There has been an increasing demand for automatic classification of digital signal formats during the past decades, which seems to be a continouning trend in future too. Most of the previously proposed classifiers can only classify a few kinds of digital signals and/or a low order of digital signals. In addition, They usually require a high level of Signal to Noise Ratio (SNR). This paper presents a hybrid intelligent system for recognition of digital signal types, including three main modules: a feature extraction module, a classifier module, i.e., a Probabilistic Neural Networks (PNN), and an optimization module. Simulation results validate the high recognition accuracy of the proposed system even at low SNRs.

Full-Text [PDF 1442 kb]   (659 Downloads)