Data compensating method for flash memory

1. A data compensating method for a flash memory, the flash memory comprising a plurality of cells, each cell having a first storing state or a second storing state, the first storing state having a low voltage level, the second storing state having a high voltage level, the data compensating method comprising steps of:

acquiring a first threshold voltage distribution curve of the cells of the flash memory with a first storing state;

acquiring a second threshold voltage distribution curve of the cells of the flash memory with a second storing state, wherein the first threshold voltage distribution curve and the second threshold voltage distribution curve are adjacent to each other;

setting a statistic voltage range;

calculating a first occurrence probability of a first type ICI pattern of the first storing state according to the statistic voltage range and the first threshold voltage distribution curve;

calculating a second occurrence probability of the first type ICI pattern of the second storing state according to the statistic voltage range and the second threshold voltage distribution curve; and

compensating storing states of central cells corresponding to the first type ICI pattern according to the first occurrence probability and the second occurrence probability during a read cycle;

wherein the calculation of the first occurrence probability comprises steps of: acquiring a first number of the central cells corresponding to the first type ICI pattern according to the first threshold voltage distribution curve; acquiring a second number of the central cells corresponding to the first type ICI pattern and having a threshold voltage within the statistic voltage range according to the first threshold voltage distribution curve and the statistic voltage range; and dividing the second number by the first number, thereby acquiring the first occurrence probability.

2. The data compensating method as claimed in claim 1 , wherein if a product of the first occurrence probability and one minus the second occurrence probability is larger than a first predetermined value, the central cells corresponding to the first type ICI pattern are changed from the second storing state to the first storing state, and a data correcting operation is performed by a hard decoding unit.

3. The data compensating method as claimed in claim 1 , wherein if a product of the first occurrence probability and one minus the second occurrence probability is larger than a first predetermined value, the data compensating method further comprises a step of providing a near-first storing state indicator to a soft decoding unit, wherein a data correcting operation is performed by the soft decoding unit.

4. The data compensating method as claimed in claim 1 , wherein if the first occurrence probability is larger than the second occurrence probability, the central cells corresponding to the first type ICI pattern are changed from the second storing state to the first storing state, and a data correcting operation is performed by a hard decoding unit.

5. The data compensating method as claimed in claim 1 , wherein if the first occurrence probability is larger than the second occurrence probability, the data compensating method further comprises a step of providing a near-first storing state indicator to a soft decoding unit, wherein a data correcting operation is performed by the soft decoding unit.

6. The data compensating method as claimed in claim 1 , further comprising steps of:

acquiring a third number of the central cells corresponding to the first type ICI pattern according to the second threshold voltage distribution curve;

acquiring a fourth number of the central cells corresponding to the first type ICI pattern and having a threshold voltage within the statistic voltage range according to the second threshold voltage distribution curve and the statistic voltage range; and

dividing the fourth number by the third number, thereby acquiring the second occurrence probability.

7. The data compensating method as claimed in claim 1 , wherein a slicing voltage for distinguishing the first storing state from the second storing state lies within the statistic voltage range.

8. A data compensating method for a flash memory, the flash memory comprising a plurality of cells, each cell having a first storing state or a second storing state, the first storing state having a first threshold voltage distribution curve, the second storing state having a second threshold voltage distribution curve, the data compensating method comprising steps of:

setting a statistic voltage range, wherein the statistic voltage range is between a first median threshold voltage of the first threshold voltage distribution curve and a second median threshold voltage of the second threshold voltage distribution curve;

calculating a first occurrence probability of a first type ICI pattern of the first storing state corresponding to a threshold voltage within the statistic voltage range;

calculating a second occurrence probability of the first type ICI pattern of the second storing state corresponding to the threshold voltage within the statistic voltage range; and

compensating storing states of central cells corresponding to the first type ICI pattern and having the threshold voltage within the statistic voltage range according to the first occurrence probability and the second occurrence probability during a read cycle;

wherein the calculation of the first occurrence probability comprises steps of: acquiring a first number of the central cells corresponding to the first type ICI pattern according to the first threshold voltage distribution curve; acquiring a second number of the central cells corresponding to the first type ICI pattern and having a threshold voltage within the statistic voltage range according to the first threshold voltage distribution curve and the statistic voltage range; and dividing the second number by the first number, thereby acquiring the first occurrence probability.

9. The data compensating method as claimed in claim 8 , further comprising steps of:

acquiring a third number of the central cells corresponding to the first type ICI pattern according to the second threshold voltage distribution curve;

acquiring a fourth number of the central cells corresponding to the first type ICI pattern and having a threshold voltage within the statistic voltage range according to the second threshold voltage distribution curve and the statistic voltage range; and

dividing the fourth number by the third number, thereby acquiring the second occurrence probability.

10. The data compensating method as claimed in claim 8 , wherein a slicing voltage for distinguishing the first storing state from the second storing state lies within the statistic voltage range.

11. The data compensating method as claimed in claim 8 , wherein if a product of the first occurrence probability and one minus the second occurrence probability is larger than a first predetermined value, the central cells corresponding to the first type ICI pattern and having the threshold voltage within the statistic voltage range are changed from the second storing state to the first storing state.

12. The data compensating method as claimed in claim 8 , wherein if the first occurrence probability is larger than the second occurrence probability, the central cells corresponding to the first type ICI pattern and having the threshold voltage within the statistic voltage range are changed from the second storing state to the first storing state.

13. The data compensating method as claimed in claim 8 , wherein if the first occurrence probability is larger than the second occurrence probability, the data compensating method further comprises a step of providing a near-first storing state indicator to a soft decoding unit, wherein a data correcting operation is performed by the soft decoding unit.