Re: [OctDev] Bchenco - Communication Toolbox

[David B. <ad...@gm...>]

Marco Maso wrote:
> Hi everybody, my name is Marco and this is my first message to the ML.
> I've decided to subscribe because i have a little problem usign the
> bchenco function provided with the communications toolbox of
> octave-forge (versio 1.0.6). I'm currently using Octave 3.0.2. As i can
> read on the Function Reference page this function:
> 
> "Encodes the message msg using a [n,k] BCH coding. The variable msg is a
> binary array with k columns and an arbitrary number of rows. Each row
> of msg represents a single symbol to be coded by the BCH coder. The
> coded message is returned in the binary array code containing n columns
> and the same number of rows as msg."
> 
> And after:
> 
> "By default the generator polynomial used in the BCH coding is based on
> the properties of the Galois Field GF(2^m). This default generator
> polynomial can be overridden by a polynomial in g. Suitable generator
> polynomials can be constructed with bchpoly."
> 
> I'm developing a software that needs a BCH encoder with the following
> parameters (among the others...):
> 
> - BCH uncoded word length = Kbch = 32208
> - BCH coded word length = Nbch = 32400
> 
> - BCH uncoded word length = Kbch = 7302
> - BCH coded word length = Nbch = 7200
> 
> In the first case i use a g polynomial obtained by multiplying 12
> polynomials of grade 16. This polynomial generator works correctly with
> the other 5 couples (n,k) i need to test on my simulator.
> 
> In the second case In the first case i use a g polynomial obtained by
> multiplying 12 polynomials of grade 14. This polynomial generator works
> correctly with the other 6 couples (n,k) i need to test on my simulator.
> 
> Obviously thes are not numbers invented by myself but i'm using a
> manual as a reference so i'm sure they are correct.
> 
> 
> In both cases i obtain this error:
> 
> error: bchenco: can not find valid generator polynomial for parameters
> error: evaluating assignment expression near line 140, column 3
> 
> I can't properly understand where the problem can be even because i
> can't find the source code of this function...
> 
> Can anybody explain me why i receive such an error?
> 
> Should i post the coefficients of my g polynomial?
> 
> Last question (stupid) : when working with octave a polynomial
> insertion has always to be done from LSB to MSB am i right?
> 
> Thanks a lot
> 
> Marco  
> 
> 

Could you supply a short example demonstrating this issue?

D.