DVB-S2 - BCH and LDPC coding fixed codeword length (modulation toolkit)

[diablo_17_bj]

If you open dvb-s2 scheme in Matlab( command:  "commdvbs2") , you'll see that you can define n,k,t matrix for BCH encoder however you want, and you can get good results, but here, in LabVIEW, you can't do that.

[lis@jiang]

It doesn't look like n, k, and t should be able to be configured for any arbitrary value.  I can't speak for other programming languages outside of LabVIEW, but it looks like these values are calculated in relation to each other.  This looks like a pretty good explanation of how these values are calculated.  The textbook Radio System Design for Telecommunication, which can be found on Google books, also discusses this in less detail.  Which specific n, k, and t values are you trying to use?

Lisa

[diablo_17_bj]

On that .jpg file which is attached in my first post on the first page, there is a table which defines parameters for BCH and LDPC encoders for different code rates. I know that i can't put any value for n,k and t, but some specific values, neither can do that in Matlab, but Matlab can perform shortening and puntcturing of BCH output, and that's why I have a problem in LabVIEW: because I need to shorten input and output my BCH encoder to satisfy the norm which defines DVB-S2.

[lis@jiang]

Hi diablo_17_bj

Sorry, I had initially forgotten about your first post.

The BCH Encoder VI in LabVIEW does limit the input of n as n = 2^m - 1.  I'm not sure if this is an intentional limitation, but I'm asking the Toolkit developers about it.

Stay tuned!

Lisa J

[Sarah_Y]

Hi diablo_17_bj,

Unfortunately, the BCH modes as mentioned in the DVB-S2 standard are not in the Modulation Toolkit.  Since the standard also has some specifics like the multiplication of different polynomials, it will most likely need to be implemented by looking at the standard.  Modulation Toolkit is a starting point for different modulation types, but does not have every modulation standard built in, which can cause difficulties like you are seeing.

[diablo_17_bj]

Yes, I can see that. Either the LDPC encoder is not working properly: it doesn't give correct output for different code rates, except for code rate R=1/2.

[Vimal_F]

Let me look into this diablo and touch base with Lisa.

Thanks

V

[gkhnclk123]

Hi,

I have another question to you about LDPC Encoding/decoding which is similar to this topic. I am using LDPC channel coding in my project. As  you know, The LDPC blocks of labview Modulation toolkit uses parity check matrices at the transmitter and receiver to encode and decode bitstream. However, I recognized that there are some inconsistency in terms of the size of the input/output bitstream. For ex, I am using a parity check matrix with size of 144x576. Let’s say that the size of our bitstream to be encoded is 1000 bits. The encoded bitstream  has size of 1152 bits, which is 2 times of 576, in this case. After decoding the 1152 bits, I get a decoded output bitstream with size of 864. The original data is truncated and we are losing about 140 bits of our data. You can check this results by using MT LDPC Encode-Decode application by changing the parameters which is stated at  C:\Program Files (x86)\National Instruments\LabVIEW 2013\examples\Modulation\simulation examples\Programming

Thanks,

Gokhan Celik

Research Assistant

[SeaSick]

Hi Gokhan,

Thank you for posting your question on the NI forums. 

You may receive a higher response rate from the community if you create a new forum for your question since this is an older forum that is not monitored as frequently.

Feel free to link this forum in your future post in order to give some background information to your new forum. 

Regards