Soft-decision decoding of block codes using neural network

It has been shown that soft decision decoding of any (n, k) linear block code can be accomplished using the Viterbi algorithm applied to a trellis with no more than q^n-k states. However, when (n-k) is large, the number of states becomes unmanageable and implementing such a receiver is impractical. A soft-decision decoder based on neural networks is presented. It is shown that the complexity of the proposed neural decoder depends only on the length of the code. Simulation results are presented which show promising error rates over a wide range of signal to noise ratio and the neural decoder bridges the gap in performance between the hard-decision decoding (HDD) and maximum likelihood decoding (MLD)