Low-density parity-check (LDPC) data decoding using iteration-variable accuracy
Low-Density Parity-Check (LDPC) data decoding using iteration-variable accuracy is performed by segmenting a LDPC encoded data block of probability values by dividing the LDPC encoded data block into a plurality of data probability value segments and a plurality of parity probability value segments, each probability value of the LDPC encoded data block representing a likelihood between binary values, decoding the LDPC encoded data block by adjusting, according to an iteration-variable accuracy parameter, the probability values of the LDPC encoded data block based on a parity-check matrix, the parity-check matrix defining correspondence among data probability value segments and parity probability value segments, and concatenating likely binary values that satisfy the parity-check matrix associated with the probability values of each data probability value segment to form a decoded data block. The iteration-variable accuracy parameter represents a tradeoff between accuracy and computational efficiency.
1. A method comprising:
segmenting, using a segmentation processor, a Low-Density Parity-Check (LDPC) encoded data block of probability values by dividing the LDPC encoded data block into a plurality of data probability value segments and a plurality of parity probability value segments, each probability value of the LDPC encoded data block representing a likelihood between binary values;
decoding, using a decoding processor, the LDPC encoded data block by adjusting, according to an iteration-variable accuracy parameter, the probability values of the LDPC encoded data block based on a parity-check matrix, the parity-check matrix defining correspondence among data probability value segments and parity probability value segments, wherein the decoding includes
comparing subsets of probability values, each subset including probability values that are related according to the parity-check matrix, the related probability values of each subset being from corresponding data probability value segments and parity probability value segments,
wherein the probability values of each subset correspond to original binary values, a sum total of which is even, and
wherein each probability value is included in more than one subset; and
adjusting each probability value in each subset based on one or more minimum confidence probability values among other probability values in the subset,
wherein each probability value is adjusted to increase confidence in a likely binary value in response to determining that a sum total of the likely binary values in the subset is even, and
wherein each probability value is adjusted to decrease confidence in the likely binary value in response to determining that the sum total of the likely binary values in the subset is not even; and
concatenating, using a concatenating processor, likely binary values that satisfy the parity-check matrix associated with the probability values of each data probability value segment to form a decoded data block.
2. The method of claim 1 , wherein the decoding further includes
determining, for each subset, an adjustment basis quantity, the adjustment basis quantity representing a quantity of minimum confidence probability values among the other probability values in the subset on which to base the adjusting;
wherein a quantity of the one or more minimum confidence probability values is the adjustment basis quantity; and
wherein the adjustment basis quantity is less than or equal to a quantity of the other probability values in the subset.
3. The method of claim 2 , wherein the decoding further includes
performing iterations of adjusting each probability value in each subset;
determining, for each subsequent iteration of adjusting, the iteration-variable accuracy parameter based on an iteration value and an error rate of at least the previous iteration; and
applying, for each subset, the adjustment basis quantity that is the lesser between the iteration-variable accuracy parameter and one less than a quantity of probability values in the subset.
4. The method of claim 3 , wherein
the error rate is a proportion of subsets in which the sum total of the likely binary values in the subset is even to all of the subsets.
5. The method of claim 3 , wherein
the error rate is a proportion of probability values in subsets in which the sum total of the likely binary values in the subset is even to all of the probability values.
6. The method of claim 3 , wherein the determining the iteration-variable accuracy parameter further includes
determining the iteration-variable accuracy parameter further based on a signal-to-noise ratio.
7. The method of claim 3 , wherein the iterations of adjusting are performed until likely binary values associated with the probability values satisfy the parity-check matrix or until a predetermined number of iterations are performed.
8. A non-transitory computer-readable medium including instructions stored thereon that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
segmenting, using a segmentation processor, a Low-Density Parity-Check (LDPC) encoded data block of probability values by dividing the LDPC encoded data block into a plurality of data probability value segments and a plurality of parity probability value segments, each probability value of the LDPC encoded data block representing a likelihood between binary values;
decoding, using a decoding processor, the LDPC encoded data block by adjusting, according to an iteration-variable accuracy parameter, the probability values of the LDPC encoded data block based on a parity-check matrix, the parity-check matrix defining correspondence among data probability value segments and parity probability value segments, wherein the decoding includes
comparing subsets of probability values, each subset including probability values that are related according to the parity-check matrix, the related probability values of each subset being from corresponding data probability value segments and parity probability value segments,
wherein the probability values of each subset correspond to original binary values, a sum total of which is even, and
wherein each probability value is included in more than one subset; and
adjusting each probability value in each subset based on one or more minimum confidence probability values among other probability values in the subset,
wherein each probability value is adjusted to increase confidence in a likely binary value in response to determining that a sum total of the likely binary values in the subset is even, and
wherein each probability value is adjusted to decrease confidence in the likely binary value in response to determining that the sum total of the likely binary values in the subset is not even; and
concatenating, using a concatenating processor, likely binary values that satisfy the parity-check matrix associated with the probability values of each data probability value segment to form a decoded data block.
9. The computer-readable medium of claim 8 , wherein the decoding further includes
determining, for each subset, an adjustment basis quantity, the adjustment basis quantity representing a quantity of minimum confidence probability values among the other probability values in the subset on which to base the adjusting;
wherein a quantity of the one or more minimum confidence probability values is the adjustment basis quantity; and
wherein the adjustment basis quantity is less than or equal to a quantity of the other probability values in the subset.
10. The computer-readable medium of claim 9 , wherein the decoding further includes
performing iterations of adjusting each probability value in each subset;
determining, for each subsequent iteration of adjusting, the iteration-variable accuracy parameter based on an iteration value and an error rate of at least the previous iteration; and
applying, for each subset, the adjustment basis quantity that is the lesser between the iteration-variable accuracy parameter and one less than a quantity of probability values in the subset.
11. The computer-readable medium of claim 10 , wherein
the error rate is a proportion of subsets in which the sum total of the likely binary values in the subset is even to all of the subsets.
12. The computer-readable medium of claim 10 , wherein
the error rate is a proportion of probability values in subsets in which the sum total of the likely binary values in the subset is even to all of the probability values.
13. The method of claim 10 , wherein the determining the iteration-variable accuracy parameter further includes
determining the iteration-variable accuracy parameter further based on a signal-to-noise ratio.
14. The method of claim 10 , wherein the iterations of adjusting are performed until likely binary values associated with the probability values satisfy the parity-check matrix or until a predetermined number of iterations are performed.
15. A device comprising:
a segmentation processor configured to divide an LDPC encoded data block of probability values into a plurality of data probability value segments and a plurality of parity probability value segments, each probability value of the LDPC encoded data block representing a likelihood between binary values;
a decoding processor configured to
adjust, according to an iteration-variable accuracy parameter, the probability values of the LDPC encoded data block based on a parity-check matrix, the parity-check matrix defining correspondence among data probability value segments and parity probability value segments,
compare subsets of probability values, each subset including probability values that are related according to the parity-check matrix, the related probability values of each subset being from corresponding data probability value segments and parity probability value segments,
wherein the probability values of each subset correspond to original binary values, a sum total of which is even, and
wherein each probability value is included in more than one subset, and
adjust each probability value in each subset based on one or more minimum confidence probability values among other probability values in the subset,
wherein each probability value is adjusted to increase confidence in a likely binary value in response to determining that a sum total of the likely binary values in the subset is even, and
wherein each probability value is adjusted to decrease confidence in the likely binary value in response to determining that the sum total of the likely binary values in the subset is not even; and
a concatenating processor configured to concatenate likely binary values that satisfy the parity-check matrix associated with the probability values of each data probability value segment to form a decoded data block.
16. The device of claim 15 , wherein the decoding processor is further configured to
determine, for each subset, an adjustment basis quantity, the adjustment basis quantity representing a quantity of minimum confidence probability values among the other probability values in the subset on which to base the adjusting;
wherein a quantity of the one or more minimum confidence probability values is the adjustment basis quantity; and
wherein the adjustment basis quantity is less than or equal to a quantity of the other probability values in the subset.
17. The device of claim 15 , wherein the device further includes, among the circuitry:
a parameter processor configured to transmit a plurality of parameters to each of the segmentation processor, the decoding processor, and the concatenating processor,
wherein the plurality of parameters corresponds to the encoded data block.