| 摘要 |
A method for decoding multiword information comprises steps (a) to (h). In step (a), a multiword information cluster including high protective words and low protective words is provided, wherein the multiword information, high protective words and low protective words can be ECC data, BIS data and LDC data, respectively. In step (b), the low protective words are partitioned into multiple groups. In step(c), the low protective words are de-interleaved, so as to generate a low protective word cluster including multiple segments corresponding to the multiple groups. In step (d), any error of the low protective words is detected, so as to generate segment erasure indicators with localities. In step (e), the low protective words and the segment erasure indicators are stored into a first memory, e.g., a DRAM. In step (f), the segment erasure indicators are stored into a second memory, e.g., a SRAM. In step (g), erasure bits of the low protective word are generated based on the segment erasure indicators. In step (h), the low protective words read from the first memory are decoded by means of the erasure bits.
|
