In order to automatically classify the modulation types of both digital and analog communication signals, nine feature parameters based on the first statistical moments of the received signals were proposed. All those feature parameters were calculated using the conventional signal processing methods. The calculation process is less complicated and it is suitable for real time online analysis. Next, an automatic modulation recognition algorithm based on the decision-theoretic approach was also developed and its realization was also presented in the form of a flowchart. Computer simulations shows that all the aforementioned types of communication signals have been recognized with an average success rate ≥97% at SNR ≥9dB and the method proposed is suitable for the practical application of signal detection and fast recognition in non-cooperation communication systems.