In this work, we are interested in the problem of the coding channel with impulsive noise, whose goal is to correct the errors due to the channel at the time of a numeric transmission. In particular, the convolutional turbocodes constitutes the last progress in this domain and it is very close to the theoretical limit predicted by Shannon in 1948. The objective of this work is to study and to test this coding with the modulation BPSK, to improve the bit error rate (BER) in presence a additive white impulsive GGD noise using the MAP algorithm.-In this work, we are interested in the problem of the coding channel with impulsive noise, whose goal is to correct the errors due to the channel at the time of a numeric transmission. In particular, the convolutional turbocodes constitutes the last progress in this domain and it is very close to the theoretical limit predicted by Shannon in 1948. The objective of this work is to study and to test this coding with the modulation BPSK, to improve the bit error rate (BER) in presence a additive white　impulsive　GGD noise using the MAP algorithm.