This project aims to apply the artificial intelligence in Satellite Communications to improve the noise and interference removal from the signal. The received satellite signal will be firstly processed by noise and interference removal from the signal by implementing DFT or ANN-based wavelets processing. The implementation in both cases will be based on FPGAs. After the noise is reduced and the interference is removed, the demodulation process will be performed on the signal.

The received satellite signal will be converted from the analog pattern to the digital pattern as soon as it is received by ADC. The processing on the received signal either by DFT or by ANNs will be applied on the digital pattern of the signal. This digital pattern is only a graphical representation of the received analog signal not the original digital signal which is modulated into the analog signal. The processes of the noise reduction and intereference removal will be done based on the assigned frequency band and frequency values within this band.

The Doppler frequencies will be removed from the received signal by the processing of DFT. After the Doppler frequency is removed, a correction process will be done to recover the original signal. There will be two approaches to achieve this correction process: The first approach is arithmetic approach by applying complex mathematical processing on the received signal. The second approach is by the artificial intelligence. The first approach is faster because the second approach of the artificial intelligence will be based on mathematical calculations so that it will consume the resources of the antenna receiver. However, the advantages of the artificial intelligence are that the models can be substituted based on the principle of trial and error and it is the best alternative when it is not possible to substitute the mathematical models are these mathematical models are too complex to be substituted.