Method to increase the usable word width of a memory providing an error correction scheme
AssigneeNXP B.V.
InventorsSoenke Ostertun, Bjorn Fay, Vitaly Ocheretny
Life of this patent
Estimating the term…
Abstract
Various embodiments relate to a method for storing and reading data from a memory. Data words stored in the memory may be grouped, and word specific parity information and shared parity information is generated, and the shared parity information is distributed among the group of words. During reading of a word, if more errors are detected than can be corrected with word parity data, the shared parity data is retrieved and used to make the error corrections.
Claims10
A method for reading data stored in a memory with parity information shared across G data words, comprising:
- reading a first data word including data bits, a word parity information, and a distributed parity information;
- checking the word parity information based upon the read data bits of the first word;
- when checking the word parity information indicates more errors than could be corrected with the word parity information then for each of a second through G data words:
- reading the data word including data bits, a word parity information, and distributed parity information;
- determining a third parity information for the first word including
- combining the distributed parity information of each of the second to G data words;
- determining a fourth parity information based upon the data bits of the read second to G data words;
- and
- calculating the third parity information based upon the fourth parity information and the combined distributed parity information of each of the second to G data words;
- and
- correcting the read data bits of the first word based upon the third parity information.
A method for reading data stored in a memory that stores a group of G data words with w data bits per data word, comprising:
- generating a first generator polynomial g(x);
- generating a second generator polynomial g′(x) that is based upon the first generator polynomial g(x) with a lesser degree;
- reading a first word of the group of G data words including w data bits, first split parity bits p′(x), and a first portion of parity information Q;
- checking the first split parity bits p′(x)based upon the read data bits of the first word of the group of G data words;
- when checking the first split parity bits p′(x) indicates more errors than could be corrected with the first split parity bits p′(x):
- reading second to G data words including of the group of G data words including w data bits, first split parity bits p′(x), and remaining portions of parity information Q;
- calculating Q by combining the portions of parity information Q from each of the G data words;
- generating second split parity bits q(x) of the first data word based upon Q and the 2 through G second split parity bits q(x) from each of the G data words;
- reconstructing the first parity bits p (x) of the first word of the group of G data words based upon q(x), g′(x), and p′(x) of the first data word of the group of G data words;
- and
- correcting the first word of the group of G data words based upon the first parity bits p(x).
Dependencies are inferred from the claim text.