An enhanced SOFM method for automatic recognition and identification of digital modulations

This paper enhances the author´s previous work with adding memory and other methods to SOFM recognition system to speed up the process and improve the accuracy of the automatically recognition. The focus of this research is the various M-ary QAM and PSK modulated signal formats subjected to data contamination by noise and fading. Distinguish from other studies, the complexity of nonlinear channel situations and certain critical use cases have been considered in the estimation of the unknown constellation. Successful recognition results have been achieved with low SNR for BPSK, QPSK, PSK-16 and QAM-16, and modest SNR from 10 to 20 dB for higher-level QAM schemes up to QAM-256.