Robust Automatic Modulation Classification Using Convolutional Deep Neural Network Based on Scalogram Information
This study proposed a two-stage method, which combines a convolutional neural network (CNN) with the continuous wavelet transform (CWT) for multiclass modulation classification. The modulation signals’ time-frequency information was first extracted using CWT as a data source. The convolutional neura...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Computers |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-431X/11/11/162 |
| _version_ | 1827644713828614144 |
|---|---|
| author | Ahmed Mohammed Abdulkarem Firas Abedi Hayder M. A. Ghanimi Sachin Kumar Waleed Khalid Al-Azzawi Ali Hashim Abbas Ali S. Abosinnee Ihab Mahdi Almaameri Ahmed Alkhayyat |
| author_facet | Ahmed Mohammed Abdulkarem Firas Abedi Hayder M. A. Ghanimi Sachin Kumar Waleed Khalid Al-Azzawi Ali Hashim Abbas Ali S. Abosinnee Ihab Mahdi Almaameri Ahmed Alkhayyat |
| author_sort | Ahmed Mohammed Abdulkarem |
| collection | DOAJ |
| description | This study proposed a two-stage method, which combines a convolutional neural network (CNN) with the continuous wavelet transform (CWT) for multiclass modulation classification. The modulation signals’ time-frequency information was first extracted using CWT as a data source. The convolutional neural network was fed input from 2D pictures. The second step included feeding the proposed algorithm the 2D time-frequency information it had obtained in order to classify the different kinds of modulations. Six different types of modulations, including amplitude-shift keying (<i>ASK</i>), phase-shift keying (<i>PSK</i>), frequency-shift keying (<i>FSK</i>), quadrature amplitude-shift keying (QASK), quadrature phase-shift keying (QPSK), and quadrature frequency-shift keying (QFSK), are automatically recognized using a new digital modulation classification model between 0 and 25 dB SNRs. Modulation types are used in satellite communication, underwater communication, and military communication. In comparison with earlier research, the recommended convolutional neural network learning model performs better in the presence of varying noise levels. |
| first_indexed | 2024-03-09T18:24:20Z |
| format | Article |
| id | doaj.art-4735bf169d644c988bcdca6ce1d69e79 |
| institution | Directory Open Access Journal |
| issn | 2073-431X |
| language | English |
| last_indexed | 2024-03-09T18:24:20Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Computers |
| spelling | doaj.art-4735bf169d644c988bcdca6ce1d69e792023-11-24T08:01:40ZengMDPI AGComputers2073-431X2022-11-01111116210.3390/computers11110162Robust Automatic Modulation Classification Using Convolutional Deep Neural Network Based on Scalogram InformationAhmed Mohammed Abdulkarem0Firas Abedi1Hayder M. A. Ghanimi2Sachin Kumar3Waleed Khalid Al-Azzawi4Ali Hashim Abbas5Ali S. Abosinnee6Ihab Mahdi Almaameri7Ahmed Alkhayyat8Ministry of Migration and Displaced, Baghdad 10011, IraqDepartment of Mathematics, College of Education, Al-Zahraa University for Women, Karbala 56001, IraqBiomedical Engineering Department, College of Engineering, University of Warith Al-Anbiyaa, Karbala 56001, IraqBig Data and Machine Learning Lab, South Ural State University, 454080 Chelyabinsk, RussiaDepartment of Medical Instruments Engineering Techniques, Al-Farahidi University, Baghdad 10011, IraqCollege of Information Technology, Imam Ja’afar Al-Sadiq University, Al-Muthanna 66002, IraqAltoosi University College, Najaf 54001, IraqDepartment of Automation and Applied Informatics, Budapest University of Technology and Economics, 1111 Budapest, HungaryFaculty of Engineering, The Islamic University, Najaf 54001, IraqThis study proposed a two-stage method, which combines a convolutional neural network (CNN) with the continuous wavelet transform (CWT) for multiclass modulation classification. The modulation signals’ time-frequency information was first extracted using CWT as a data source. The convolutional neural network was fed input from 2D pictures. The second step included feeding the proposed algorithm the 2D time-frequency information it had obtained in order to classify the different kinds of modulations. Six different types of modulations, including amplitude-shift keying (<i>ASK</i>), phase-shift keying (<i>PSK</i>), frequency-shift keying (<i>FSK</i>), quadrature amplitude-shift keying (QASK), quadrature phase-shift keying (QPSK), and quadrature frequency-shift keying (QFSK), are automatically recognized using a new digital modulation classification model between 0 and 25 dB SNRs. Modulation types are used in satellite communication, underwater communication, and military communication. In comparison with earlier research, the recommended convolutional neural network learning model performs better in the presence of varying noise levels.https://www.mdpi.com/2073-431X/11/11/162modulationdeep learningwavelet transformmulticlass classification |
| spellingShingle | Ahmed Mohammed Abdulkarem Firas Abedi Hayder M. A. Ghanimi Sachin Kumar Waleed Khalid Al-Azzawi Ali Hashim Abbas Ali S. Abosinnee Ihab Mahdi Almaameri Ahmed Alkhayyat Robust Automatic Modulation Classification Using Convolutional Deep Neural Network Based on Scalogram Information Computers modulation deep learning wavelet transform multiclass classification |
| title | Robust Automatic Modulation Classification Using Convolutional Deep Neural Network Based on Scalogram Information |
| title_full | Robust Automatic Modulation Classification Using Convolutional Deep Neural Network Based on Scalogram Information |
| title_fullStr | Robust Automatic Modulation Classification Using Convolutional Deep Neural Network Based on Scalogram Information |
| title_full_unstemmed | Robust Automatic Modulation Classification Using Convolutional Deep Neural Network Based on Scalogram Information |
| title_short | Robust Automatic Modulation Classification Using Convolutional Deep Neural Network Based on Scalogram Information |
| title_sort | robust automatic modulation classification using convolutional deep neural network based on scalogram information |
| topic | modulation deep learning wavelet transform multiclass classification |
| url | https://www.mdpi.com/2073-431X/11/11/162 |
| work_keys_str_mv | AT ahmedmohammedabdulkarem robustautomaticmodulationclassificationusingconvolutionaldeepneuralnetworkbasedonscalograminformation AT firasabedi robustautomaticmodulationclassificationusingconvolutionaldeepneuralnetworkbasedonscalograminformation AT haydermaghanimi robustautomaticmodulationclassificationusingconvolutionaldeepneuralnetworkbasedonscalograminformation AT sachinkumar robustautomaticmodulationclassificationusingconvolutionaldeepneuralnetworkbasedonscalograminformation AT waleedkhalidalazzawi robustautomaticmodulationclassificationusingconvolutionaldeepneuralnetworkbasedonscalograminformation AT alihashimabbas robustautomaticmodulationclassificationusingconvolutionaldeepneuralnetworkbasedonscalograminformation AT alisabosinnee robustautomaticmodulationclassificationusingconvolutionaldeepneuralnetworkbasedonscalograminformation AT ihabmahdialmaameri robustautomaticmodulationclassificationusingconvolutionaldeepneuralnetworkbasedonscalograminformation AT ahmedalkhayyat robustautomaticmodulationclassificationusingconvolutionaldeepneuralnetworkbasedonscalograminformation |