Processor instructions for iterative decoding operations
A storage circuit is configured to store multiple vectors associated with variable and check nodes of an iterative decoding operation. As part of the iterative decoding operation, a processor circuit is configured to retrieve, from the storage circuit, an intermediate value vector, a first estimation vector, a second estimation vector, and a sign vector, and determine an absolute value of the intermediate value vector. The processor circuit is also configured, using the retrieved vectors, to generate updated values for the first and second estimation vectors as part of determining a bit estimate for a check node included in the iterative decoding operation.
1. An apparatus, comprising:
a storage circuit configured to store a plurality of vectors associated with an iterative decoding operation; and
one or more processor cores, wherein a given processor core of the processor cores is configured to:
retrieve a subset of the plurality of vectors that includes a first intermediate value vector, a first estimation vector, a second estimation vector, and a sign vector;
determine an absolute value of the first intermediate value vector;
generate an updated value for the first estimation vector using the absolute value of the first intermediate value vector;
generate an updated value for the second estimation vector using the absolute value of the first intermediate value vector, the updated value of the first estimation vector, and a current value of the second estimation vector; and
generate an updated value for the sign vector of the plurality of vectors using a current value of the sign vector and a sign of the first intermediate value vector.
2. The apparatus of claim 1 , wherein to generate the updated value for the first estimation vector, the given processor core is further configured to:
determine a difference between the absolute value of the first intermediate value vector and the updated value for the first estimation vector;
determine a product of the difference and a coefficient; and
generate a corrected value for the first estimation vector using the updated value of the first estimation vector and the product.
3. The apparatus of claim 1 , wherein to generate the updated value for the first estimation vector, the given processor core is further configured to generate a corrected value for the first estimation vector using a coefficient and the updated value of the first estimation vector.
4. The apparatus of claim 1 , wherein the given processor core is further configured to:
determine an absolute value of a second intermediate value vector of the plurality of vectors; and
generate, in parallel to a generation of the updated value for the first estimation vector, an updated value for a third estimation vector of the plurality of vectors using a current value of the third estimation vector and the absolute value of the second intermediate value vector.
5. The apparatus of claim 1 , wherein the first intermediate value vector is included in a plurality of intermediate vectors stored in a corresponding one of a plurality of registers included in the storage circuit, and wherein the given processor core is further configured to:
change respective registers in which the plurality of intermediate vectors are stored; and
retrieve the first intermediate value vector from its new storage location prior to a determination of the absolute value of the first intermediate value vector.
6. The apparatus of claim 1 , wherein to generate the updated value for the first estimation vector, the given processor core is further configured to:
perform a comparison of the updated value for the first estimation vector to a threshold value; and
adjust the updated value for the first estimation vector using results of the comparison.
7. A method, comprising:
receiving, by a given processor core of a plurality of processor cores, a first intermediate value vector, a first estimation vector, a second estimation vector, and a sign vector;
determining, by the given processor core, an absolute value of the first intermediate value vector;
generating, by the given processor core, an updated value for the first estimation vector with a minimum of a current value of the first estimation vector and the absolute value of the first intermediate value vector;
generating, by the given processor core, an updated value for the second estimation vector with a minimum of the updated value of the first estimation vector, the absolute value of the first intermediate value vector, and a current value of the second estimation vector; and
generating, by the given processor core, an updated value for the sign vector using a current value of the sign vector and a sign of the first intermediate value vector.
8. The method of claim 7 , wherein generating the updated value for the first estimation vector includes:
determining, by the given processor core, a difference between the absolute value of the first intermediate value vector and the updated value for the first estimation vector;
determining, by the given processor core, a product of the difference and a coefficient; and
generating, by the given processor core, a corrected value for the first estimation vector using the updated value of the first estimation vector and the product.
9. The method of claim 7 , wherein generating the updated value for the first estimation vector includes generating, by the given processor core, a corrected value for the first estimation vector using a coefficient and the updated value of the first estimation vector.
10. The method of claim 7 , further comprising:
receiving a second intermediate value vector, a third estimation vector, a fourth estimation vector, and a second sign vector;
determining a second absolute value of the second intermediate value vector; and
generating, in parallel to generating the updated value for the first estimation vector, an updated value for the third estimation vector with a second minimum of a current value of the third estimation vector and the second absolute value of the second intermediate value vector.
11. The method of claim 7 , wherein the first intermediate value vector is included in a plurality of intermediate value vectors stored in corresponding ones of a plurality of registers, and wherein the method further includes:
changing respective registers in which the plurality of intermediate value vectors are stored; and
retrieving, by the given processor core, the first intermediate value vector from its new storage location prior to a determination of the absolute value of the first intermediate value vector.
12. The method of claim 7 , wherein generating the updated value for the first estimation vector includes:
performing, by the given processor core, a comparison of the updated value for the first estimation vector to a threshold value; and
adjusting, by the given processor core, the updated value for the first estimation vector using results of the comparison.
13. The method of claim 12 , wherein adjusting the updated value of the first estimation vector includes selecting a minimum value of the updated value for the first estimation vector and the threshold value.