Behavior of neural maximum likelihood decoder in the hard and soft decoding of telecommunication channels

The aim of the paper is to present the main experiments carried out in order to study the behavior of a neural device designed to decode block error correcting codes. The decoder presents several important advantages, like low decoding time, ability to change online the code, simple implementation and last, but not least, possibilities to exceed the correcting performances of the classical decoders. The paper presents the main results obtained through computer simulations both for hard and soft decoding.