Abstract
A node synchronisation technique for rate (ν − l)/ν convolutional codes is presented which can resolve a loss of node synchronisation with only a single trial. The algorithm, which is implemented external to the decoder, is based on extracting evidence in support of the in-sync or off-sync hypotheses from the syndrome and other sequences associated with the convolutionally encoded received channel symbols. Simulation results are presented for rate R = (ν − l)/ν punctured codes derived from the well known standard NASA, constraint length 7, rate 1/2 code.

This publication has 0 references indexed in Scilit: