In modern data communication networks , self adaptive equalizers are increasingly required. This is due to the fact that this type of equalizers avoids' retransmission of a learning data sequence each time a receiver enters the network. The problem Of self adaptive equalization has not yet received sufficient study in the literature where cases of moderate data transmission channels are considered. In this paper we present a self adaptive equalizer capable on equalization of arbitrarily severe distortions in the amplitude-frequency response of the chahael . The idea of equalizer relies on the fact that in practical applications, the data signal has a flat spectrum. This enables the identification of the amplitude -frequency response of the channel on the basis of measuring the power spectrum of the received signal . The FFT technique is used to perform that measurement and the corresponding equalization.Computer simulations of the proposed equalizer have shown its ability to deal with drastically distorted channels.