Standards/BCH
From UIT
(Difference between revisions)
(→Coder) |
|||
Line 24: | Line 24: | ||
The following picture shows a BCH encoder with 7 code bits : | The following picture shows a BCH encoder with 7 code bits : | ||
− | [[File:BCH encoder schematic.svg|center|BCH encoder (63, 56)]] | + | [[File:BCH encoder schematic.svg|center|800 px|BCH encoder (63, 56)]] |
= Decoder = | = Decoder = | ||
+ | The following picture shows the decoder for the previous encoder : | ||
+ | |||
+ | [[File:BCH decoder schematic.svg|center|800 px]] | ||
[[Category:Standards]] [[Category:Codes]] | [[Category:Standards]] [[Category:Codes]] |
Latest revision as of 14:37, 7 June 2017
|
BCH codes have been invented by Bose, Chaudhuri and Hocquenghem. They are used in communication systems for error correction. Similarly to Hamming codes, they are able to detect a certain number of errors and to correct a smaller quantity of them.
These codes are often implemented in addition to convolutional codes.
Definitions
A BCH code is specified by:
-
n
: total code length -
k
: information data length -
m
: parity bits number -
G
: the generator polynomial
The coded BCH data is often filled with '0' bits up to a length which is a multiple of 8, thus allowing to be transmitted as bytes.
Coder
The parity bits are created with the help of a Linear Feedback Shift Register (LFSR).
The following picture shows a BCH encoder with 7 code bits :
Decoder
The following picture shows the decoder for the previous encoder :