Title :
A class of Hopfield decodable codes
Author_Institution :
Dept. of Electr. Eng., Linkoping Univ., Sweden
Abstract :
A class of error correcting codes is presented that can be decoded by recurrent neural networks of the Hopfield style. The codes are Euclidean-space codes with information rate less than one bit per dimension. They are related to, but considerably generalize, the Hopfield decodable codes by Yuan et al. Simulations indicate that the Hopfield decoder is nearly optimal for these codes
Keywords :
Hopfield neural nets; error correction codes; Euclidean-space codes; Hopfield decodable codes; Hopfield neural nets; error correcting codes; recurrent neural networks; Binary codes; Decoding; Error correction codes; Euclidean distance; Hamming distance; Hopfield neural networks; Information rates; Neural networks; Parity check codes; Recurrent neural networks;
Conference_Titel :
Neural Networks for Processing [1993] III. Proceedings of the 1993 IEEE-SP Workshop
Conference_Location :
Linthicum Heights, MD
Print_ISBN :
0-7803-0928-6
DOI :
10.1109/NNSP.1993.471880
