LDPC and BCH channel coding sim

[alexpas]

Hello,

I am trying to make a DVB-S2 simulation using  LDPC and BCH channel encoding. Is it possible to put values corresponding to the DVBS2 standard when using these 2 toolboxes (MT LDPC Generate Regular Parity Check Matrix & MT BCH)? The parameters for LDPC and the values n,k,t for BCH for the DVBS2 standard are N_ldpc=64800, K_ldpc=43200 and N_bch=43200, K_bch=43040. Can Labview Communications handle these values, please help !

The simulation works for default values for LDPC IrregularParity Matrix (100,50,3) and BCH (7,4,1) but it appears that for values higher than n=1000 it cannot handle it or turns back errors.

[DECONN]

Hey alexpas,

Most of the details behind the Modulation GVI's can be found in the Help documents (for example see here). However, I found a few more details looking at the LabVIEW Modulation Toolkit help. They mention the following:

---

MT BCH Encoder

n,k,t - specifies the values of the code word length (n), data word length (k), and error correcting capacity (t), in bits. The n, k, and t values for BCH encoding are interrelated such that specifying two values constrains the third value. Select from a set of well-known (n,k,t) triples that are widely used for BCH encoding or specify an arbitrary triple by entering three comma-separated integers. The default value is 7, 4, 1.

Details:

The prepopulated n,k,t parameter contains all valid codes for values of n up to 511. However, the Modulation Toolkit supports BCH encoding and decoding up to Galois field sizes of GF(65,536). Therefore, you can specify BCH code word lengths up to 65,535 by entering valid values of the code word length (n), data word length (k), and error correction capacity (t) separated by commas inside the string combo box.

Note: Not all n,k,t combinations specify valid BCH codes. The BCH encoder/decoder VIs perform validation for values of the code word length (n). If n is not of the form 2m-1, the VI returns an error. The only validation criterion for the data word length (k) and error-correcting capacity (t) is that they must be positive integers strictly lesser than n. The Modulation Toolkit does not guarantee specifications of BCH n,k,t values that satisfy the above criterion but are not valid BCH codes.

Values of standard (n,k,t) triples are defined in Wicker, Stephen B. Error Control Systems for Digital Communications and Storage. Englewood Cliffs: Prentice-Hall, 1995.

---

I'm not intimately familiar with specifics on the algorithms so I don't know if this applies directly to the issues you were seeing but I thought it may be worth mentioning. Where exactly in the code are the errors emanating from and what is the specific error and code? This may tell us a little more about what may be happening.