Adaptive coded-modulation for intelligent optical transport networks

1. A computer implemented method for dynamic data rate adjustment within a cascaded forward error correction FEC for optical communications, the method comprising the steps of:

i) subjecting data communicated over an optical network to a forward error correction in an encoding or decoding of the data, the encoding or decoding employing a codeword;

ii) re-encoding part of the codeword for generating a subsequent codeword where an actual code rate is tuned by adjusting a size of data encoded to provide re-encoded data; and

iii) dynamically changing the re-encoded data size to achieve cascaded rate adaptive FEC for communication of the data over the optical network;

wherein the encoding comprises a cascaded low-density parity-check LDPC encoder that generates a 1 st codeword by multiplying a vector of M zeros and K−M information symbols with G, using a regular (N, K) nonbinary LDPC code as a component code G denotes a generator matrix; and

wherein to generate the 2 nd codeword, the encoder first encapsulates the last M symbols of the first codeword and (K−M) incoming information symbols into a K-symbol sequence and then multiplies the K-symbol sequence with G.

2. The method of claim 1 , wherein the encoding includes a (N, K) nonbinary LDPC code with a code rate R=K/N, where a row vector of K information symbols is denoted as u, and an encoder of the component nonbinary LDPC code generates the codeword c of N symbols according to c=u·G where G is a K×N matrix composed of elements from a corresponding Galois field GF(q) while q is chosen to be a constellation size.

3. The method of claim 1 , wherein all the codewords following generation of the 2 nd codeword are generated in such a cascaded way by re-encoding the last M symbols of its previous codeword.

4. The method of claim 1 , wherein step iii) comprises adjusting a value of M according to real-time system status, where M denotes a number of symbols re-encoded, M can be an arbitrary positive number no larger than K information symbols and an actual code rate R′=(K−M)/(N−M)≦R, where R is a an unadjusted code rate.

5. The method of claim 4 , wherein M is set to be zero where no previous symbol is re-encoded and the actual code rate is R′=R=K/N, K representing information symbols and N representing a number of symbols generated by the codeword.

6. The method of claim 1 , wherein responsive to real-time optical-signal-to-noise ratio (OSNR) requirement in the optical network, M is set to be a proper positive integer to guarantee a transmission reliability with an error-correction capability increasing as M goes up.

7. A non-transitory storage medium with instructions for enabling a computer to implement a method for dynamic data rate adjustment within a cascaded forward error correction FEC for optical communications, the method comprising the steps of:

i) subjecting data communicated over an optical network to a forward error correction in an encoding or decoding of the data, the encoding or decoding employing a codeword;

ii) re-encoding part of the codeword for generating a subsequent codeword where an actual code rate is tuned by adjusting a size of data encoded to provide re-encoded data; and

iii) dynamically changing the re-encoded data size to achieve cascaded rate adaptive FEC for communication of the data over the optical network;

wherein to generate the 2 nd codeword, the encoder first encapsulates the last M symbols of the first codeword and (K−M)incoming information symbols into a K-symbol sequence and then multiplies the K-symbol sequence with G; and

wherein the encoding comprises a cascaded low-density parity-check LDPC encoder that generates a 1 st codeword by multiplying a vector of M zeros and K−M information symbols with G, using a regular (N, K) nonbinary LDPC code as a component code G denotes a generator matrix.

8. The storage medium of claim 7 , wherein the encoding includes a (N, K) nonbinary LDPC code with a code rate R=K/N, where a row vector of K information symbols is denoted as u, and an encoder of the component nonbinary LDPC code generates the codeword c of N symbols according to c=u·G where G is a K×N matrix composed of elements from a corresponding Galois field GF(q) while q is chosen to be a constellation size.

9. The storage medium of claim 7 , wherein all the codewords following generation of the 2 nd codeword are generated in such a cascaded way by re-encoding the last M symbols of its previous codeword.

10. The storage medium of claim 7 , wherein step iii) comprises adjusting a value of M according to real-time system status, where M denotes a number of symbols re-encoded, M can be an arbitrary positive number no larger than K information symbols and an actual code rate R′=(K−M)/(N−M)≦R, where R is a an unadjusted code rate.

11. The storage medium of claim 10 , wherein M is set to be zero where no previous symbol is re-encoded and the actual code rate is R′=R=K/N, K representing information symbols and N representing a number of symbols generated by the codeword.

12. The storage medium of claim 7 , wherein, wherein responsive to real-time optical-signal-to-noise ratio (OSNR) requirement in the optical network, M is set to be a proper positive integer to guarantee a transmission reliability with an error-correction capability increasing as M goes up.