Automatic determination of digital modulation types with different noises using Convolutional Neural Network based on time-frequency information

被引:55
作者
Daldal, Nihat [1 ]
Comert, Zafer [2 ]
Polat, Kemal [1 ]
机构
[1] Abant Izzet Baysal Univ, Dept Elect & Elect Engn, Fac Engn, TR-14280 Bolu, Turkey
[2] Samsun Univ, Dept Software Engn, Samsun, Turkey
关键词
Modulation type classification; Digital modulation; Deep learning; Convolutional neural network (CNN); Short-time Fourier transform (STFT); CLASSIFICATION; SIGNALS;
D O I
10.1016/j.asoc.2019.105834
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this study, a novel digital modulation classification model has been proposed for automatically recognizing six different modulation types including amplitude shift keying (ASK), frequency shift keying (FSK), phase-shift keying (PSK), quadrate amplitude shift keying (QASK), quadrate frequency shift keying (QFSK), and quadrate phase-shift keying (QPSK). The determination of modulation type is significant in military communication, satellite communication systems, and submarine communication. To classify the modulation types, we have proposed a two-stage hybrid method combining short-time Fourier transform (STFT) and convolutional neural network (CNN). In the first stage, as the data source, the time-frequency information from these modulation signals have been extracted with STFT. This information has been obtained as 2D images to feed the input of the CNN deep learning method. In the second stage, the obtained 2D time-frequency information has been given to the input of the CNN algorithm to classify the modulation types. In this work, noises at various SNR values from 0 dB to 25 dB were created and added to the modulated signals. Even in the presence of noise, the proposed hybrid deep learning model achieved excellent results in the noised-modulation signals. (C) 2019 Elsevier B.V. All rights reserved.
引用
收藏
页数:10
相关论文
共 40 条
[1]   Automatic modulation classification based on high order cumulants and hierarchical polynomial classifiers [J].
Abdelmutalab, Ameen ;
Assaleh, Khaled ;
El-Tarhuni, Mohamed .
PHYSICAL COMMUNICATION, 2016, 21 :10-18
[2]  
Akyon F.C., 2018, 2018 26th Signal Processing and Communications Applications Conference (SIU), P1
[3]   Multi-category EEG signal classification developing time-frequency texture features based Fisher Vector encoding method [J].
Alcin, Omer F. ;
Siuly, Siuly ;
Bajaj, Varun ;
Guo, Yanhui ;
Sengur, Abdulkadir ;
Zhang, Yanchun .
NEUROCOMPUTING, 2016, 218 :251-258
[4]  
Ali A, 2017, 2017 COMPUTING CONFERENCE, P294, DOI 10.1109/SAI.2017.8252117
[5]   Unsupervised feature learning and automatic modulation classification using deep learning model [J].
Ali, Afan ;
Fan Yangyu .
PHYSICAL COMMUNICATION, 2017, 25 :75-84
[6]   Automatic modulation classification of digital modulation signals with stacked autoencoders [J].
Ali, Afan ;
Fan Yangyu ;
Liu, Shu .
DIGITAL SIGNAL PROCESSING, 2017, 71 :108-116
[7]  
[Anonymous], BIYOMEDIKAL GORUNTUL
[8]  
[Anonymous], SAYISAL HABERLESME
[9]   Automatic identification of digital modulation types [J].
Azzouz, EE ;
Nandi, AK .
SIGNAL PROCESSING, 1995, 47 (01) :55-69
[10]  
Bouchou M, 2017, 2017 17TH IEEE INTERNATIONAL CONFERENCE ON COMMUNICATION TECHNOLOGY (ICCT 2017), P28, DOI 10.1109/ICCT.2017.8359478