Soft decoding error prediction
A memory system having a memory block and a memory controller in communication with the memory block. The memory controller is configured to: read and decode codewords from the memory block, determine a fail bit count (FBC), a strong correct (SC) rate indicating a percentage of failed bits correctable through log likelihood ratios (LLRs), and a number of spare bytes in the codewords decoded from the memory, predict a soft decoding error based on a fixed FBC, a fixed SC rate, and the number of spare bytes, and determine soft errors in the codewords read from the memory block based on the predicted soft decoding error.
1. A memory system comprising:
a memory block; and
a memory controller in communication with the memory block,
wherein the memory controller is configured to:
read and decode codewords from the memory block,
determine a fail bit count (FBC), a strong correct (SC) rate indicating a percentage of failed bits correctable through log likelihood ratios (LLRs), and a number of spare bytes in the codewords decoded from the memory block,
predict a soft decoding error based on a fixed FBC, a fixed SC rate, and the number of spare bytes, and
determine soft errors in the codewords read from the memory block based on the predicted soft decoding error.
2. The memory system of claim 1 , wherein the memory controller is configured to relocate data in the codewords from the memory block having the soft errors determined based on the predicted soft decoding error.
3. The memory system of claim 1 , wherein the memory controller is configured to prepare, based on the FBC, the SC rate, and the number of spare bytes, a two-dimensional soft decoding error prediction look-up table representing a failure probability for the codewords read from the memory block.
4. The memory system of claim 3 , wherein the memory controller is configured to determine soft decoding failure based on entries in the look-up table.
5. The memory system of claim 3 , wherein the entries in the two-dimensional soft decoding error prediction table comprise:
ranges, for the number of spare bytes, the SC rate, and the FBC, which are indicative of the soft decoding failure.
6. The memory system of claim 5 , wherein the memory controller is configured to determine the ranges based on a correlation between a codeword failure rate and the FBC at different raw bit error rates.
7. The memory system of claim 6 , wherein the different raw bit error rates comprise different sets of the LLRs for correction of the codewords read from the memory block.
8. The memory system of claim 1 , wherein the memory controller is configured to:
stress test a memory component at different operating conditions to determine the FBC and the SC rate; and
determine, based on the stress testing at the different operating conditions, a maximum FBC for the memory component.
9. The memory system of claim 8 , wherein the memory controller is configured to:
declare the memory component qualified when the maximum FBC is below a threshold; and
declare the memory component not-qualified when the maximum FBC is at or above the threshold.
10. The memory system of claim 1 , wherein the memory block comprises a NAND memory, and the wherein the memory controller is configured to:
provide a NAND reliability specification comprising at different spare byte percentages a) a max raw bit error rate RBER, b) the FBC for the NAND memory, and c) the SC rate.
11. A method for identifying soft errors in a memory block, comprising:
reading and decoding codewords from the memory block;
determining a fail bit count (FBC), a strong correct (SC) rate indicating a percentage of failed bits correctable through log likelihood ratios (LLRs), and a number of spare bytes in the codewords decoded from the memory block;
predicting a soft decoding error in the codewords read from the memory block based on a fixed FBC, a fixed SC rate, and the number of spare bytes; and
determining the soft errors based on the predicted soft decoding error.
12. The method of claim 11 , further comprising:
only relocating data in the codewords from the memory block having the soft errors determined based on the predicted soft decoding error.
13. The method of claim 11 , wherein the predicting a soft decoding error comprises:
preparing, based on the FBC, the SC rate, and the number of spare bytes, a two-dimensional soft decoding error prediction look-up table representing a failure probability for the codewords read from the memory block.
14. The method of claim 13 , further comprising:
determining soft decoding failure based on entries in the look-up table.
15. The method of claim 13 , wherein the entries in the two-dimensional soft decoding error prediction table comprise:
ranges, for the number of spare bytes, the SC rate, and the FBC, which are indicative of the soft decoding failure.
16. The method of claim 15 , further comprising determining the ranges based on a correlation between a codeword failure rate and the FBC at different raw bit error rates.
17. The method of claim 16 , wherein the different raw bit error rates comprise different sets of the LLRs for correction of the codewords read from the memory block.
18. The method of claim 11 , further comprising:
stress testing a memory component at different operating conditions to determine the FBC and the SC rate; and
determining, based on the stress testing at the different operating conditions, a maximum FBC for the memory component.
19. The method of claim 18 , further comprising:
declaring the memory component qualified when the maximum FBC is below a threshold; and
declaring the memory component not-qualified when the maximum FBC is at or above the threshold.
20. The method of claim 11 , wherein the memory block comprises a NAND memory, and the method further comprises:
providing a NAND reliability specification comprising at different spare byte percentages a) a max raw bit error rate RBER, b) the FBC for the NAND memory, and c) the SC rate.