JP2010020813A - Information decoder and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To suitably decode encoded data by relatively simple processing in an information decoder. <P>SOLUTION: The information decoder includes a deviation detection means (220) for detecting two-dimensional position deviation between input data and reference data to be input, a correction amount generation means (230) for generating a correction amount to correct the detected deviation, a coefficient generation means (240) for generating a weight coefficient to be used for weighting when the input data are demodulated or decoded according to the level of the correction amount, and a demodulation/decoding means (250) for demodulating/decoding the input data by using the weight coefficient. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technical field of an information decoding apparatus and method for decoding information encoded by a code having error correction capability such as an LDPC (Low density parity check) code.

  As this type of information decoding apparatus, there is an apparatus that decodes data more preferably using a weighting factor when decoding encoded data. For example, Patent Document 1 discloses a technique for obtaining a weighting factor based on a probability value obtained by a probability calculating means when decoding data encoded using an LDPC code. Patent Document 2 also discloses a technique of performing weighting on likelihood values that are input for LDPC decoding.

  On the other hand, data encoded by an LDPC code or the like is two-dimensionally modulated and recorded on a recording medium such as a holographic memory. When reproducing two-dimensionally modulated data, the detected two-dimensional modulation data may be distorted. For example, Patent Document 3 discloses a technique of detecting a border (that is, both ends of a recording area boundary) from a luminance value and performing geometric correction in order to correct such distortion.

JP 2005-347883 A JP 2006-13714 A Japanese Patent Laying-Open No. 2005-302282

  However, in the technologies according to Patent Documents 1 and 2 described above, since the LDPC decoding itself is iterative decoding, a temporary probability value as a decoding result is used as a weight amount (so-called feedback configuration). There is a risk that the throughput will decrease. That is, the information decoding apparatus according to the above-described technique has a technical problem that the decoding process becomes complicated.

  The present invention has been made in view of the above-described problems, for example, and it is an object of the present invention to provide an information decoding apparatus and method capable of suitably decoding encoded data with relatively simple processing. To do.

  In order to solve the above-mentioned problem, an information decoding apparatus according to the present invention is an information decoding apparatus that demodulates and decodes data that is encoded with a code having error correction capability and that is two-dimensionally modulated. A deviation detecting means for detecting a two-dimensional positional deviation from the reference data to be inputted, a correction amount generating means for generating a correction amount for correcting the detected deviation, and a magnitude of the correction amount In accordance therewith, weighting factor generating means for generating a weighting factor used for weighting when demodulating and decoding the input data, and demodulating and decoding means for demodulating and decoding the input data using the weighting factor With.

  According to the information decoding apparatus of the present invention, during the operation, data encoded by a code having error correction capability and two-dimensionally modulated are demodulated and decoded. The code having error correction capability is, for example, an LDPC code, and the encoded data is two-dimensionally modulated and recorded in a recording medium such as a holographic memory. When reproducing such data, image data that has been two-dimensionally modulated by a light receiving element such as a line sensor or a CCD (Charge Coupled Device) is read, and then demodulated and decoded.

  In the present invention, first, a shift of the position between the input data and the reference data to be input is detected by the shift detection means. That is, it is detected how much the input data is deviated from the reference data. The term “deviation” here means two-dimensional positional deviation such as enlargement / reduction, rotation, positional deviation, distortion, etc., compared with recorded image data that has been two-dimensionally modulated and its reproduced image data. is doing. The “reference data” is data serving as a reference for detecting a deviation, and is typically data indicating coordinates serving as a reference for data to be input.

  When a shift of the input data is detected, a correction amount for correcting the detected shift is generated in the correction amount generation means. That is, a correction amount for reducing the difference between the input data and the reference data is generated. The correction amount is typically indicated by the difference in coordinates between the input data and the reference data. In addition, a plurality of pixels are generated for each pixel of the input two-dimensional modulation data.

  When the correction amount is generated, the weighting coefficient generation unit generates a weighting coefficient used for weighting when demodulating and decoding data. The weighting coefficient is typically a coefficient having a value of 1 or less, and can be weighted by multiplication.

  Particularly in the present invention, the weighting coefficient is generated according to the magnitude of the correction amount. Specifically, the larger the correction amount, the smaller the weighting factor is generated. If the correction amount is generated as a relatively large value, it can be considered that the data corresponding to the correction amount has a large deviation from the reference data. On the other hand, if the correction amount is generated as a relatively small value, it can be considered that the data corresponding to the correction amount has a small deviation from the reference data. From the above results, by generating a weighting coefficient that decreases as the correction amount increases, the weight for data with a large deviation can be reduced, and the weight for data with a small deviation can be increased.

  The demodulation / decoding means demodulates and decodes the input data using the above-described weighting factor. For demodulation and decoding, for example, a likelihood value is first output as a result of demodulation, and decoding using an error correction code is performed based on the likelihood value. The weighting factor may be used for processing for obtaining a likelihood value in demodulation, or may be used in decoding together with the obtained likelihood value. In other words, the demodulation / decoding means performs at least one of the demodulation and decoding processes using the weighting coefficient.

  As described above, since the weighting factor is generated corresponding to the correction amount (that is, the deviation of the input data), as a result, the error correction capability at the time of decoding is increased, and the bit error rate (bit Error Rate: bER) can be improved. Further, as described above, the correction amount can be obtained relatively easily from the deviation of the input data, so that the processing can be prevented from becoming complicated. Furthermore, the bit error rate can be more effectively improved by correcting the deviation of the actually input data using the correction amount.

  As described above, according to the information decoding apparatus of the present invention, it is possible to suitably decode encoded data by a relatively simple process.

  In one aspect of the information decoding apparatus of the present invention, the demodulating / decoding means includes demodulating means for demodulating the input data, and decoding means for decoding the demodulated data using the weighting factor.

  According to this aspect, the demodulating / decoding means includes the demodulating means and the decoding means, and during the operation of the apparatus, the demodulating means and the decoding means perform processing for demodulating and decoding input data, respectively. Here, particularly in this aspect, the weighting coefficient is used when the decoding unit decodes the data demodulated by the demodulation unit.

  More specifically, the input data is first input to the demodulation unit, and the likelihood value is output as a demodulation result from the demodulation unit. The likelihood value and the weight coefficient output from the demodulation means are input to the decoding means. In the decoding means, the data is decoded using the likelihood value multiplied by the weight coefficient. Therefore, the data to be decoded is data affected by the weighting factor. Thus, in this aspect, since the weighting coefficient is used for decoding, the error correction capability is reliably improved. Therefore, it is possible to improve the bit error rate.

  In an aspect in which the decoding unit uses a weighting factor, the weighting factor generation unit may be configured to generate the weighting factor for each modulation symbol in the two-dimensional modulation that is an output unit of the demodulation unit. .

  If comprised in this way, a weighting coefficient will be produced | generated for every modulation symbol in the two-dimensional modulation which is an output unit of a demodulation means. That is, a weighting factor is generated so as to correspond to the output of the demodulating means. Therefore, at the time of decoding, each of the likelihood values output from the demodulating means can be reliably multiplied by the weight coefficient. Therefore, it is possible to improve the error correction capability and improve the bit error rate.

  In another aspect of the information decoding apparatus of the present invention, the demodulating and decoding means includes a demodulating means for demodulating the input data using the weighting factor, and a decoding means for decoding the demodulated data. .

  According to this aspect, the demodulating / decoding means includes the demodulating means and the decoding means, and during the operation of the apparatus, the demodulating means and the decoding means perform processing for demodulating and decoding input data, respectively. Here, in this embodiment, in particular, the weighting coefficient is used when demodulating the data input in the demodulating means.

  More specifically, a weighting factor is input to the demodulating means together with the input data. The demodulating means obtains a likelihood value from the data multiplied by the weighting coefficient. The likelihood value here is a weighted likelihood value because it is obtained from the data multiplied by the weighting coefficient. The weighting likelihood value output from the demodulation unit is input to the decoding unit, and the data is decoded. Thus, in this aspect, since the weighting coefficient is used for demodulation, the error correction capability in decoding using the demodulation result is reliably improved. Therefore, it is possible to improve the bit error rate.

  In the aspect in which the demodulating unit uses a weighting factor, the weighting factor generating unit may be configured to generate the weighting factor for each pixel in the two-dimensional modulation that is an input unit of the demodulating unit.

  If comprised in this way, a weighting coefficient will be produced | generated for every pixel in the two-dimensional modulation which is an input unit of a demodulation means. That is, a weighting factor is generated so as to correspond to the input of the demodulating means. Therefore, at the time of demodulation, each input data can be reliably multiplied by a weighting factor. Therefore, it is possible to improve the error correction capability and improve the bit error rate.

  In another aspect of the information decoding apparatus of the present invention, the weight coefficient generation means selects one process based on a predetermined threshold from a plurality of processes for generating the weight coefficient, and the process performs the Generate a weighting factor.

  According to this aspect, the weight coefficient generation means has a plurality of processes for generating the weight coefficient, and selects one process from the processes to generate the weight coefficient. For example, the weight coefficient generation means has a plurality of mathematical formulas, coefficients, constants, etc. for generating the weight coefficient, and generates a weight coefficient by selecting one of them and using it.

  Which process is selected is determined by a predetermined threshold set in advance. The predetermined threshold is a threshold for a parameter that affects the bit error rate, such as a correction amount, for example, and a weighting factor that can effectively reduce the bit error rate is appropriately generated in the device design stage. Is set. Specifically, the predetermined threshold value can be set by experimentally determining the relationship between the bit error rate and the parameter for the threshold value.

  As described above, by selecting an appropriate process from a plurality of processes when generating the weighting factor, the bit error rate can be effectively improved.

  In another aspect of the information decoding apparatus of the present invention, the weighting factor generating means generates the weighting factor when the correction amount is within a predetermined range.

  According to this aspect, the weight coefficient generation means determines whether or not the correction amount is within a predetermined range before generating the weight coefficient. Then, a weighting coefficient is generated when the correction amount is within a predetermined range. If the correction amount is not within the predetermined range, no correction amount is generated. Therefore, weighting coefficients are not used for demodulation and decoding in the demodulation and decoding means. Here, the “predetermined range” is a range in which the effect of improving the bit error rate by using the weighting factor can be greatly obtained, and is set in advance at the time of design.

  According to the research of the present inventor, the effect that the bit error rate is greatly improved by using the weighting factor is obtained remarkably when the correction amount is within the predetermined range, and the correction amount is within the predetermined range. If it is not, it has been found to be small. Specifically, the predetermined range can be determined by comparing the bit error rates with and without using a weighting factor.

  Effective selection of whether to generate weighting factors (that is, whether to perform demodulation and decoding using weighting factors) according to a predetermined range effectively increases processing by generating weighting factors. Can be prevented. Therefore, the bit error rate can be effectively improved with simpler processing.

  In another aspect of the information decoding apparatus of the present invention, the information decoding apparatus further comprises correction means for correcting the detected deviation in the input data using the correction amount, and the demodulation and decoding means includes the corrected data. Is demodulated and decoded.

  According to this aspect, the input data is corrected using the correction amount before being demodulated and decoded by the demodulation and decoding means. That is, the deviation from the detected reference data is reduced. As the deviation from the reference data is reduced, the input data becomes more accurate. This improves the bit error rate.

  In order to solve the above problems, an information decoding method of the present invention is an information decoding method for demodulating and decoding data that has been encoded with a code having error correction capability and that has been two-dimensionally modulated. A deviation detecting step for detecting a two-dimensional positional deviation from the reference data to be input, a correction amount generating step for generating a correction amount for correcting the detected deviation, and a magnitude of the correction amount Accordingly, a weighting factor generating step for generating a weighting factor used for weighting when demodulating and decoding the input data, and a demodulation decoding step for demodulating and decoding the input data using the weighting factor Including.

  According to the information decoding method of the present invention, since the weighting coefficient used for weighting when demodulating and decoding the input data is generated according to the magnitude of the correction amount, similarly to the information decoding apparatus described above, The encoded data can be suitably decoded by a relatively simple process.

  It should be noted that the information decoding method of the present invention can also take various aspects similar to those of the information decoding apparatus described above.

  The effect | action and other gain of this invention are clarified from the best form for implementing demonstrated below.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

<First Embodiment>
An information decoding apparatus according to the first embodiment will be described with reference to FIGS. In the following description, a case where two-dimensional modulation data encoded by an LDPC code is reproduced from a holographic memory will be described as an example.

  First, the configuration of the information decoding apparatus according to the first embodiment will be described with reference to FIG. 1 and FIG. FIG. 1 is a block diagram showing the overall configuration of the information decoding apparatus according to the first embodiment, and FIG. 2 is a block diagram showing the configuration of the demodulation / decoding unit in the information decoding apparatus according to the first embodiment. . In FIG. 2, for convenience of explanation, each component shown in FIG. 1 is omitted as appropriate.

  In FIG. 1, the information decoding apparatus according to the first embodiment includes a reproducing head 210, a marker detection unit 220, a geometric correction unit 230, a weight coefficient generation unit 240, and a demodulation decoding unit 250. Yes.

  The reproducing head 210 is configured to include, for example, a light emitting unit, a lens, and a light receiving element, and an image obtained by irradiating the holographic memory 100 with laser light is detected by the light receiving element and is two-dimensionally modulated image data. (Hereinafter referred to as “page” as appropriate).

  The marker detection unit 220 is an example of the “deviation detection unit” of the present invention, and detects a marker position deviation (predetermined preset pattern) existing at a predetermined position on the read page. , And the page inclination and the like are detected. Also, for each data position on the page, a shift is calculated from a plurality of marker detection positions.

  The geometric correction unit 230 is an example of the “correction amount generation unit” and the “correction unit” in the present invention, and generates a correction amount for reducing the shift based on the detected page shift. In addition, the page shift is actually corrected using the generated correction amount. However, as long as the correction amount is generated, the weighting coefficient can be generated as will be described later. Therefore, even when the geometric correction unit 230 does not perform correction using the correction amount, the effect according to the present embodiment can be obtained. In other words, the geometric correction unit 230 may be configured to generate only a correction amount and not actually perform correction. In this case, “corrected data” to be described later may be read as “reproduced data” or the like. The correction amount generated in the geometric correction unit 230 is used for correction and is also output to the weight coefficient generation unit 240.

  The weight coefficient generation unit 240 is an example of the “weight coefficient generation unit” of the present invention, and generates a weight coefficient used for data demodulation and decoding based on the correction amount generated by the geometric correction unit 230.

  The demodulator / decoder 250 is an example of the “demodulator / decoder” of the present invention, and demodulates and decodes the data to be output from the geometric corrector 230 and outputs the data. At the time of demodulation and decoding, the weighting factor generated by the weighting factor generator 240 is used.

  As shown in FIG. 2, the demodulation / decoding unit 250 includes a demodulation unit 251 which is an example of the “demodulation unit” of the present invention, and an LDPC decoding unit 252 which is an example of the “decoding unit” of the present invention. . The demodulation unit 251 demodulates the data (for example, the corrected luminance value) corrected by the geometric correction unit 230 and outputs the demodulation result as a likelihood value. The LDPC decoding unit 252 decodes and outputs data using the likelihood value and the weighting coefficient.

  Next, the operation of the information decoding apparatus according to the first embodiment will be described with reference to FIGS. 3 to 19 in addition to FIGS. FIG. 3 is a flowchart showing an operation flow of the information decoding apparatus according to the first embodiment, and FIG. 4 is a plan view showing an example of a page to be read.

  1 and 3, when the information decoding apparatus according to the first embodiment operates, the reproducing head 210 first reads data recorded in the holographic memory 100 (step S11).

  As shown in FIG. 4, the data recorded in the holographic memory 100 is read out as a page 500 that is two-dimensionally modulated image data. The page 500 includes a data area 510 containing data to be recorded and a position detection marker 520.

  When the page 500 is read, the marker detection unit 220 detects the position of the marker 520 on the page 500 (step S12). By detecting the position of the reference marker 520, the exact position of the page 500 can be detected. In addition, as shown in FIG. 4, by detecting a plurality of markers 520 a to 520 d, the inclination of the page 500 can be determined. Therefore, it is possible to know how much the read page 500 is deviated from the state to be read.

  Hereinafter, the shift of the page 500 at the time of reading will be described in more detail with reference to FIGS. 5 and 6. FIG. 5 is a plan view showing an example of reproduction data when there is no deviation, and FIG. 6 is a plan view showing an example of reproduction data when there is a deviation. Note that the data in the page 500 is represented as white or black pixels.

  In FIG. 5, when there is no deviation between the data recorded in the holographic memory 100 and the light receiving element in the reproducing head 210, white or black in the page 500 for each pixel (thick line in the drawing) of the light receiving element. Are detected without positional deviation. As a result, the reproduced image shown on the right side of the drawing has only white or black luminance values. Therefore, in such a case, demodulation and decoding can be accurately performed without correction.

  In FIG. 6, when there is a deviation between the data recorded in the holographic memory 100 and the light receiving element in the reproducing head 210, the reproduced image captured by the light receiving element is shown in a thick line indicating each pixel of the light receiving element. The luminance value corresponds to the black and white ratio. That is, the reproduced image is an image including gray in addition to white and black. Therefore, unlike the case shown in FIG. 5, it is difficult to accurately demodulate and decode the reproduced image.

  The information decoding apparatus according to this embodiment performs correction for reducing the shift in the geometric correction unit 230 with respect to the shift of the page 500 described above. Hereinafter, the generation and correction of the correction amount will be described with reference to FIGS. FIG. 7 is a table showing the coordinates of the data position in each pixel of the data area, and FIG. 8 is a table showing the correction amount in each pixel. 7 and 8 show only positions and correction amounts for some of the plurality of pixels for convenience of explanation, and the following tables are similarly omitted as appropriate.

  1 and 3, when a page shift is detected by the marker detection unit 210, the geometric correction unit 230 first generates a correction amount based on the shift amount of each pixel in the data area 510 in the page 500 (step S13). ).

  In FIG. 7, it is assumed here that data of adjacent 4 × 4 pixels is detected at coordinates as shown in the table. In the table, the values on the left side are the coordinates in the x direction in FIG. 4, and the values on the right side are the coordinates in the y direction. That is, (15.20, 24.30) in the upper left indicates that the coordinate in the x direction is 15.20 and the coordinate in the y direction is 24.30. The integer part in the table corresponds to one pixel. For example, data whose coordinate in the x direction is “1” larger indicates data adjacent to the right side in the x direction.

  In FIG. 8, the geometric correction unit 230 generates a geometric correction amount by removing an integer part from each value shown in FIG. That is, the geometric correction amount is a decimal part in coordinates. Therefore, if there is no deviation, the decimal part of the value in the table is “0”.

  Returning to FIGS. 1 and 3, when the geometric correction unit 230 generates the geometric correction amount, the geometric correction amount is used to correct the misalignment of the page (step S14). Thereby, for example, data having variations in luminance values as shown in FIG. 6 can be corrected into accurate data as shown in FIG.

  When the shift of the page 500 is corrected, the weighting coefficient generation unit 240 generates a weighting coefficient used in demodulation and decoding from the geometric correction amount used in the correction (step S15). The weighting coefficient is a coefficient for weighting information to be demodulated and decoded.

  Hereinafter, the generation of the weight coefficient will be described with reference to FIGS. 9 to 11. FIG. 9 is a table showing values δx and δy obtained from the correction amount in each pixel, and FIG. 10 is a table showing values selδ obtained from δx and δy for each modulation symbol. FIG. 11 is a table showing weighting factors for each modulation symbol.

  In FIG. 9, the weight coefficient generation unit 240 obtains δx and δy for each pixel from the geometric correction amount shown in FIG. δx and δy can be obtained by the following equations (1) and (2), respectively, where gx is a geometric correction amount in each pixel and gy is a geometric correction amount in the y direction.

δx = | gx−0.5 | (1)
δy = | gy−0.5 | (2)
Δx and δy obtained as described above are smaller as the geometric correction amount is larger. That is, the larger the deviation when detected, the smaller the value. Here, “0.5” is subtracted because the absolute value of the deviation amount is 0.5 at the maximum (for example, data shifted by 0.5 in the x direction is 0 in the −x direction. This is because it can be regarded as data shifted by 5).

  In FIG. 10, the weight coefficient generation unit 240 obtains selδ for each modulation symbol that is a modulation unit in the two-dimensional modulation from δx and δy. Therefore, for example, for (2, 4) modulated data, selδ is obtained for every 2 × 2 pixels that are modulation symbol units. selδ can be obtained by the following equation (3).

selδ = min (min (δx), min (δy)) (3)
Note that min (n) means that the minimum value is selected from a plurality of values.

  According to Equation (3) described above, one selδ is obtained for each modulation symbol. That is, here, one selδ is obtained for every four pixels. Specifically, one selδ is obtained for every four pixels surrounded by a broken line in FIG. In the upper left four pixels in FIG. 9, min (δx) is 0.29 and min (δy) is 0.18. Furthermore, since 0.29 <0.18, selδ here is 0.18.

  The calculation method of selδ using equation (3) is merely an example, and selδ can be obtained using another equation. For example, selδ can also be obtained using any of the following formulas (4) to (9).

selδ = min (mean (δx), mean (δy)) (4)
selδ = mean (mean (δx), mean (δy)) (5)
selδ = min (δx) + min (δy) (6)
selδ = mean (δx) + mean (δy) (7)
selδ = ((min (δx)) ² + (min (δy)) ² ) ^1/2 (8)
selδ = ((mean (δx)) ² + (mean (δy)) ² ) ^1/2 (9)
Here, mean (n) represents the average value calculation.

  In FIG. 11, the weighting factor generation unit 240 generates a weighting factor using selδ obtained as described above. When generating the weighting factor, in addition to selδ, predetermined constants gain and offset are used. The weighting factor w can be obtained using the following equation (10).

w = gain × selδ + offset (10)
Note that the values of gain and offset are design matters and can be appropriately changed depending on the level to be weighted. Further, gain and offset may be selected from a plurality of values using the value of selδ or the like as a threshold value. When the equation (10) is applied to selδ shown in FIG. 10 with gain = 1 and offset = 0.5, the weighting coefficient shown in FIG. 11 is obtained.

  In FIG. 2, the weighting coefficient generated as described above is input to the LDPC decoding unit 252 in the demodulation decoding unit 250. The demodulating / decoding unit 250 first demodulates the data using the corrected luminance value indicating the data corrected by the demodulating unit 251 (step S16).

  Hereinafter, data demodulation will be described with reference to FIGS. FIG. 12 is a table showing the corrected luminance value in each pixel, and FIG. 13 is a conceptual diagram showing a modulation rule and a method for determining a code when obtaining a likelihood value. FIG. 14 is a table showing likelihood values output by demodulation.

  12 and 13, the corrected luminance value as shown in FIG. 12 is output to the demodulation unit 251 as a result of the correction performed by the geometric correction unit 230. However, the luminance value takes a value of 0.0 (black) to 1.0 (white). When the demodulation unit 251 uses the (2, 4) modulation rule as shown in FIG. 13, one symbol is in units of four pixels, and four corrected luminance values P1 to P4 at that time (where P1> P2> P3> When P4) is obtained, the likelihood value λ is obtained using the following equation (11).

λ _n = sign (P1-P2) (11)
Sign (n) represents code selection, and the code is based on a modulation rule as shown in FIG. FIG. 13 shows modulation / demodulation of (2, 4) modulation. When the input data at the time of recording is “00”, the modulation pixel is modulated into a recording image in which only the upper left part is white and the others are black, Record. The decoding apparatus shown in FIGS. 1 and 2 receives a corrected luminance value in which the upper left is white and the other is black as a reproduced image (corrected luminance value) in the demodulator, and the likelihood code that is output from the demodulator is “ -"Is output. Here, when “−” is “0” and “+” is hard-decided to “1”, the sign of likelihood matches the input data at the time of recording.

  The position of the pixel represented by white can be determined from the position of the pixel (P1) having the highest luminance value. Therefore, since the modulation symbol at the upper left in FIG. 12 has a maximum luminance value of 0.7, only the upper left portion can be determined as data represented by white. Accordingly, the sign of the likelihood value is determined as “-”. When the likelihood value is obtained in this way, a value as shown in FIG. 14 is obtained from the corrected luminance value shown in FIG.

  Returning to FIG. 2, when the likelihood value is output from the demodulation unit 251, the LDPC decoding unit 252 performs data decoding using the likelihood value and the weighting coefficient. Hereinafter, data decoding will be described with reference to FIGS. 15 to 19. FIG. 15 is a table showing weighted likelihood values in each modulation symbol. FIG. 16 is a diagram illustrating an example of a parity check matrix that defines an LDPC matrix, and FIG. 17 is a Tanner graph corresponding to the parity check matrix in FIG. FIG. 18 is a graph showing the relationship between the signal-to-noise ratio and the bit error rate, and FIG. 19 is a graph showing the relationship between the geometric correction amount and the bit error rate.

  2 and 3, the LDPC decoding unit 252 first multiplies the likelihood value by a weighting factor (step S17). As a result, the likelihood value becomes a weighted likelihood value. For example, when the likelihood value shown in FIG. 14 is multiplied by the weighting coefficient shown in FIG. 11, a weighted likelihood value as shown in FIG. 15 is obtained.

  When the weighted likelihood value is obtained, the LDPC decoding unit 252 decodes data using the weighted likelihood value (step S17). When decoding data encoded by the LDPC code, a check matrix H as shown in FIG. 16 is used. Specifically, data is decoded by repeatedly exchanging information called a message on a bipartite graph called a Tanner graph that has a one-to-one correspondence with the check matrix H as shown in FIG.

  Note that the check nodes in the Tanner graph correspond to the rows of the check matrix H, respectively. Variable nodes correspond to the columns of the check matrix H, respectively. A line connecting the check node and the variable node is called a branch (edge), and this corresponds to the element ‘1’ in the check matrix H. Messages are exchanged between check nodes and variable nodes along the branches.

  The decoded data is output as reproduction data from the LDPC decoding unit 252 to the outside (step S18). In the information decoding apparatus according to the present embodiment, when decoding data, the bit error rate in the decoded data is reduced by using a weighting factor corresponding to the corrected luminance value and the geometric correction amount.

  As shown in FIGS. 18 and 19, the bit error rate when the weighted likelihood value λw multiplied by the weighting factor w is used is more reliable than when the likelihood value λ not multiplied by the weighting factor w is used. To drop. This is because the corrected image after being corrected by the geometric correction unit 230 is a corrected image by interpolation from adjacent pixels, and therefore, a different corrected image may be generated from the normal image due to the influence of each pixel noise or the like. Is due to being. In other words, a corrected image from a reproduced image with no positional deviation has less influence from adjacent pixels, and the influence of noise is less than that of a corrected image with positional deviation. By applying a weighting factor to the likelihood value, it is possible to reflect the degree of influence of the positional deviation described above. Therefore, it is possible to effectively reduce the bit error rate.

  Further, as shown in FIGS. 18 and 19, the range in which the effect of reducing the bit error rate is significantly obtained is determined by the data. Therefore, for example, by using the geometric correction amount as a threshold value, a weighting factor is not partially generated (that is, depending on the value of the geometric correction amount, the weighting factor is not used), thereby preventing processing complexity. can do. Specifically, in FIG. 19, the bit error rate is significantly reduced in the range of 0.16 <selδ <0.3. Therefore, it is only necessary to generate the weighting coefficient when 0.16 <selδ <0.3 is satisfied.

  As described above, according to the information decoding apparatus according to the first embodiment, it is possible to suitably decode the encoded data by a relatively simple process.

<Second Embodiment>
Next, an information decoding apparatus according to the second embodiment will be described with reference to FIGS. The second embodiment differs from the first embodiment described above in the configuration and operation of the demodulation / decoding unit, and the other configurations are generally the same. Therefore, a different part from 1st Embodiment is demonstrated in detail here, and description is abbreviate | omitted suitably about another overlapping part. In the following drawings, the same reference numerals are assigned to the same components as those in the first embodiment shown in FIGS. 1 to 19.

  First, the configuration of the information decoding apparatus according to the second embodiment will be described with reference to FIG. FIG. 20 is a block diagram showing the configuration of the demodulation / decoding unit in the information decoding apparatus according to the second embodiment.

  In FIG. 20, the decoding demodulation unit 250 in the information decoding apparatus according to the second embodiment includes a demodulation unit 251 and an LDPC decoding unit 252. The demodulating unit 251 demodulates the data (corrected luminance value) corrected by the geometric correcting unit 230 using a weighting factor, and outputs the result as a likelihood value. The LDPC decoding unit 252 decodes and outputs data using the likelihood value output from the demodulation unit 251.

  As described above, in the information decoding apparatus according to the second embodiment, the weighting coefficient is input to the demodulation unit 251 instead of the LDPC decoding unit 252 in the demodulation decoding unit 250. The configuration other than the decoding demodulation unit 250 in the information decoding apparatus according to the second embodiment is substantially the same as that of the information decoding apparatus according to the first embodiment shown in FIG.

  Next, the operation of the information decoding apparatus according to the second embodiment will be described with reference to FIGS. FIG. 21 is a flowchart showing an operation flow of the information decoding apparatus according to the second embodiment.

  1 and 21, when the information decoding apparatus according to the second embodiment operates, the reproducing head 210 first reads out data recorded in the holographic memory 100 (step S21). When the page 500 is read, the marker detection unit 220 detects the position of the marker 520 on the page 500 (step S22). Further, the geometric correction unit 230 generates a correction amount based on the shift amount of each pixel of the data area 510 in the page 500 (step S23), and corrects the data shift based on the generated correction amount (step S24). ). The processing so far is the same as the processing (step S11 to step S14) according to the first embodiment shown in FIG.

  Subsequently, the weight coefficient generation unit 240 generates a weight coefficient corresponding to the magnitude of the correction amount (step S25). Hereinafter, the generation of the weighting coefficient will be described with reference to FIGS. 22 and 23 in addition to FIGS. 8 and 9. FIG. 22 is a table showing the value selδ obtained from δx and δy for each pixel, and FIG. 23 is a table showing the weighting coefficient for each pixel.

  8 and 9, the weighting coefficient generation unit 240 obtains δx and δy in each pixel from the geometric correction amount used to correct the data shift, as in the information decoding apparatus according to the first embodiment. . Here, in the information decoding apparatus according to the second embodiment, selδ is obtained for each pixel from δx and δy. This is because an object to be multiplied by the weighting coefficient is a luminance correction value (that is, a value indicated for each pixel) that is an input of the demodulation unit. For example, selδ can be obtained using the following equation (12).

selδ = min (δx, δy) (12)
That is, the smaller value of δx and δy in each pixel is selδ. According to the equation (12) described above, selδ as shown in FIG. 22 is obtained from δx and δy shown in FIG. It can also be obtained by using any of the following formulas (13) to (15).

selδ = mean (δx, δy) (13)
selδ = δx + δy (14)
selδ = ((δx) ² + (δy) ² ) ^1/2 (15)
Subsequently, the weighting factor generation unit 240 generates a weighting factor using selδ obtained as described above. When generating the weighting factor, in addition to selδ, predetermined constants gain and offset are used. The weighting factor w can be obtained using the following equation (16).

w = gain × selδ + offset (16)
Note that the values of gain and offset are design matters and can be appropriately changed depending on the level to be weighted. Further, gain and offset may be selected from a plurality of values using the value of selδ or the like as a threshold value. If the equation (16) is applied to selδ shown in FIG. 22 with gain = 1 and offset = 0.5, the weighting coefficient shown in FIG. 23 is obtained.

  Returning to FIG. 21, in the information decoding apparatus according to the second embodiment, the weight coefficient is multiplied by the corrected luminance value that is the input of the demodulator 251 in the demodulator / decoder 250 (step S26). Then, the demodulation unit 251 performs demodulation using the weighted corrected luminance value multiplied by the weight coefficient (step S27).

  Hereinafter, data demodulation in the demodulation / decoding unit 250 will be described with reference to FIGS. 24 and 25 in addition to FIGS. 12 and 23. FIG. 24 is a table showing weighted corrected luminance values corresponding to each pixel, and FIG. 25 is a table showing likelihood values output by demodulation.

  In the demodulation, the weighting coefficient shown in FIG. 23 is input to the demodulator 251 in addition to the corrected luminance value shown in FIG. The demodulator 251 multiplies the corrected luminance value P by a weight coefficient w for each pixel. Thereby, a weighted correction luminance value Pw as shown in FIG. 24 is obtained.

  Further, the demodulation unit 251 obtains a likelihood value λ for each modulation symbol by applying the above-described equation (11) to the weighted corrected luminance value Pw. For example, from the weighted corrected luminance value Pw shown in FIG. 24, a likelihood value as shown in FIG. 25 is obtained for each symbol surrounded by a dotted line. In addition, since the likelihood value calculated | required here is calculated | required from a weighted correction | amendment luminance value, it is a weighted likelihood value.

  Returning to FIG. 21, when the weighted likelihood value is obtained, the LDPC decoding unit 252 decodes the data using the weighted likelihood value (step S27). The decoded data is output as reproduction data from the LDPC decoding unit 252 to the outside (step S28). In the information decoding apparatus according to the second embodiment, when demodulating data, the bit error rate in the decoded data is reduced by using a weighting factor corresponding to the corrected luminance value and the geometric correction amount.

  As shown in FIGS. 26 and 27, the bit error rate when using the weighted corrected luminance value Pw multiplied by the weighting factor w is more reliable than when the corrected luminance value P not multiplied by the weighting factor w is used. To drop. This is because the influence of the positional deviation of the page 500 can be reflected in demodulation and decoding by applying a weighting coefficient to the luminance correction value. As a result, the decoding apparatus according to the second embodiment can effectively reduce the bit error rate.

  In addition, in the information decoding apparatus according to the second embodiment, since the weight coefficient is generated in units of pixels, it is possible to cope with so-called decimal oversampling. That is, when the data recorded in the holographic memory 100 is read out by the reproducing head 210, the recorded two-dimensional modulation data pixels and the pixels of the light receiving elements in the reproducing head 210 do not seem to have a one-to-one correspondence. Even in such a case, it is possible to reliably reduce the bit error rate.

  As described above, according to the information decoding apparatus according to the second embodiment, it is possible to suitably decode the encoded data as in the first embodiment described above.

  The present invention is not limited to the above-described embodiment, and can be appropriately changed without departing from the gist or concept of the invention that can be read from the claims and the entire specification, and information accompanying such a change. Decoding devices and methods are also included in the technical scope of the present invention.

It is a block diagram which shows the whole structure of the information decoding apparatus which concerns on 1st Embodiment. It is a block diagram which shows the structure of the demodulation decoding part in the information decoding apparatus which concerns on 1st Embodiment. It is a flowchart which shows the flow of operation | movement of the information decoding apparatus which concerns on 1st Embodiment. It is a top view which shows an example of the page read. It is a top view which shows an example of the reproduction | regeneration data when the gap has not arisen. It is a top view which shows an example of the reproduction | regeneration data when the gap has arisen. It is a table | surface which shows the coordinate of the data position in each pixel of a data area. It is a table | surface which shows the correction amount in each pixel. It is a table | surface which shows the value (delta) x and (delta) y calculated | required from the correction amount in each pixel. It is a table | surface which shows value seldelta calculated | required from (delta) x and (delta) y for every modulation symbol. It is a table | surface which shows the weighting coefficient with respect to each modulation symbol. It is a table | surface which shows the correction | amendment luminance value in each pixel. It is a conceptual diagram which shows the determination method of the code | symbol at the time of calculating | requiring likelihood value. It is a table | surface (the 1) which shows the likelihood value output by demodulation. It is a table | surface which shows the weighted likelihood value in each modulation symbol. It is a figure which shows an example of the check matrix which prescribes | regulates an LDPC matrix. 17 is a Tanner graph corresponding to the parity check matrix of FIG. It is a graph (the 1) which shows the relationship between a signal to noise ratio and a bit error rate. It is a graph (the 1) which shows the relationship between geometric correction amount and a bit error rate. It is a block diagram which shows the structure of the demodulation decoding part in the information decoding apparatus which concerns on 2nd Embodiment. It is a flowchart which shows the flow of operation | movement of the information decoding apparatus which concerns on 2nd Embodiment. It is a table | surface which shows value seldelta calculated | required from (delta) x and (delta) y for every pixel. It is a table | surface which shows the weighting coefficient with respect to each pixel. It is a table | surface which shows the weighted correction | amendment luminance value corresponding to each pixel. It is a table | surface (the 2) which shows the likelihood value output by demodulation. It is a graph (the 2) which shows the relationship between a signal to noise ratio and a bit error rate. It is a graph (the 2) which shows the relationship between geometric correction amount and a bit error rate.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100 ... Holographic memory, 210 ... Reproducing head, 220 ... Marker detection part, 230 ... Geometric correction part, 240 ... Weight coefficient generation part, 250 ... Demodulation decoding part, 251 ... Demodulation part, 252 ... LDPC decoding part, 500 ... Page , 510 ... Data recording area, 520 ... Marker

Claims (9)

An information decoding apparatus that demodulates and decodes data that is encoded by a code having error correction capability and that is two-dimensionally modulated,
A deviation detecting means for detecting a two-dimensional positional deviation between the input data and the reference data to be input;
Correction amount generating means for generating a correction amount for correcting the detected deviation;
Weight coefficient generation means for generating a weight coefficient used for weighting when demodulating and decoding the input data according to the magnitude of the correction amount;
An information decoding apparatus comprising: demodulation and decoding means for demodulating and decoding the input data using the weighting factor. The demodulation and decoding means includes
Demodulation means for demodulating the input data;
The information decoding apparatus according to claim 1, further comprising: a decoding unit that decodes the demodulated data using the weighting factor.   3. The information decoding apparatus according to claim 2, wherein the weight coefficient generation means generates the weight coefficient for each modulation symbol in the two-dimensional modulation that is an output unit of the demodulation means. The demodulation and decoding means includes
Demodulating means for demodulating the input data using the weighting factor;
The information decoding apparatus according to claim 1, further comprising: a decoding unit that decodes the demodulated data.   5. The information decoding apparatus according to claim 4, wherein the weighting coefficient generation unit generates the weighting coefficient for each pixel in the two-dimensional modulation that is an input unit of the demodulation unit.   The weighting factor generating means selects one process based on a predetermined threshold from a plurality of processes for generating the weighting factor, and generates the weighting coefficient by the one process. The information decoding device according to any one of 1 to 5.   The information decoding apparatus according to claim 1, wherein the weighting coefficient generation unit generates the weighting coefficient when the correction amount is within a predetermined range. A correction means for correcting the detected deviation in the input data using the correction amount;
The information decoding apparatus according to any one of claims 1 to 7, wherein the demodulation and decoding means demodulates and decodes the corrected data. An information decoding method for demodulating and decoding data encoded by a code having error correction capability and two-dimensionally modulated,
A deviation detection step of detecting a two-dimensional positional deviation between the input data and the reference data to be input;
A correction amount generating step for generating a correction amount for correcting the detected deviation;
A weighting factor generating step for generating a weighting factor used for weighting when demodulating and decoding the input data according to the magnitude of the correction amount;
A demodulation decoding step of demodulating and decoding the input data using the weighting factor.

Images