Modulation Recognition of Digital Signal Based on Deep Auto-Ancoder Network

Automated Modulation Classification (AMC) shows great significance for any receiver that has little knowledge of the modulation scheme of the received signal. A useful digital signal modulation recognition scheme inspired by the deep auto-encoder network is proposed in this investigation. In our pro...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2017 IEEE International Conference on Software Quality, Reliability and Security Companion (QRS-C) S. 256 - 260
Hauptverfasser: Tu Ya, Yun Lin, Hui Wang
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.07.2017
Schlagworte:
Algorithm design and analysis
Classification algorithms
cyclic feature
deep autoencoder network
Feature extraction
Modulation
Modulation recognition
Signal processing algorithms
Signal to noise ratio
Wavelet transforms
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Automated Modulation Classification (AMC) shows great significance for any receiver that has little knowledge of the modulation scheme of the received signal. A useful digital signal modulation recognition scheme inspired by the deep auto-encoder network is proposed in this investigation. In our proposed method, there are two deep auto-encoder networks. The system extracts the original features of the signal one by one to complete the recognition of unknown modulation signals, according to different modulation signals has different cyclic spectrum characteristics and wavelet characteristics. Finally, the effectiveness of the system is verified by simulation. The system can identify nine typical signals, which are 2FSK, 4FSK, 8FSK, BPSK, QPSK, 16QAM, 64QAM, 2ASK, MSK. The recognition accuracy can achieve 85% when signal to noise ratio is higher than 0 dB. The results indicate that digital signal modulation recognition based on deep auto-encoder network is feasible and accuracy rate.
DOI:10.1109/QRS-C.2017.50