Abstract

In modern non-cooperative communication fields, signal modulation identification is a crucial issue. This paper presents a new modulation recognition algorithm based on information entropy fusion technology for interclass and intraclass modulation type identification. The decision is made based on extraction of a variety of entropy features and calculation of the feature distances between unknown signal and typical signals with the established information fusion model. Besides SNR, the algorithm needs no further information for the received signal such as signal bandwidth or carrier frequency. Theoretical analysis and simulation results show that the algorithm is practically valuable for its high recognition accuracy, less computation load, good anti-noise performance, and the stability characteristics.

Paper Details