Automatic modulation recognition based on mixed-type features
Xinrui Jiang, Hui Chen, Yaodong Zhao, Wen-Qin Wang
Abstract
The existing modulation classification method using instantaneous features is poor for low SNRs, and the high-order cumulant features-based modulation recognition algorithm is only applicable to some types of communication modulation signals. To overcome these problems, we propose a mixed features-based modulation recognition algorithm, which refines instantaneous features and high-order cumulant feature, and the back propagation (BP) neural network is adopted as a classifier to perform experiments. The experimental results show that our proposed mixed features-based modulation recognition method can improve the recognition rate for more kinds of signals.
Topics & Concepts
Modulation (music)Pattern recognition (psychology)Artificial intelligenceComputer scienceClassifier (UML)Artificial neural networkFeature extractionFeature (linguistics)Speech recognitionAcousticsPhysicsLinguisticsPhilosophyWireless Signal Modulation Classification