Systems and methods for directed soft data perturbation in layered decoding
View Patent ↗Systems and method relating generally to data processing, and more particularly to systems and methods for perturbing soft data in a layered decoder system.
1. A data processing system, the system comprising:
a layered decoder circuit operable to apply a data decode algorithm to a first layer of a decoder input to yield a first set of hard decision data and a first syndrome, and to a second layer of a decoder input to yield a second set of hard decision data and a second syndrome;
a data perturbation circuit operable to maintain a list of continuous layers exhibiting a zero value syndrome, wherein the list of continuous layers exhibiting a zero value syndrome includes hard decision data for each layer in the list.
2. The system of claim 1 , wherein the data perturbation circuit is further operable to:
empty the list of continuous layers exhibiting a zero value syndrome when the second syndrome is non-zero.
3. The system of claim 1 , wherein the list of continuous layers exhibiting a zero value includes the first set of hard decision data, the data perturbation circuit is further operable to:
compare the second set of hard decision data with corresponding elements in the first set of hard decision data to yield a set of conflicts, wherein the set of conflicts indicates locations where the second set of hard decision data does not agree with a corresponding element in the first set of hard decision data.
4. The system of claim 3 , wherein the data perturbation circuit is further operable to:
dampen a soft decision corresponding to an element of the second hard decision data at a location in the set of conflicts.
5. The system of claim 4 , wherein the data perturbation circuit includes a multiplier circuit, and wherein dampening the soft decision corresponding to the element of the second hard decision data at the location in the set of conflicts includes multiplying the soft decision by a scalar value.
6. The system of claim 5 , wherein the scalar value is user programmable.
7. The system of claim 5 , wherein the scalar value varies as a function of the number of layers represented in the list of continuous layers exhibiting a zero value syndrome.
8. The system of claim 1 , wherein the data perturbation circuit is further operable to:
augment the list of continuous layers exhibiting a zero value syndrome to include the second set of hard decision data when the second syndrome is non-zero.
9. The system of claim 1 , wherein the data perturbation circuit is further operable to:
determine that the second syndrome is non-zero; and
perform at least one decision flip in the second set of hard decision data to yield a zero value for the second syndrome and an updated set of hard decision data.
10. The system of claim 9 , wherein the data perturbation circuit is further operable to:
augment the list of continuous layers exhibiting a zero value syndrome to include the updated set of hard decision data.
11. The system of claim 1 , wherein the data decode algorithm is a low density parity check algorithm.
12. The system of claim 1 , wherein the system is implemented as part of a device selected from a group consisting of: a communication device, and a storage device.
13. The system of claim 1 , wherein the system is implemented as part of an integrated circuit.
14. A method for data processing, the method comprising:
applying a data decode algorithm by a data decoder circuit to a first layer of a decoder input to yield a first set of hard decision data and a first syndrome, and to a second layer of a decoder input to yield a second set of hard decision data and a second syndrome; and
maintaining a list of continuous layers exhibiting a zero value syndrome, wherein the list of continuous layers exhibiting a zero value syndrome includes at least the first set of hard decision data.
15. The method of claim 14 , the method further comprising:
emptying the list of continuous layers exhibiting a zero value syndrome when the second syndrome is non-zero.
16. The method of claim 14 , wherein the list of continuous layers exhibiting a zero value includes the first set of hard decision data, the method further comprising:
comparing the second set of hard decision data with corresponding elements in the first set of hard decision data to yield a set of conflicts, wherein the set of conflicts indicates locations where the second set of hard decision data does not agree with a corresponding element in the first set of hard decision data.
17. The method of claim 16 , the method further comprising:
dampening a soft decision corresponding to an element of the second hard decision data at a location in the set of conflicts.
18. The method of claim 14 , the method further comprising:
augmenting the list of continuous layers exhibiting a zero value syndrome to include the second set of hard decision data when the second syndrome is non-zero.
19. The method of claim 14 , the method further comprising:
determining that the second syndrome is non-zero; and
performing at least one decision flip in the second set of hard decision data to yield a zero value for the second syndrome and an updated set of hard decision data.
20. A storage device, the storage device comprising:
a storage medium;
a data access circuit operable to access information from the storage medium as a processing input;
a data processing circuit including:
a layered decoder circuit operable to apply a data decode algorithm to a first layer of a decoder input to yield a first set of hard decision data and a first syndrome, and to a second layer of a decoder input to yield a second set of hard decision data and a second syndrome, wherein the decoder input is derived from the processing input;
a data perturbation circuit operable to maintain a list of continuous layers exhibiting a zero value syndrome, wherein the list of continuous layers exhibiting a zero value syndrome includes at least the first set of hard decision data.