A simple scheme for correcting blocks of multiple adjacent bit errors
There are several practical scenarios where the behavior of a communications channel can be somehow predictable. For instance, in nanometric technologies the increase of integration density makes memory devices more sensitive to radiation effects that can cause multiple bit upsets (MBUs) in adjacent...
Saved in:
| Published in: | 2014 International Symposium on Information Theory and its Applications pp. 176 - 180 |
|---|---|
| Main Authors: | , |
| Format: | Conference Proceeding |
| Language: | English |
| Published: |
IEICE
01.10.2014
|
| Subjects: | |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | There are several practical scenarios where the behavior of a communications channel can be somehow predictable. For instance, in nanometric technologies the increase of integration density makes memory devices more sensitive to radiation effects that can cause multiple bit upsets (MBUs) in adjacent cells. In this paper we use random error correcting codes to deal with such particular channels. We propose a very simple and efficient scheme for correcting blocks of adjacent errors. Our approach consists in applying a transformation to the code in order to adapt it to the channel. Undoing the transformation at the receiving end converts blocks of adjacent errors (of any length) into low weight error vectors that will subsequently be corrected by a bounded distance decoding algorithm. The proposed method can be applied, without noticeable loss of performance, to any existing encoding-decoding scheme. Moreover, and in a list decoding fashion, we obtain stronger results for binary cyclic codes. |
|---|