Abstract | ||
---|---|---|
We present a novel autoencoder-based approach for designing codes that provide unequal error protection (UEP) capabilities. The proposed approach, based on a generalization of an autoencoder loss function, provides a versatile framework for the design of message-wise and bit-wise UEP codes. Using an associated weight vector, the generalized loss function can be used to trade off error probabilitie... |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/LCOMM.2021.3108845 | IEEE Communications Letters |
Keywords | DocType | Volume |
Error probability,Error correction codes,Decoding,Receivers,Transmitters,Training,Neural networks | Journal | 25 |
Issue | ISSN | Citations |
11 | 1089-7798 | 0 |
PageRank | References | Authors |
0.34 | 0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Vukan Ninkovic | 1 | 0 | 0.34 |
Dejan Vukobratović | 2 | 378 | 36.82 |
Christian Häger | 3 | 1 | 3.06 |
Henk Wymeersch | 4 | 12 | 1.96 |
Amat Alexandre Graell i | 5 | 456 | 65.56 |