Distributed Decoding of Convolutional Network Error Correction Codes

摘要

A Viterbi-like decoding algorithm is proposed in this paper for generalizedconvolutional network error correction coding. Different from classical Viterbialgorithm, our decoding algorithm is based on minimum error weight rather thanthe shortest Hamming distance between received and sent sequences. Networkerrors may disperse or neutralize due to network transmission and convolutionalnetwork coding. Therefore, classical decoding algorithm cannot be employed anymore. Source decoding was proposed by multiplying the inverse of networktransmission matrix, where the inverse is hard to compute. Starting from theMaximum A Posteriori (MAP) decoding criterion, we find that it is equivalent tothe minimum error weight under our model. Inspired by Viterbi algorithm, wepropose a Viterbi-like decoding algorithm based on minimum error weight ofcombined error vectors, which can be carried out directly at sink nodes and cancorrect any network errors within the capability of convolutional network errorcorrection codes (CNECC). Under certain situations, the proposed algorithm canrealize the distributed decoding of CNECC.