octave-dev

[OctDev] Bchenco - Communication Toolbox

[Marco M. <de...@gm...>]

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  

-- 
Per favore non mandatemi allegati in Word o PowerPoint.
Si veda http://www.gnu.org/philosophy/no-word-attachments.html

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.

Re: [OctDev] Bchenco - Communication Toolbox

[Marco M. <de...@gm...>]

Il giorno Thu, 11 Sep 2008 09:24:01 +0200
David Bateman <ad...@gm...> ha scritto:

> 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.


Ok! So i give you some consistent data:

I use this polynomial generator ( given from LSB to MSB, you don't
have answered to that question so i believe this is correct):

 Columns 1 through 20:

   1   1   1   0   0   1   1   1   1   0   1   0   1   0   1   0   0
1   0   0

 Columns 21 through 40:

   0   0   0   0   0   1   1   0   0   1   1   0   1   1   1   0   1
1   1   1

 Columns 41 through 60:

   1   0   1   0   0   0   0   1   1   1   1   0   0   0   1   0   1
1   0   0

 Columns 61 through 80:

   0   0   0   0   1   0   0   1   0   0   0   1   0   0   0   1   0
0   0   0

 Columns 81 through 100:

   1   0   1   0   1   1   0   0   0   0   1   1   1   0   1   1   0
0   0   1

 Columns 101 through 120:

   1   0   1   1   0   0   1   1   0   1   0   0   1   1   1   1   0
0   1   1

 Columns 121 through 140:

   0   0   0   0   1   0   1   0   0   0   1   1   1   0   0   0   1
0   0   0

 Columns 141 through 160:

   1   0   1   0   0   0   1   1   1   0   1   0   0   0   1   0   0
0   0   1

 Columns 161 through 180:

   1   1   0   0   0   0   0   1   0   1   1   1   0   0   0   0   0
1   1   0

 Columns 181 through 193:

   0   1   0   0   0   1   1   1   0   0   1   0   1


As you can see this polynomial has grade 192.


I call this polynomial g, with this i try to generate the BCH code with
this command:

code = bchenco(ones(1, 32208), 32400, 32208, g);

this gives:

octave:140> e = bchenco(ones(1,32208),32400,32208,g);
error: bchenco: can not find valid generator polynomial for parameters
error: evaluating assignment expression near line 140, column 3

Obviously the part "line 140" si referred to the octave prompt and
change everytime i call this function

I'm codifying a ones word but it's not important, i'm just testing.

Another valid couple (n, k) i'm trying is (48600, 48408) and so:

octave:145> code = bchenco(ones(1,48408),48600,48408,fliplr(g));
octave:146> length(code)
ans =  48600

And this is the result with other 4 known couples, no problems at all.

With another polynomial generator, let's call it g_1, i have other 7
known couples to test, and i have a problem with only one couple her
too:

g_1 = 
 
 Columns 1 through 20:

   1   0   1   0   0   1   0   1   1   0   1   0   0   0   0   0   1
0   0   1

 Columns 21 through 40:

   1   0   0   0   1   0   0   0   1   0   1   1   1   1   1   0   1
0   1   1

 Columns 41 through 60:

   1   1   1   0   0   1   1   1   1   1   1   1   0   0   0   1   0
1   0   0

 Columns 61 through 80:

   1   0   1   0   1   0   0   1   0   1   1   0   0   0   0   0   1
0   0   1

 Columns 81 through 100:

   1   1   0   0   0   1   0   1   1   1   0   0   0   1   0   0   1
0   1   1

 Columns 101 through 120:

   0   0   1   1   0   1   0   0   0   1   1   0   0   1   0   0   1
1   0   1

 Columns 121 through 140:

   1   0   0   1   0   1   1   0   0   0   0   1   1   0   0   1   0
1   0   1

 Columns 141 through 160:

   0   1   1   1   1   1   0   1   1   0   1   1   0   1   0   0   0
1   1   0

 Columns 161 through 169:

   0   0   0   0   0   0   1   0   1



This polynomial has grade 168. Again:

octave:148> code = bchenco(ones(1, 7032), 7200, 7032, g_1);
error: bchenco: can not find valid generator polynomial for parameters
error: evaluating assignment expression near line 148, column 3


Another valid couple (n, k) i'm trying is (3072, 3240) and so:

octave:148> code = bchenco(ones(1, 3072), 3240, 3072, g_1);
octave:149> length(code)
ans =  3240


And this is the result with other 5 known couples, no problems at all.


I hope this is what you asked for.

Thanks

M.

-- 
Per favore non mandatemi allegati in Word o PowerPoint.
Si veda http://www.gnu.org/philosophy/no-word-attachments.html

Re: [OctDev] Bchenco - Communication Toolbox

[David B. <Dav...@mo...>]

Marco Maso wrote:
> Il giorno Thu, 11 Sep 2008 09:24:01 +0200
> David Bateman <ad...@gm...> ha scritto:
>
>   
>> 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.
>>     
>
>
> Ok! So i give you some consistent data:
>
> I use this polynomial generator ( given from LSB to MSB, you don't
> have answered to that question so i believe this is correct):
>
>  Columns 1 through 20:
>
>    1   1   1   0   0   1   1   1   1   0   1   0   1   0   1   0   0
> 1   0   0
>
>  Columns 21 through 40:
>
>    0   0   0   0   0   1   1   0   0   1   1   0   1   1   1   0   1
> 1   1   1
>
>  Columns 41 through 60:
>
>    1   0   1   0   0   0   0   1   1   1   1   0   0   0   1   0   1
> 1   0   0
>
>  Columns 61 through 80:
>
>    0   0   0   0   1   0   0   1   0   0   0   1   0   0   0   1   0
> 0   0   0
>
>  Columns 81 through 100:
>
>    1   0   1   0   1   1   0   0   0   0   1   1   1   0   1   1   0
> 0   0   1
>
>  Columns 101 through 120:
>
>    1   0   1   1   0   0   1   1   0   1   0   0   1   1   1   1   0
> 0   1   1
>
>  Columns 121 through 140:
>
>    0   0   0   0   1   0   1   0   0   0   1   1   1   0   0   0   1
> 0   0   0
>
>  Columns 141 through 160:
>
>    1   0   1   0   0   0   1   1   1   0   1   0   0   0   1   0   0
> 0   0   1
>
>  Columns 161 through 180:
>
>    1   1   0   0   0   0   0   1   0   1   1   1   0   0   0   0   0
> 1   1   0
>
>  Columns 181 through 193:
>
>    0   1   0   0   0   1   1   1   0   0   1   0   1
>
>
> As you can see this polynomial has grade 192.
>
>
> I call this polynomial g, with this i try to generate the BCH code with
> this command:
>
> code = bchenco(ones(1, 32208), 32400, 32208, g);
>
> this gives:
>
> octave:140> e = bchenco(ones(1,32208),32400,32208,g);
> error: bchenco: can not find valid generator polynomial for parameters
> error: evaluating assignment expression near line 140, column 3
>
> Obviously the part "line 140" si referred to the octave prompt and
> change everytime i call this function
>
> I'm codifying a ones word but it's not important, i'm just testing.
>
> Another valid couple (n, k) i'm trying is (48600, 48408) and so:
>
> octave:145> code = bchenco(ones(1,48408),48600,48408,fliplr(g));
> octave:146> length(code)
> ans =  48600
>
> And this is the result with other 4 known couples, no problems at all.
>
> With another polynomial generator, let's call it g_1, i have other 7
> known couples to test, and i have a problem with only one couple her
> too:
>
> g_1 = 
>  
>  Columns 1 through 20:
>
>    1   0   1   0   0   1   0   1   1   0   1   0   0   0   0   0   1
> 0   0   1
>
>  Columns 21 through 40:
>
>    1   0   0   0   1   0   0   0   1   0   1   1   1   1   1   0   1
> 0   1   1
>
>  Columns 41 through 60:
>
>    1   1   1   0   0   1   1   1   1   1   1   1   0   0   0   1   0
> 1   0   0
>
>  Columns 61 through 80:
>
>    1   0   1   0   1   0   0   1   0   1   1   0   0   0   0   0   1
> 0   0   1
>
>  Columns 81 through 100:
>
>    1   1   0   0   0   1   0   1   1   1   0   0   0   1   0   0   1
> 0   1   1
>
>  Columns 101 through 120:
>
>    0   0   1   1   0   1   0   0   0   1   1   0   0   1   0   0   1
> 1   0   1
>
>  Columns 121 through 140:
>
>    1   0   0   1   0   1   1   0   0   0   0   1   1   0   0   1   0
> 1   0   1
>
>  Columns 141 through 160:
>
>    0   1   1   1   1   1   0   1   1   0   1   1   0   1   0   0   0
> 1   1   0
>
>  Columns 161 through 169:
>
>    0   0   0   0   0   0   1   0   1
>
>
>
> This polynomial has grade 168. Again:
>
> octave:148> code = bchenco(ones(1, 7032), 7200, 7032, g_1);
> error: bchenco: can not find valid generator polynomial for parameters
> error: evaluating assignment expression near line 148, column 3
>
>
> Another valid couple (n, k) i'm trying is (3072, 3240) and so:
>
> octave:148> code = bchenco(ones(1, 3072), 3240, 3072, g_1);
> octave:149> length(code)
> ans =  3240
>
>   
>

With the default primitive polynomial, there is no valid BCH code for 
your values of N and K. You need to either change N and K or try a 
different valid primitive polynomial. For example,

M = 13
N = 2^M - 1; % Next value of N above 7200
K = N - 168;
p = primpoly(13, 'all');
prim = NaN
for i = 1: length(p)
  try
     g = bchpoly(N, K, p(i))
     prim = p(i);
     break;
  catch
     continue
  end
end

this is rather slow but finds that there is in fact NO valid generator 
with any of the primitive polynomials of 2^13-1 and your values of N and 
K, and prim will end up being NaN. I'd suggest you change your value of 
K to N - 169 and then yes there is a valid BCH code that can be found. 
In particular the one given by

N = 2^13 - 1;
g = bchpoly(N, N - 169);

Note not all values of N abd K are valid for BCH codes.

D.

[OctDev] Bchenco - Communication Toolbox

[marco m. <de...@gm...>]

Hi, last month i had a doubt about Bchenco and thanks to your help i've
managed to solve my problems...but now i have again that silly last
question:

In & out parameters in octave function are always LSB on the left?
I mean, referring to bchenco fro example:

bchenco([1 0 1 1],7,4)

1 0 1 1 = 11    or   1 0 1 1 = 13

i thought that since bchpoly gives a polynomial with the LSB on the left all
the other functions would work coherently...am i wrong?

Thanks

Marco

Re: [OctDev] Bchenco - Communication Toolbox

[David B. <Dav...@mo...>]

marco maso wrote:
> Hi, last month i had a doubt about Bchenco and thanks to your help 
> i've managed to solve my problems...but now i have again that silly 
> last question:
>
> In & out parameters in octave function are always LSB on the left?
> I mean, referring to bchenco fro example:
>
> bchenco([1 0 1 1],7,4)
>
> 1 0 1 1 = 11    or   1 0 1 1 = 13
>
> i thought that since bchpoly gives a polynomial with the LSB on the 
> left all the other functions would work coherently...am i wrong?
>
> Thanks
>
> Marco
>
All of this is documented.. Type "doc comms" at the prompt to get the on 
line version of the manual or look at it online at

http://octave.sourceforge.net/comm/local/comms_4.html#SEC13

In short

octave:2> bchpoly(14,6)
ans =

1 0 0 0 1 0 1 1 1

which is

1 + x ^ 4 + x ^ 6 + x ^ 7 + x ^ 8

so yes it is LSB first. So the above case "1 0 1 1" = 13. The functions 
are suppose to be coherent in their definitions, if they aren't then 
please give an example of the incoherence..

Regards
David



D.


-- 
David Bateman                                Dav...@mo...
Motorola Labs - Paris                        +33 1 69 35 48 04 (Ph) 
Parc Les Algorithmes, Commune de St Aubin    +33 6 72 01 06 33 (Mob) 
91193 Gif-Sur-Yvette FRANCE                  +33 1 69 35 77 01 (Fax) 

The information contained in this communication has been classified as: 

[x] General Business Information 
[ ] Motorola Internal Use Only 
[ ] Motorola Confidential Proprietary