ISI effects in a hybrid ICA-SVM modulation recognition algorithm

Automatic modulation recognition is a topic of interest in many fields including signal surveillance, multi-user detection and radio frequency spectrum monitoring. In this paper the effect of inter symbol interference (ISI) on the performance of a recently proposed hybrid ICA-SVM modulation recognition algorithm is studied. The algorithm combines elements of cyclo-spectral analysis, ICA and SVM algorithms to distinguish between different types of continuous phase modulations (CPM) and phase shift keying modulations. Although ISI decreases the achievable SVM margin, combining the ICA-SVM hybrid approach with feature equalization and a modified SVM kernel allows reliable classification to occur when ISI is present.