Low-density parity-check decoding with de-saturation
A saturation metric that represents a degree of saturation in a low-density parity-check (LDPC) decoding system that uses a fixed-point number representation is determined. The saturation metric is compared against a saturation threshold. In the event the saturation metric exceeds the saturation threshold, at the end of a decoding iteration, a message is more aggressively attenuated compared to when the saturation metric does not exceed the saturation threshold in order to produce an attenuated message. In the event the saturation metric does not exceed the saturation threshold, at the end of the decoding iteration, the message is less aggressively attenuated compared to when the saturation metric does exceed the saturation threshold in order to produce the attenuated message.
1. A decoder, comprising:
a saturation monitor configured to:
determine a saturation metric that represents a degree of saturation in a low-density parity-check (LDPC) decoding system that uses a fixed-point number representation; and
compare the saturation metric against a saturation threshold; and
a multiplexer and attenuation block configured to:
in the event the saturation metric exceeds the saturation threshold, at the end of a decoding iteration, more aggressively attenuate a message compared to when the saturation metric does not exceed the saturation threshold in order to produce an attenuated message; and
in the event the saturation metric does not exceed the saturation threshold, at the end of the decoding iteration, less aggressively attenuate the message compared to when the saturation metric does exceed the saturation threshold in order to produce the attenuated message.
2. The decoder recited in claim 1 , wherein:
a de-saturation attenuation factor (α ds ) and a de-saturation attenuation rounding (β ds ) are used to more aggressively attenuate the message; and
an attenuation factor (α) and an attenuation rounding (β) are used to less aggressively attenuate the message, wherein α ds is within a range of
1
2
α
and
3
4
α
and β ds is within a range or
1
2
β
and
3
4
β
.
3. The decoder recited in claim 1 , wherein:
a de-saturation attenuation factor (α ds ) and a de-saturation attenuation rounding (β ds ) are used to more aggressively attenuate the message;
an attenuation factor (α) and an attenuation rounding (β) are used to less aggressively attenuate the message; and
at least one of α, α ds , β, and β ds is selected using density evolution optimization.
4. The decoder recited in claim 1 , wherein:
the decoding iteration is associated with a min-sum decoding iteration and the min-sum decoding iteration includes a critical timing path; and
in the event the saturation metric exceeds the saturation threshold, the message is more aggressively attenuated outside of the critical timing path such that the more aggressive attenuation of the message does not add additional delay to the critical timing path in the min-sum decoding iteration.
5. The decoder recited in claim 1 , wherein:
the decoding iteration is associated with a min-sum decoding iteration and the min-sum decoding iteration includes one or more existing pipeline stages; and
in the event the saturation metric exceeds the saturation threshold, the message is more aggressively attenuated outside of the existing pipeline stages such that the more aggressive attenuation of the message does not affect the existing pipeline stages in the min-sum decoding iteration.
6. A method, comprising:
determining a saturation metric that represents a degree of saturation in a low-density parity-check (LDPC) decoding system that uses a fixed-point number representation;
comparing the saturation metric against a saturation threshold;
in the event the saturation metric exceeds the saturation threshold, at the end of a decoding iteration, more aggressively attenuating a message compared to when the saturation metric does not exceed the saturation threshold in order to produce an attenuated message; and
in the event the saturation metric does not exceed the saturation threshold, at the end of the decoding iteration, less aggressively attenuating the message compared to when the saturation metric does exceed the saturation threshold in order to produce the attenuated message.
7. The method recited in claim 6 , wherein:
a de-saturation attenuation factor (α ds ) and a de-saturation attenuation rounding (β ds ) are used to more aggressively attenuate the message; and
an attenuation factor (α) and an attenuation rounding (β) are used to less aggressively attenuate the message, wherein α ds is within a range of
1
2
α
and
3
4
α
and β ds is within a range or
1
2
β
and
3
4
β
.
8. The method recited in claim 6 , wherein:
a de-saturation attenuation factor (α ds ) and a de-saturation attenuation rounding (β ds ) are used to more aggressively attenuate the message;
an attenuation factor (α) and an attenuation rounding (β) are used to less aggressively attenuate the message; and
at least one of α, α ds , β, and β ds is selected using density evolution optimization.
9. The method recited in claim 6 , wherein:
the decoding iteration is associated with a min-sum decoding iteration and the min-sum decoding iteration includes a critical timing path; and
in the event the saturation metric exceeds the saturation threshold, the message is more aggressively attenuated outside of the critical timing path such that the more aggressive attenuation of the message does not add additional delay to the critical timing path in the min-sum decoding iteration.
10. The method recited in claim 6 , wherein:
the decoding iteration is associated with a min-sum decoding iteration and the min-sum decoding iteration includes one or more existing pipeline stages; and
in the event the saturation metric exceeds the saturation threshold, the message is more aggressively attenuated outside of the existing pipeline stages such that the more aggressive attenuation of the message does not affect the existing pipeline stages in the min-sum decoding iteration.
11. A computer program product, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:
determining a saturation metric that represents a degree of saturation in a low-density parity-check (LDPC) decoding system that uses a fixed-point number representation;
comparing the saturation metric against a saturation threshold;
in the event the saturation metric exceeds the saturation threshold, at the end of a decoding iteration, more aggressively attenuating a message compared to when the saturation metric does not exceed the saturation threshold in order to produce an attenuated message; and
in the event the saturation metric does not exceed the saturation threshold, at the end of the decoding iteration, less aggressively attenuating the message compared to when the saturation metric does exceed the saturation threshold in order to produce the attenuated message.
12. The computer program product recited in claim 11 , wherein:
a de-saturation attenuation factor (α ds ) and a de-saturation attenuation rounding (β ds ) are used to more aggressively attenuate the message; and
an attenuation factor (α) and an attenuation rounding (β) are used to less aggressively attenuate the message, wherein α ds , is within a range of
1
2
α
and
3
4
α
and β ds is within a range of
1
2
β
and
3
4
β
.
13. The computer program product recited in claim 11 , wherein:
a de-saturation attenuation factor (α ds ) and a de-saturation attenuation rounding (β ds ) are used to more aggressively attenuate the message;
an attenuation factor (α) and an attenuation rounding (β) are used to less aggressively attenuate the message; and
at least one of α, α ds , β, and β ds is selected using density evolution optimization.
14. The computer program product recited in claim 11 , wherein:
the decoding iteration is associated with a min-sum decoding iteration and the min-sum decoding iteration includes a critical timing path; and
in the event the saturation metric exceeds the saturation threshold, the message is more aggressively attenuated outside of the critical timing path such that the more aggressive attenuation of the message does not add additional delay to the critical timing path in the min-sum decoding iteration.
15. The computer program product recited in claim 11 , wherein:
the decoding iteration is associated with a min-sum decoding iteration and the min-sum decoding iteration includes one or more existing pipeline stages; and
in the event the saturation metric exceeds the saturation threshold, the message is more aggressively attenuated outside of the existing pipeline stages such that the more aggressive attenuation of the message does not affect the existing pipeline stages in the min-sum decoding iteration.