Title :
Regaining synchronization using neural networks to detect watermark codes
Author :
Heymann, R. ; Ferreira, H.C.
Author_Institution :
Dept. of Electr. & Electron. Eng. Sci., Univ. of Johannesburg, Johannesburg
Abstract :
A binary neural network configuration is presented which is capable of correcting synchronization errors in a received bit sequence. The neural network uses the bit positions of a watermark sequence, which is inserted into the sequence before transmission, to determine whether an insertion or a deletion has occurred.
Keywords :
error statistics; neural nets; synchronisation; watermarking; binary neural network; regaining synchronization; synchronization error; watermark code detection; watermark sequence; Africa; Block codes; Electronic mail; Error correction; Error correction codes; Hidden Markov models; Information theory; Neural networks; Sampling methods; Watermarking;
Conference_Titel :
Information Theory and Its Applications, 2008. ISITA 2008. International Symposium on
Conference_Location :
Auckland
Print_ISBN :
978-1-4244-2068-1
Electronic_ISBN :
978-1-4244-2069-8
DOI :
10.1109/ISITA.2008.4895499
