Me. Buckley et Sb. Wicker, The design and performance of a neural network for predicting turbo decoding error with application to hybrid ARQ protocols, IEEE COMMUN, 48(4), 2000, pp. 566-576
It is shown that a neural network can be trained to observe the cross entro
py of the outputs of component decoders in a turbo error control system and
to accurately predict the presence of errors in the decoded data. The neur
al network can be used as a:trigger for retransmission requests at either t
he beginning or the conclusion of the decoding process, providing improved
reliability and throughput performance at a lower average decoding complexi
ty than turbo decoding with cyclic redundancy check error detection.