Impulse Response) filter 51, a Viterbi decoding circuit 52, a specific pattern detection circuit 53, and a tap coefficient update circuit 54. For simplification of explanation, a part of the configuration is omitted.

  The FIR filter 51 performs waveform equalization on the input reproduction signal sequence and outputs an equalized signal sequence. The equalized signal sequence is obtained by a convolution operation between the tap coefficient of the FIR filter 51 and the reproduction signal sequence.

  The Viterbi decoding circuit 52 performs Viterbi decoding of the equalized signal sequence output from the FIR filter 51 based on the assumed PR characteristic, and outputs a decoded bit sequence, and at the same time, the path metrics of the two paths joined in the Viterbi decoding process Calculate and output the difference. This path metric difference is called SAM (Sequenced Amplitude Margin), and is well known, for example, by Non-Patent Document 1.

  The specific pattern detection circuit 53 determines whether or not the decoded bit string decoded by the Viterbi decoding circuit 52 matches a predetermined specific pattern, and transmits a match signal to the tap coefficient update circuit 54 if they match. .

  Each time the coincidence signal is transmitted from the specific pattern detection circuit 53, the tap coefficient update circuit 54 performs a predetermined calculation on the input reproduction signal sequence, path metric difference, and pre-update tap coefficient, thereby updating the post-update tap coefficient. And the tap coefficient of the FIR filter 51 is updated.

  The update amount of the tap coefficient in the above is very small, but the tap coefficient gradually converges to a certain value by repeating this update operation. In the equalized signal equalized by the convergence tap coefficient, the mean square error of the path metric difference with respect to the ideal value is the smallest, and at this time, the error rate of the Viterbi decoded bit is the best. .

As described above, in the prior art waveform equalizer 50, the tap coefficient of the FIR filter 51 is updated so that the mean square error of the path metric difference is minimized, so that the error rate is adaptively minimized. It is possible to do.
JP 2004-63036 A (publication date February 26, 2004) T.Perkins, "A Window Margin Like Procedure for Evaluating PRML Channel Performance"; IEEE Transactions on Magnetics, Vol.31, No2, 1995, p1109-1114

  In the above configuration, the tap coefficient of the FIR filter 51 is determined so as to minimize the mean square error of the path metric difference with respect to the ideal value, and therefore the tap coefficient value itself is not limited. However, when the FIR filter 51 is actually realized by a circuit, the range of tap coefficient values that can be expressed according to the bit width specifications of a register that holds the tap coefficient, an adder used for equalization, a multiplier, and the like. There are constraints on

  For example, when an 8-bit register is used as a register for holding a tap coefficient as a fixed point in increments of 1/16 = 0.0625, the range of tap coefficient values that can be expressed is ± 8, and a value having an absolute value larger than this is required. I can't hold it. In order to express a larger range than this, there is no choice but to increase the step size of the fixed point or increase the number of bits of the register. However, in the former case, there is a problem that an approximation error (rounding error) of the set tap coefficient value becomes large, and as a result, non-optimal equalization is performed, and in the latter case, the circuit scale increases. Occurs. That is, it is considered that there is always a premise that the range of tap coefficient values is finite.

  Therefore, when the tap coefficient that minimizes the mean square error of the path metric difference with respect to the ideal value is found to be larger than 8, the tap coefficient updating circuit 54 uses the tap coefficient for the FIR filter 51. Cannot be set correctly, and equalization characteristics cannot be optimized.

  The present invention has been made in view of the above problems, and its purpose is to stabilize the error rate in consideration of restrictions imposed on values that determine equalization characteristics such as tap coefficient values. It is to be able to reduce.

  A waveform equalization apparatus according to the present invention is a waveform equalization apparatus that adapts an equalization characteristic while waveform equalizing a Viterbi-decodeable input signal sequence, and in order to solve the above problem, the input signal An equalization means for generating an equalized signal sequence by performing the waveform equalization on the sequence, and a path for detecting a path metric difference between a correct path and an error path in a Viterbi decoding process for the equalized signal sequence Metric difference detection means, target value setting means for setting a target value of the path metric difference, and equalization adaptation means for performing the adaptation based on an error with respect to the detected target value of the path metric difference. The target value setting means sets the target value to a value smaller than an ideal path metric difference that is a path metric difference based on an ideal input signal sequence assumed by Viterbi decoding. To have.

  Further, a waveform equalization method according to the present invention is a waveform equalization method for adapting the equalization characteristics while equalizing the waveform of an input signal sequence that can be Viterbi-decoded. An equalization step for generating an equalized signal sequence by performing the waveform equalization on the input signal sequence, and detecting a path metric difference between a correct path and an error path in the Viterbi decoding process for the equalized signal sequence A path metric difference detection step, and an equalization adaptation step for performing the adaptation based on an error of the detected path metric difference with respect to a target value set as a target value of the path metric difference. And the target value is set to a value smaller than an ideal path metric difference that is a path metric difference based on an ideal input signal sequence assumed by Viterbi decoding. .

  In the above configuration and method, the path metric difference between the correct path and the error path in the Viterbi decoding process for the equalized signal sequence is detected, and equalization characteristics are adapted using the path metric difference. As the path metric difference, an ideal path metric difference (ideal path metric difference) is determined, but the actually detected path metric difference has variations with respect to the ideal path metric difference. In order to reduce the error rate in Viterbi decoding, it is only necessary to reduce the variation of the actually detected path metric difference with respect to the ideal path metric difference. To that end, the target value of the path metric difference is set to the ideal path metric difference. It is conceivable to set the difference.

  However, in reality, the target value of the path metric difference is not only set to the ideal path metric difference, but the target value of the path metric difference is set to a value smaller than the ideal path metric difference. It has been found that the same low error rate can be maintained as when the difference is set. Therefore, it is not necessary to fix the target value of the path metric difference to the ideal path metric difference.

  On the other hand, in order to perform waveform equalization, amplification is performed individually for each input signal in the input signal sequence. Usually, the absolute value of the amplification factor (absolute amplification factor) when performing this amplification is actually There is a finite range of possible values. This is because the upper limit value of the absolute amplification factor is determined due to a configuration used for amplification (for example, a register that holds a tap coefficient when an FIR filter is used). Therefore, if the convergence value of the absolute gain due to the adaptation of the equalization characteristics becomes a value that exceeds the above upper limit value, such a value should actually be set as the gain. Therefore, the equalization characteristic cannot be optimized and the error rate is deteriorated. Therefore, it is desirable that the absolute gain has a smaller convergence value.

  Here, it is found that the convergence value of the absolute amplification factor is proportional to the change of the target value of the path metric difference, that is, the convergence value of the absolute amplification factor is reduced by decreasing the target value. It was. Therefore, by reducing the target value, the risk that the convergence value of the absolute gain will reach the upper limit value can be reduced.

  Therefore, in the above configuration and method, the target value of the path metric difference is set to a value smaller than the ideal path metric difference. As a result, it is possible to reduce the risk that the absolute gain reaches the upper limit while maintaining the same low error rate as when the target value of the path metric difference is set to the ideal path metric difference. As a result, the error rate can be stably reduced.

  In the waveform equalizer according to the present invention, it is preferable that the target value setting means sets the target value to a value larger than ½ of the ideal path metric difference. Thereby, the deterioration of the error rate due to setting the target value too small can be suppressed.

  A waveform equalization apparatus according to the present invention is a waveform equalization apparatus that adapts an equalization characteristic while waveform equalizing a Viterbi-decodeable input signal sequence, and in order to solve the above problem, the input signal An equalization means for generating an equalized signal sequence by performing the waveform equalization on the sequence, and a path for detecting a path metric difference between a correct path and an error path in a Viterbi decoding process for the equalized signal sequence Metric difference detection means, target value setting means for setting a target value of the path metric difference, and equalization adaptation means for performing the adaptation based on an error with respect to the detected target value of the path metric difference. The target value setting means changes the target value in accordance with the equalization characteristic.

  Further, the waveform equalization apparatus according to the present invention is a waveform equalization method that adapts the equalization characteristics while waveform equalizing an input signal sequence that can be Viterbi-decoded. An equalization step for generating an equalized signal sequence by performing the waveform equalization on the input signal sequence, and detecting a path metric difference between a correct path and an error path in the Viterbi decoding process for the equalized signal sequence A path metric difference detection step, and an equalization adaptation step for performing the adaptation based on an error of the detected path metric difference with respect to a target value set as a target value of the path metric difference, And a target value update step of changing the target value in accordance with the equalization characteristic.

  In the above configuration and method, the path metric difference between the correct path and the error path in the Viterbi decoding process for the equalized signal sequence is detected, and equalization characteristics are adapted using the path metric difference. As the path metric difference, an ideal path metric difference (ideal path metric difference) is determined, but the actually detected path metric difference has variations with respect to the ideal path metric difference. In order to reduce the error rate in Viterbi decoding, it is only necessary to reduce the variation of the actually detected path metric difference with respect to the ideal path metric difference. To that end, the target value of the path metric difference is set to the ideal path metric difference. It is conceivable to set the difference.

  However, in reality, the target value of the path metric difference is not only set to the ideal path metric difference, but the target value of the path metric difference is set to a value smaller than the ideal path metric difference. It has been found that the same low error rate can be maintained as when the difference is set. Therefore, it is not necessary to fix the target value of the path metric difference to the ideal path metric difference.

  On the other hand, in order to perform waveform equalization, amplification is performed individually for each input signal in the input signal sequence. Usually, the absolute value of the amplification factor (absolute amplification factor) when performing this amplification is actually There is a finite range of possible values. This is because the upper limit value of the absolute amplification factor is determined due to a configuration used for amplification (for example, a register that holds a tap coefficient when an FIR filter is used). Therefore, if the convergence value of the absolute gain due to the adaptation of the equalization characteristics becomes a value that exceeds the above upper limit value, such a value should actually be set as the gain. Therefore, the equalization characteristic cannot be optimized and the error rate is deteriorated. Therefore, it is desirable that the absolute gain has a smaller convergence value.

  Here, it is found that the convergence value of the absolute amplification factor is proportional to the change of the target value of the path metric difference, that is, the convergence value of the absolute amplification factor is reduced by decreasing the target value. It was. Therefore, by reducing the target value, the risk that the convergence value of the absolute gain will reach the upper limit value can be reduced.

  Therefore, in the above configuration and method, the target value is changed according to the equalization characteristic. Thereby, for example, the ideal path metric difference is set as the initial value of the target value, the change of the amplification factor due to the adaptation of the equalization characteristic is monitored, and the absolute amplification factor is likely to reach the above upper limit value. In such a case, it is possible to avoid the absolute gain reaching the upper limit value by changing the target value. As a result, the error rate can be stably reduced.

  In this way, the target value setting means can change the target value to be a value smaller than an ideal path metric difference that is a path metric difference based on an ideal input signal sequence assumed by Viterbi decoding. desirable.

  Further, it is desirable that the target value setting means changes the target value so as to be a value larger than ½ of the ideal path metric difference. Thereby, the deterioration of the error rate due to setting the target value too small can be suppressed.

  The waveform equalization apparatus according to the present invention is the waveform equalization apparatus, wherein the equalization means sequentially associates each input signal of the input signal sequence with a plurality of equalization coefficients, and And a signal sequence after equalization by performing a convolution operation with each input signal associated with each equalization coefficient, and the target value setting means uses the equalization adaptation means, etc. The target value may be changed according to the absolute value of the equalization coefficient during the equalization characteristic adaptation operation. In this case, the equalization coefficient is the amplification factor described above.

  An information reproducing apparatus according to the present invention includes any one of the above-described waveform equalizing apparatuses and reproducing means for reproducing the input signal sequence from an information recording medium.

  In addition, a communication apparatus according to the present invention includes any one of the waveform equalization apparatuses described above and a reception unit that receives the input signal sequence transmitted via a communication path.

  As described above, in the information reproducing apparatus and communication apparatus provided with the waveform equalizer of the present invention, the error rate can be stably suppressed to a low level due to the effect of the waveform equalizer of the present invention.

  The waveform equalization program of the present invention is a waveform equalization program for operating any one of the above waveform equalization apparatuses, and is a program for causing a computer to function as each of the above means.

  With the above configuration, the waveform equalization apparatus can be realized by realizing each means of the waveform equalization apparatus with a computer. Therefore, as the above waveform equalizer, the error rate can be stably kept low.

  In addition, a computer-readable recording medium on which the waveform equalization program of the present invention is recorded records a program for causing a computer to function as each means.

  With the above configuration, the waveform equalization apparatus can be realized on the computer by the waveform equalization program read from the recording medium.

  The waveform equalization apparatus according to the present invention includes an equalization unit that generates an equalized signal sequence by performing the waveform equalization on the input signal sequence, and a correct answer in a Viterbi decoding process for the equalized signal sequence A path metric difference detecting means for detecting a path metric difference between a path and an error path, a target value setting means for setting a target value of the path metric difference, and an error with respect to the detected target value of the path metric difference. Equalization adaptation means for performing the adaptation, wherein the target value setting means determines that the target value is greater than an ideal path metric difference that is a path metric difference based on an ideal input signal sequence assumed by Viterbi decoding. The configuration is set to a small value.

  The waveform equalization method according to the present invention includes an equalization step of generating an equalized signal sequence by performing the waveform equalization on the input signal sequence, and a Viterbi decoding process for the equalized signal sequence A path metric difference detecting step for detecting a path metric difference between a correct path and an error path in the path, and based on an error of the detected path metric difference with respect to a target value set as a target value of the path metric difference An equalization adaptation step for performing the adaptation, and setting the target value to a value smaller than an ideal path metric difference that is a path metric difference based on an ideal input signal sequence assumed by Viterbi decoding It is.

  In the above configuration and method, the target value of the path metric difference is set to a value smaller than the ideal path metric difference. As a result, it is possible to reduce the risk that the absolute gain reaches the upper limit while maintaining the same low error rate as when the target value of the path metric difference is set to the ideal path metric difference. As a result, there is an effect that the error rate can be stably reduced.

  The waveform equalization apparatus according to the present invention includes an equalization unit that generates an equalized signal sequence by performing the waveform equalization on the input signal sequence, and a correct answer in a Viterbi decoding process for the equalized signal sequence A path metric difference detecting means for detecting a path metric difference between a path and an error path, a target value setting means for setting a target value of the path metric difference, and an error with respect to the detected target value of the path metric difference. Equalization adaptation means for performing the adaptation, and the target value setting means is configured to change the target value in accordance with the equalization characteristics.

  Further, the waveform equalization apparatus according to the present invention includes an equalization step of generating an equalized signal sequence by performing the waveform equalization on the input signal sequence, and a Viterbi decoding process for the equalized signal sequence A path metric difference detecting step for detecting a path metric difference between a correct path and an error path in the path, and based on an error of the detected path metric difference with respect to a target value set as a target value of the path metric difference The method includes an equalization adaptation step for performing the adaptation, and a target value update step for changing the target value in accordance with the equalization characteristics.

  In the above configuration and method, the target value is changed according to the equalization characteristic. Thereby, for example, the ideal path metric difference is set as the initial value of the target value, the change of the amplification factor due to the adaptation of the equalization characteristic is monitored, and the absolute amplification factor is likely to reach the above upper limit value. In such a case, it is possible to avoid the absolute gain reaching the upper limit value by changing the target value. As a result, there is an effect that the error rate can be stably reduced.

Embodiment 1
A first embodiment of the present invention will be described with reference to FIGS.

  FIG. 1 is a block diagram showing a configuration of an optical disc reproducing apparatus 20 to which the waveform equalization apparatus and the waveform equalization method of this embodiment are applied. The optical disc reproducing device 20 is a device for reproducing the optical disc 1, and includes an optical pickup 2, an A / D converter 3, and a waveform equalizer 20a. The waveform equalizer 20 a includes an FIR filter 4, a Viterbi decoding circuit 5, delay adjustment circuits 6 and 7, a specific pattern detection circuit 8, a target value register 9, and a tap coefficient update circuit 10.

  The optical disc 1 (information recording medium) has a run length limited code with d = 1 such as a (1, 7) RLL (Run Length Limited) code, that is, the shortest mark length is 2T (T is one channel bit time of a reproduction signal) In the following, the recording mark row of the modulation method is recorded.

  The optical pickup 2 (reproducing means) reproduces an analog reproduction waveform having a reproduction signal sequence from the optical disc 1. The optical pickup 2 includes a semiconductor laser (not shown), various optical components, a photodiode, and the like. The optical pickup 2 condenses the laser beam emitted from the semiconductor laser onto the optical disc 1, reflects it with a recording mark recorded on the optical disc 1, and converts the reflected light into an electrical signal with a photodiode to perform analog reproduction. A waveform (hereinafter simply referred to as “reproduced waveform”) is output.

  The A / D converter 3 performs A / D conversion of the reproduction waveform output from the optical pickup 2 at the timing of the channel frequency clock. The A / D converter 3 then outputs a digital reproduction signal after A / D conversion (hereinafter referred to as “reproduction signal string”).

  The FIR filter 4 (equalizing means) generates an equalized signal sequence by performing waveform equalization on the reproduced signal sequence. The FIR filter 4 has two delay elements of time T, three gain variable amplifiers (gains are c (0, n), c (1, n), c (2, n), respectively), and an adder. It is a 3-tap digital filter with one. Here, gains c (0, n), c (1, n), and c (2, n) are tap coefficients (amplification factor and equalization coefficient), and by changing these values, the FIR filter 4 etc. The conversion characteristics can be changed. Note that the number of taps is set to 3 for simplification of explanation, and in fact, a higher performance filter can be obtained by using a larger number of taps. The FIR filter 4 performs waveform equalization on the reproduced waveform using the tap coefficient, and outputs an equalized signal y (i−1, n). The meaning of “n” will be described later.

  The Viterbi decoding circuit 5 (path metric difference detecting means) outputs the equalized signal y (i−1, n) output from the FIR filter 4 based on the PR (1, 2, 1) characteristic having a waveform interference width of 3T. ) And a decoding bit string b (i) of a recording mark recorded on the optical disc 1 is output, and at the same time, a path metric difference s (n) between two paths joined in the Viterbi decoding process is calculated and output. (Details will be described later).

  The delay adjusting circuits 6 and 7 each of the reproduced signal sequence u (i, n) and the path metric difference s (n) due to the delay corresponding to the path memory length in the Viterbi decoding circuit 5 and the decoded bit sequence b (i). This is to correct the time difference between the two and synchronize.

  The specific pattern detection circuit 8 uses the decoded bit strings b (i-4), b (i-3),..., B (i) decoded by the Viterbi decoding circuit 5 as “00111”, “00011” as specific patterns. , “11000”, or “11100” is determined.

  The target value register 9 (target value setting means) sets the target value ds of the path metric difference s (n), and stores this target value ds. The target value ds is set smaller than an ideal value of a path metric difference s (n) described later.

The tap coefficient update circuit 10 (equalization adapting means), every time the specific pattern detection circuit 8 detects the specific pattern,
c (k, n + 1) = c (k, n) ± [mu] {s (n) -ds} {u (-2-k, n) + 2u (-1-k, n) + u (-k, n )}… (1)
Thus, a new tap coefficient c (k, n + 1) is obtained, and the tap coefficient of the FIR filter 4 is updated. The above-described processing in the tap coefficient updating circuit 10 represents an error of the path metric difference s (n) detected in the Viterbi decoding circuit 5 with respect to the target value ds set by the target value register 9 as a tap coefficient. Is a process of updating the tap coefficient so that the function representing the mean square value approaches the minimum value and performing equalization adaptation. “N” is a value corresponding to the number of times of detection of the specific pattern, and means that the tap coefficient is updated for each detected n-th specific pattern.

  Here, the decoding method by the Viterbi decoding circuit 5 will be described in more detail with reference to FIGS.

  As shown in FIG. 2, the reproduction waveform of an ideal 1T mark having a PR (1, 2, 1) characteristic and free from distortion and noise has a sample level ratio of 1: 2: 1 for each channel clock. become. The reproduction waveform of the 2T or more mark is obtained by superimposing the reproduction waveforms of the 1T mark. For example, the sample level ratio of the 2T mark is 1: 3: 3: 1, the sample level ratio of the 3T mark is 1: 3: 4: 3: 1, and the sample level ratio of the 4T mark is 1: 3: 4: 4. : 3: 1. In this manner, an ideal reproduction waveform can be assumed for an arbitrary bit string, and five levels of 0, 1, 2, 3, and 4 are taken as ideal sample levels. Here, for convenience, if the sample level is normalized so that the maximum amplitude of the sample level is ± 1, the ideal sample level becomes five levels of −1, −0.5, 0, +0.5, and +1. .

  FIG. 3 is a trellis diagram showing Viterbi decoding for realizing data detection by the PRML method. In FIG. 3, S (00), S (01), S (10), and S (11) each represent a state. For example, the state S (01) indicates that the previous bit is 0 and the current bit is 1. . A line connecting the states is called a “branch”, and this branch represents a state transition. For example, a bit string “001” can be represented by a branch of S (00) → S (01). The numerical value next to each branch represents the ideal sample level expected in each state transition. For example, since the branch of S (00) → S (00) represents a bit string “000”, −1 (sample level before normalization is 0) is the ideal sample level. Note that there is no branch of S (01) → S (10) and S (10) → S (01) because there is no bit string of “010” and “101” due to the run length limitation of d = 1. It reflects that.

  In the trellis diagram, a route that passes through each state at each time so that branches are continuous is called a “path”. Considering all paths generated from any state through any state is equivalent to considering all possible bit strings. Therefore, the ideal waveform expected for all paths is compared with the reproduced waveform actually reproduced from the optical disc 1, and the path having the ideal waveform closest to the reproduced waveform, that is, having the smallest Euclidean distance from the reproduced waveform is searched. Then, the most likely maximum likelihood path can be determined as the correct path.

  The procedure of Viterbi decoding using a trellis diagram will be specifically described. At an arbitrary time, two paths join the states S (00) and S (11), and one path is connected to S (01) and S (10). For the states S (00) and S (11) where the two paths merge, if the Euclidean distance between the ideal waveform and the reproduction waveform of each path to be merged is left as a surviving path, at any time A total of four paths remain, one for each of the four states.

  The square of the Euclidean distance between the ideal waveform of the path and the playback waveform is called “path metric”, and the path metric is made up of the branch metric obtained as the square of the difference between the ideal sample level for each branch and the sample level of the playback waveform. Calculated by accumulating for all branches.

  In this way, every time a sample value of a reproduction waveform, that is, a reproduction signal is input, the procedure for determining the surviving path by comparing the magnitudes of the path metrics of the two paths joining the same state is repeated. As large paths are deceived, the surviving paths converge to a single path with the smallest path metric. By making this a correct path, the data bit string recorded on the optical disc 1 is correctly reproduced.

  The number of state transitions up to the time when the correct answer path is determined and the decoded bit string is output with respect to the input time of the reproduction signal is referred to as “path memory length”. In general, the path memory length is long enough for the surviving paths to converge to one.

  Here, considering the conditions under which Viterbi decoding is performed correctly, the path metric of the correct path is determined in the process of determining the surviving path at each time in order for the path that converges to one finally becomes the correct path. Must be smaller than the path metric of the other path that is the error path. Therefore, by looking at the path metric difference, which is the difference between the path metrics between the two paths that face the survival and confront each other, it is possible to determine how likely the path is likely to cause an error. This path metric difference is the definition of SAM.

  In order to prevent an error in Viterbi decoding, the path metric difference obtained by subtracting the path metric of the correct path from the path metric of the error path needs to be larger than 0, and the error is less likely to occur as the path metric difference is larger. become. The Viterbi decoding circuit 5 calculates this path metric difference and outputs it as s (n).

  Now, the reproducing operation by the optical disk reproducing apparatus 20 having the above-described configuration shown in FIG. 1 will be described as follows.

  First, the optical pickup 2 emits a light beam onto the optical disc 1, whereby a reproduction waveform of a recording mark recorded on the optical disc 1 is output from the optical pickup 2. This reproduced waveform is converted into a reproduced signal sequence u (i, n) by the A / D converter 3. When the reproduction signal sequence u (i, n) is input to the FIR filter 4, waveform equalization processing is performed by the FIR filter 4 and an equalized signal y (i-1, n) is output. The equalized signal y (i−1, n) is an equalized signal corresponding to the reproduction signal u (i−1, n) (the signal before and after equalization by matching the subscript “i−1”). Represents the corresponding relationship). The equalized signal y (i−1, n) is obtained by a convolution operation between the tap coefficient c (k, n) and the reproduction signal sequence u (i−k, n).

  That is, the FIR filter 4 sequentially corresponds each reproduction signal of the reproduction signal sequence u (ik, n) to each tap coefficient for a plurality of tap coefficients c (k, n) (k = 0,1,2). In addition, an equalized signal sequence y (i−1, n) is generated by performing a convolution operation between each tap coefficient and each input signal associated with each tap coefficient.

  When the equalized signal sequence y (i−1, n) is input, the Viterbi decoding circuit 5 obtains and outputs the path metric difference s (n) as described above, and performs Viterbi decoding. The decoded bit string b (i) obtained as a result is output. That is, the Viterbi decoding circuit 5 generates a decoded bit string b (i) that is a decoding result of the reproduction signal string u (i, n). In addition, the Viterbi decoding circuit 5 performs path metrics between a correct path determined as a surviving path and an error path that confronts the correct path in the Viterbi decoding process based on the equalized signal sequence y (i−1, n). The difference s (n) is detected.

  Here, the delay adjusting circuits 6 and 7 perform the reproduction signal sequence u (i, n) and the path metric difference s (n) due to the delay corresponding to the time of the path memory length in the Viterbi decoding, and the decoded bit sequence b (i). The time difference is corrected to produce an effect of synchronization.

  The specific pattern detection circuit 8 includes “00111”, “00011”, “11000”, “11100” in which the decoded bit strings b (i-4), b (i-3),..., B (i) are specific patterns. It is determined whether or not they match, and if they match, a match signal is transmitted to the tap coefficient update circuit 10. Although the specific pattern will be described later, when an ideal waveform signal sequence that is an ideal waveform for Viterbi decoding is assumed, it is a bit sequence pattern that minimizes the path metric difference based on this ideal waveform signal sequence. . The specific pattern detection circuit 8 is a circuit that detects such a specific pattern from the decoded bit string b (i).

  When the coincidence signal is transmitted from the specific pattern detection circuit 8, the tap coefficient update circuit 10 reproduces the error {s (n) −ds} of the path metric difference s (n) with respect to the target value ds from the target value register 9 and the reproduction. Tap by the product of the first order polynomial {u (-2-k, n) + 2u (-1-k, n) + u (-k, n)} with a predetermined weight applied to the signal sequence u (ik, n) The coefficient c (k, n) is corrected. That is, the tap coefficient update circuit 10 determines the error of the FIR filter 4 based on the error {s (n) −ds} of the path metric difference s (n) actually detected by the Viterbi decoding circuit 5 with respect to the target value ds. The equalization characteristic is adapted. Specifically, the tap coefficient update circuit 10 calculates the above-described equation (1).

  Here, the correspondence between the path metric difference s (n) and the reproduced signal sequence u (i−k, n) will be described with reference to FIG. When “00111” is detected as the nth decoded bit string that matches the specific pattern, it is assumed that the reproduced waveform reproduced from the recording mark corresponding to this decoded bit string is as shown in FIG. Let u (-4, n), u (-3, n), u (-2, n), u (-1, n), u (0, n).

  In the above equation (1), μ is a minute value for controlling the responsiveness, and is called a step gain. If μ is set to an appropriate value, the tap coefficient c (k, i) converges to a predetermined value while the update operation is repeated. The sign (±) to be applied to the coefficient correction term (the second term on the right side of equation (1)) is determined by the specific pattern detected by the specific pattern detection circuit 8, and is negative in the case of “00111” or “11100”. In the case of “11000” or “00011”, it is positive.

In the equalized signal equalized by the FIR filter 4 by the tap coefficient (convergent tap coefficient) finally converged by repeating the tap coefficient updating operation according to the above equation (1), the target of the path metric difference s (n) The mean square error E [{s (n) −ds} ² ] (E [] is an expected value operator) with respect to the value ds is minimized. Since this reason is disclosed in Patent Document 1 (Japanese Patent Laid-Open No. 2004-63036), detailed description thereof is omitted. Here, in the invention disclosed in Patent Document 1, an ideal value of the path metric difference s (n) is set as the target value ds, and thereby an error rate (bit error) of a decoded bit decoded by Viterbi decoding. It can be theoretically explained that the rate is the best.

On the other hand, in the present embodiment, the target value ds is set to a value smaller than the ideal value of the path metric difference s (n). Before describing the present embodiment, first, the target value ds is set to the ideal value of the path metric difference s (n), and the mean square error E [{s (n) −ds} ² ] is minimized. The reason will be described in detail.

  FIG. 5A is a graph showing a path metric difference histogram obtained for an ideal waveform having no noise assumed in the PR (1, 2, 1) characteristic based on the bit pattern of the (1, 7) RLL code. FIG. 6 is a table showing the correspondence between each ideal value of the path metric difference and the bit pattern corresponding to each ideal value. From FIG. 5A, the ideal value of the path metric difference is 1.5, 2.5, 3.5, 4.5, 5, 6, 7, 8, 9,... (PR (1, 2, When the maximum amplitude of the sample level of 1) is normalized to ± 1), it can be seen that a plurality of discrete values are taken. The ideal value takes various values because the path metric difference of the error path starting from the same state as the correct path corresponding to the ideal waveform in the trellis diagram and joining the same state differs depending on the bit pattern. ing. In addition, the frequency of each ideal value is different because the number of types of bit patterns taking each ideal value is different as shown in FIG. 6, and the appearance of each bit pattern in the bit pattern of (1, 7) RLL code. This is because the frequency is different.

  On the other hand, when the histogram of the path metric difference obtained by actually reproducing the bit pattern of the (1,7) RLL code recorded on the optical disk is examined, the path is centered on each ideal value as shown in FIG. The distribution has a metric difference and a distribution shape in which a plurality of distributions overlap. This is caused by various noises on the reproduction signal.

  From this, it can be seen that the error rate can be reduced satisfactorily by determining the equalization characteristic so that the variation of the path metric difference of the reproduced signal with respect to each ideal value becomes small. Furthermore, considering that the path metric difference needs to be greater than 0 in order to prevent an error in Viterbi decoding, equalization is performed so that only the path metric difference of the ideal value 1.5 that is most likely to cause an error is reduced in variation. If the characteristics are determined, the error rate can be optimized.

The specific pattern detection circuit 8 detects bit patterns “00111”, “00011”, “11000”, “11100” such that the ideal value of the path metric difference is 1.5 as shown in FIG. It operates so as to minimize the mean square error E [{s (n) −ds} ² ] with respect to the target value ds only for the path metric difference s (n) of .5.

  Next, the case where the target value ds is set to a value smaller than the ideal value of the path metric difference s (n) will be considered in detail. In order to prevent an error in Viterbi decoding, it is desirable that the path metric difference be a large value. Therefore, if the target value ds is set to a small value, the entire distribution of the path metric difference becomes small (the average value approaches 0). There is a concern that the error rate deteriorates because of convergence to such equalization characteristics.

  However, from the results of experiments performed using an actual optical disc reproducing apparatus, it was confirmed that the error rate is not necessarily deteriorated by making the target value ds smaller than the ideal value of the path metric difference s (n).

  FIG. 7A shows the experimental results, and is a graph showing the relationship between the target value ds and the error rate obtained by performing equalization with the convergence tap coefficient. It can be seen that even if the target value ds is set to a value smaller than the ideal value 1.5, if the target value ds is 0.7 or more, the obtained error rate is substantially constant. Considering this reason, when the target value ds is reduced, the average value of the distribution of the path metric difference approaches 0, but the variation (dispersion) of the distribution is reduced accordingly, and as a result, the path metric difference <0. (This corresponds to the error rate) is almost unchanged.

  However, if the target value ds is made too small (less than 0.7 in the result of FIG. 7A), the error rate deteriorates. The reason for this is that when the target value ds is about ½ or less of the ideal ideal value of the path metric difference, the dispersion of the distribution cannot be further reduced and the error rate is deteriorated. It is done.

  From the above, it can be said that even if the target value ds is set to a value that is somewhat smaller than the ideal value, it is guaranteed that the equalization capability of the tap coefficient update circuit 10 is not adversely affected.

  FIG. 7B is a graph showing the relationship between the target value ds in the same experiment as FIG. 7A and the convergence tap coefficient. The number of taps of the FIR filter used in this experiment is 9 taps, and c (0), c (1),..., C (8) are tap coefficients c (k, n) (k = 0 to 8), respectively. ) Is a convergence tap coefficient.

  From FIG. 7B, it can be confirmed that the target value ds is proportional to the convergence tap coefficient. Accordingly, by setting the target value ds to be smaller than the ideal value 1.5 of the path metric difference s (n), it is possible to suppress the absolute value of the convergence tap coefficient to be small.

  Since the convergence tap coefficient can be reduced in this way, the risk that the convergence tap coefficient overflows from the capacity (number of bits) of the register that holds the tap coefficient (hereinafter referred to as “tap coefficient register”) is reduced. can do. Although not shown in FIG. 1, the tap coefficient register is provided in, for example, the FIR filter 4.

  A more detailed configuration of the FIR filter 4 will be described with reference to FIG. The FIR filter 4 includes tap coefficient registers 41a, 41b, and 41c, multipliers 42a, 42b, and 42c, and an adder 43. The new tap coefficients c (k, n + 1) (k = 0,1,2) calculated by the tap coefficient updating circuit 10 are respectively stored in the tap coefficient registers 41a, 41b, and 41c of the FIR filter 4. (k, n) (k = 0,1,2) are held and output to the corresponding multipliers 42a, 42b, and 42c. In the multipliers 42a, 42b, and 42c, tap coefficients c (k, n) (k = 0, 1, 2) held in the corresponding tap coefficient registers 41a, 41b, and 41c and the reproduction signal u (k, n) Multiply by (k = 0, 1, 2) and output to the adder 43. The adder 43 adds the outputs from the multipliers 42a, 42b, and 42c to generate and output an equalized signal y (i-1, n).

  The tap coefficient registers 41a, 41b, and 41c may be provided separately from the FIR filter 4.

  Here, consider a case where each tap coefficient register is constituted by an 8-bit register and is used as a fixed point in increments of 1/16 = 0.0625. In this case, the range of tap coefficient values that can be held in each tap coefficient register is ± 8. In order to correspond to the experimental results shown in FIGS. 7A and 7B, here, unlike the FIR filter 4 shown in FIG. 8, an FIR filter having 9 taps is assumed.

  When the target value ds is set to the ideal value 1.5 as in the conventional example, the absolute values of the convergence tap coefficients c (2) and c (3) exceed 8 from FIG. The adaptive operation cannot be performed and the error rate is deteriorated. On the other hand, by setting the target value ds to a value smaller than the ideal value 1.5, for example, 0.8, all the convergence tap coefficients are within ± 8, and the adaptation can be performed correctly. Even in this case, as shown in FIG. 7A, the error rate can be satisfactorily reduced as in the case where the target value ds is set to the ideal value 1.5.

  In this way, by setting the target value ds to a value smaller than the ideal value, the convergence tap coefficient is set to the capacity of the tap coefficient register while maintaining the same low error rate as when the target value ds is set to the ideal value. Can be reduced, and as a result, the risk that the error rate is deteriorated can be reduced.

  As described above, the waveform equalization apparatus 20a and the waveform equalization method executed in the waveform equalization apparatus 20a according to the present embodiment can change the absolute value of the convergence tap coefficient by changing the target value of the path metric difference. The value can be controlled. In particular, by setting the target value to a value smaller than the ideal value of the path metric difference, the absolute value of the convergence tap coefficient can be kept small compared to the case where the target value is set to the same value as the ideal value. In addition, it is possible to achieve a high error rate equivalent to the case where the target value is set to the same value as the ideal value.

[Embodiment 2]
In the first embodiment, the target value ds is set in advance to a value smaller than the ideal value of the path metric difference. However, the tap coefficient update circuit is caused by a change in the characteristics of the reproduction waveform caused by various state changes in the reproduction system. It is difficult to predict how large the convergence tap coefficient derived by 10 can be.

  On the other hand, as shown in FIG. 7A, if the target value ds is too small, the error rate may be deteriorated. Therefore, it can be said that it is basically desirable to set the ideal value of the path metric difference as it is as the target value ds.

  Therefore, in this embodiment, a configuration is adopted in which the value of the updated tap coefficient is monitored and the target value ds is gradually reduced when the tap coefficient is likely to exceed the allowable range.

  FIG. 9 is a configuration diagram of an optical disc playback apparatus 21 using the waveform equalization apparatus and the waveform equalization method according to the second embodiment of the present invention. Regarding the optical disc playback apparatus 21, components having the same functions as those of the optical disc playback apparatus 20 in the first embodiment are denoted by the same reference numerals and description thereof is omitted.

  The optical disc playback device 21 is different from the optical disc playback device 20 in that the waveform equalizer 21a of the optical disc playback device 21 uses the target value update circuit 11 instead of the target value register 9 in the waveform equalizer 20a of the optical disc playback device 20. It is a point that is provided. Note that the settable range of the tap coefficient c (k, n) (k = 0,1,2) of the FIR filter 4 is limited to ± 8 depending on the circuit specifications.

  The target value update circuit 11 (target value setting means) monitors the value of the tap coefficient c (k, n) currently set in the FIR filter 4 and determines whether or not it is within a predetermined range. This predetermined range is a range slightly smaller than the settable range of the tap coefficient. In this embodiment, the predetermined range is set to ± 7, for example, with respect to the tap coefficient settable range ± 8. The target value update circuit 11 fixes the target value ds set in the tap coefficient update circuit 10 if the tap coefficient is within a predetermined range ± 7. On the other hand, when it is determined that the tap coefficient exceeds the predetermined range ± 7, the target value ds is updated to a slightly smaller value. For that purpose, for example, a method of subtracting 0.1 from the target value ds at that time, or multiplying the target value ds by 0.9 is conceivable. In the present embodiment, description will be made assuming a method of subtracting 0.1. However, another method may be used as a method of updating the target value ds to a slightly smaller value.

  The flowchart of FIG. 10 shows an example of a processing procedure for updating the target value ds as described above.

  First, the target value update circuit 11 sets the target value ds to an initial value 1.5, that is, an ideal value (step S1). Subsequently, reproduction of the optical disc 1 is started (step S2). Thereafter, every time a specific pattern is detected by the specific pattern detection circuit 8, the tap coefficient c (k, n) of the FIR filter 4 is updated toward the convergence tap coefficient (step S3). This coefficient update process (step S3) and a target value ds update process (steps S5 and S6), which will be described later, are repeated until the reproduction of the predetermined range of the optical disc 1 is completed (step S4).

  When the tap coefficient c (k, n) is updated, the target value update circuit 11 determines whether or not the value of each tap coefficient c (k, n) is within a predetermined range ± 7 (step S5). ). If it is within the predetermined range ± 7, the process returns to step S3 to continue the tap coefficient updating operation. On the other hand, if the predetermined range ± 7 is exceeded, 0.1 is subtracted from the target value ds to update the target value ds (step S6), and the process returns to step S3. Then, as the target value ds becomes smaller, the convergence tap coefficient of each tap coefficient c (k, n) by the tap coefficient update circuit 10 also becomes smaller, so that an increase in the tap coefficient c (k, n) is suppressed. The update is continued while staying within the predetermined range ± 7.

  Note that the determination operation in step S5 may be performed for each tap coefficient update operation in step S3, but the determination operation in step S5 may be performed each time the tap coefficient update operation is performed a plurality of times.

  In this way, the value of the tap coefficient to be updated is monitored, and when the tap coefficient is about to exceed the allowable range, the target value is gradually reduced, so that the first embodiment is reduced. In addition to the effect, it is possible to reduce the risk that the error rate will deteriorate due to the target value ds being too small.

  Each block of the waveform equalization apparatuses 20a and 21a described in the first and second embodiments may be configured by hardware logic, or may be realized by software using a computer as follows. That is, the waveform equalization apparatuses 20a and 21a have a central processing unit (CPU) that executes instructions of a waveform equalization program that implements each function of the apparatus, a read only memory (ROM) that stores the program, and a program that stores the program. It can also be realized by a computer having a RAM (random access memory) to be developed, a storage device (recording medium) such as a memory for storing the program and various data.

  That is, an object of the present invention is to provide a computer with a recording medium in which a program code (execution format program, intermediate code program, source program) of a waveform equalization program, which is software that realizes the functions described above, is recorded in a computer-readable manner This can also be achieved by supplying and executing the program code recorded in the recording medium by the computer.

  Thus, in this specification, the means does not necessarily mean physical means, but includes cases where the functions of the means are realized by software. Further, the function of one means may be realized by two or more physical means, or the functions of two or more means may be realized by one physical means.

  In the first and second embodiments, since the ideal sample level is normalized to ± 1 in the reproduction system combining the PR (1, 2, 1) characteristic and the (1, 7) RLL code, the path metric difference is reduced. Although the minimum value of the ideal value is 1.5, the gist of the present invention is not dependent on the PR characteristics and the coding method.

  Also in the first and second embodiments, the optical disk reproducing apparatus has been described as an example of the information reproducing apparatus. However, the present invention is not limited to this, and the same effect can be achieved in an apparatus that performs PRML signal reproduction. That is, the present invention can be applied not only to a magnetic recording / reproducing apparatus such as a hard disk apparatus or a magnetic tape apparatus as another information reproducing apparatus but also to a communication apparatus such as a communication data receiving apparatus. In the communication apparatus to which the present invention is applied, a communication waveform transmitted via a wireless or wired communication path is received instead of the optical pickup 2 in the optical disk reproducing apparatus 20 in FIG. 1 and the optical disk reproducing apparatus 21 in FIG. A receiver (receiving means) may be provided.

  The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope shown in the claims, and embodiments obtained by appropriately combining technical means disclosed in different embodiments. Of course, it is included in the technical scope of the present invention.

  According to the waveform equalization apparatus, the waveform equalization method, and the waveform equalization program of the present invention, it is possible to adapt the waveform equalization with high accuracy, so that the signal reproduction of an optical disk reproduction apparatus or the like that reproduces a PRML signal is possible. Applicable to devices and communication devices.

1 is a block diagram showing a configuration of an optical disc playback apparatus according to a first embodiment of the present invention. It is a schematic diagram showing the relationship between a data bit string, a recording mark, a reproduction waveform according to PR (1, 2, 1) characteristics, and its sample level. It is a schematic diagram which shows a trellis diagram. FIG. 10 is a schematic diagram for explaining a correspondence relationship between a decoded bit string “00111”, a recording mark, a reproduction waveform, and its sample level. (A) is a histogram of the path metric difference of the ideal waveform, and (b) is a histogram of the path metric difference of the actual reproduction signal. It is a list of bit patterns corresponding to respective ideal values of path metric differences in a reproduction system in which PR (1, 2, 1) ML and (1, 7) RLL are combined. (A) is a graph showing the relationship between the target value of the path metric difference and the error rate, and (b) is a graph showing the relationship between the target value of the path metric difference and the convergence tap coefficient value. FIG. 2 is a block diagram showing in more detail the configuration of an FIR filter in the optical disc reproducing apparatus of FIG. It is a block diagram which shows the structure of the optical disk reproducing | regenerating apparatus concerning the 2nd Embodiment of this invention. 10 is a flowchart showing a flow of processing for updating a target value of a path metric difference in the optical disc playback apparatus of FIG. 9. It is a block diagram which shows the structure of the conventional waveform equalization apparatus.

Explanation of symbols

1. Optical disc (information recording medium)
2 Optical pickup (reproducing means)
3 A / D converter 4 FIR filter (equalization means)
5 Viterbi decoding circuit (path metric difference detection means)
6 Path memory length delay element 7 Path memory length delay element 8 Specific pattern detection circuit (pattern detection means)
9 Target value register (Target value setting means)
10 Tap coefficient update circuit (equalization adaptation means)
11 Target value update circuit (Target value setting means)
20 Optical disk playback device (information playback device)
20a Waveform equalizing device 21 Optical disc reproducing device (information reproducing device)
21a Waveform equalizer

Claims (12)

In the waveform equalization apparatus that adapts the equalization characteristics while equalizing the waveform of the Viterbi-decoded input signal sequence,
Equalization means for generating a signal sequence after equalization by performing the waveform equalization on the input signal sequence;
Path metric difference detection means for detecting a path metric difference between a correct path and an error path in a Viterbi decoding process for the equalized signal sequence;
Target value setting means for setting a target value of the path metric difference;
Equalization adaptation means for performing the adaptation based on an error with respect to a target value of the detected path metric difference,
The target value setting means sets the target value to a value smaller than an ideal path metric difference that is a path metric difference based on an ideal input signal sequence assumed by Viterbi decoding. .   3. The waveform equalization apparatus according to claim 2, wherein the target value setting means sets the target value to a value larger than ½ of the ideal path metric difference. In the waveform equalization apparatus that adapts the equalization characteristics while equalizing the waveform of the Viterbi-decoded input signal sequence,
Equalization means for generating a signal sequence after equalization by performing the waveform equalization on the input signal sequence;
Path metric difference detection means for detecting a path metric difference between a correct path and an error path in a Viterbi decoding process for the equalized signal sequence;
Target value setting means for setting a target value of the path metric difference;
Equalization adaptation means for performing the adaptation based on an error with respect to a target value of the detected path metric difference,
The target value setting means changes the target value according to the equalization characteristic.   The target value setting means changes the target value to be a value smaller than an ideal path metric difference that is a path metric difference based on an ideal input signal sequence assumed by Viterbi decoding. Item 4. The waveform equalizer according to Item 3.   5. The waveform equalization apparatus according to claim 4, wherein the target value setting means changes the target value so as to be a value larger than ½ of the ideal path metric difference. The equalization means sequentially associates each input signal of the input signal sequence with a plurality of equalization coefficients, and each equalization coefficient and each input signal associated with the equalization coefficient. A signal sequence after equalization is generated by performing a convolution operation,
5. The target value setting means changes the target value according to an absolute value of the equalization coefficient during an equalization characteristic adaptation operation by the equalization adaptation means. Or the waveform equalization apparatus of 5. The waveform equalizer according to any one of claims 1 to 6,
An information reproducing apparatus comprising: reproducing means for reproducing the input signal sequence from an information recording medium. The waveform equalizer according to any one of claims 1 to 6,
And a receiving means for receiving the input signal string transmitted via the communication path.   A waveform equalization program for operating the waveform equalization apparatus according to any one of claims 1 to 6, wherein the waveform equalization program causes a computer to function as each of the means.   A computer-readable recording medium on which the waveform equalization program according to claim 9 is recorded. In the waveform equalization method for adapting the equalization characteristics while equalizing the waveform of the Viterbi-decodeable input signal sequence,
An equalization step of generating a post-equalization signal sequence by performing the waveform equalization on the input signal sequence;
A path metric difference detecting step for detecting a path metric difference between a correct path and an error path in a Viterbi decoding process for the equalized signal sequence;
An equalization adaptation step for performing the adaptation based on an error of the detected path metric difference with respect to a target value set as a target value of the path metric difference,
A waveform equalization method, wherein the target value is set to a value smaller than an ideal path metric difference that is a path metric difference based on an ideal input signal sequence assumed by Viterbi decoding. In the waveform equalization method for adapting the equalization characteristics while equalizing the waveform of the input signal sequence that can be Viterbi-decoded,
An equalization step of generating a post-equalization signal sequence by performing the waveform equalization on the input signal sequence;
A path metric difference detecting step for detecting a path metric difference between a correct path and an error path in a Viterbi decoding process for the equalized signal sequence;
An equalization adaptation step for performing the adaptation based on an error of the detected path metric difference with respect to a target value set as a target value of the path metric difference;
And a target value update step of changing the target value in accordance with the equalization characteristic.

Images