Augmented Genetic Programming for automatic digital modulation classification

Automatic modulation classification (AMC) is used to identify automatically the modulation types of transmitted signals using the received data samples in the presence of noise. It is a very important process for a receiver that has no, or limited, knowledge of signals received. It is an intermediate step between signal detection and demodulation and has various civilian and military applications. In this paper we propose to use Genetic Programming (GP) with KNN classifier for automatic classification of digital modulation types for the first time. The method proposed here has been designed for BPSK, QPSK, 16QAM and 64QAM. The results from simulation experiments show that the proposed method is able to identify the above modulation types at SNRs of 10dB and 20dB. The performance of the proposed method has been compared with existing methods and it is found to provide the best results so far.