Decodes the coded message code using a BCH coder. The message length of the coder is defined in variable k, and the error correction capability of the code is defined in t.
The variable code is a binary array with n columns and an arbitrary number of rows. Each row of code represents a single symbol to be decoded by the BCH coder. The decoded message is returned in the binary array msg containing k columns and the same number of rows as code.
The use of bchdeco can be seen in the following short example.
m = 3; n = 2^m -1; k = 4; t = 1; msg = randint (10, k); code = bchenco (msg, n, k); noisy = mod (randerr (10,n) + code, 2); [dec, err] = bchdeco (msg, k, t);
Valid codes can be found using bchpoly. In general the codeword
length n should be of the form 2^m-1, where m is an
integer. However, shortened BCH codes can be used such that if
[2^m-1,k] is a valid code
[2^m-1-x,k-x]
is also a valid code using
the same generator polynomial.
By default the BCH coding is based on the properties of the Galois
Field GF(2^m). The primitive polynomial used in the Galois
can be overridden by a primitive polynomial in prim. Suitable
primitive polynomials can be constructed with primpoly. The form
of prim maybe be either a integer representation of the primitive
polynomial as given by primpoly, or a binary representation that
might be constructed like
m = 3; prim = de2bi (primpoly (m));
By default the parity symbols are assumed to be placed at the beginning of
the coded message. The variable parpos controls this positioning and
can take the values "beginning" or "end".
See also: bchpoly, bchenco, decode, primpoly.
Package: communications