Performance of FeatureBased Techniques for Automatic Digital Modulation Recognition and Classification—A Review
Abstract
:1. Introduction
2. AMR Signals
2.1. Digital Modulation
2.2. Communication Channel Model
3. Overview of FB Approach for MR
3.1. Spectral Features for MR
3.2. Statistical Features for MR
3.3. Transform Features for MR
3.4. Constellation Shape Features for MR
3.5. Critical Analysis of the FB Approach for MR
4. Overview of the Type of Classifiers Used for MR
4.1. DT
4.2. ANN
4.3. SVM
4.4. KNN
4.5. Clustering Algorithms
4.6. Complexity Analysis for Classifiers
4.7. Critical Analysis on Different Classifiers for MR
5. Performance Analysis
 Selection of appropriate features can improve system robustness towards noise effects and can be sensitive in terms of discriminating the modulation schemes, thereby enhancing the performance.
 Features and classifier types with reduced complexity are crucial for enhancing the performance of classifiers.
 Recognition is difficult when higherorder modulation types are used for higherorder QAM at low SNR.
6. Conclusions
Author Contributions
Conflicts of Interest
References
 Yang, G.Q. Modulation classification based on extensible neural networks. Math. Probl. Eng. 2017, 2017, 1–10. [Google Scholar]
 Popoola, J.J.; Olst, R.V. Development and comparative study of effects of training algorithms on performance of artificial neural network based analog and digital automatic modulation recognition. J. Eng. Sci. Technol. Rev. 2015, 8, 135–144. [Google Scholar] [CrossRef]
 Tu, Y.; Lin, Y.; Wang, J.; Kim, J.U. Semisupervised learning with generative adversarial networks on digital signal modulation classification. Comput. Mater. Continua 2018, 55, 243–254. [Google Scholar]
 Tang, B.; Tu, Y.; Zhang, Z.; Lin, Y. Digital signal modulation classification with data augmentation using generative adversarial nets in cognitive radio networks. IEEE Access 2018, 6, 15713–15722. [Google Scholar] [CrossRef]
 Azzouz, E.E.; Nandi, A.K. Procedure for automatic recognition of analogue and digital modulations. IEEE Proc. Commun. 1996, 143, 259–266. [Google Scholar] [CrossRef]
 Ali, A.; Yangyu, F.; Liu, S. Automatic modulation classification of digital modulation signals with stacked auto encoders. Digit. Signal Process. 2017, 71, 108–116. [Google Scholar] [CrossRef]
 Lee, J.H.; Kim, J.; Kim, B.; Yoon, D.; Choi, J.W. Robust automatic modulation classification technique for fading channels via deep neural network. Entropy 2017, 19, 454. [Google Scholar]
 Uys, L.Y.; Gouws, M.; Strydom, J.J.; Helberg, A.S.J. The Performance of Featurebased Classification of Digital Modulations under Varying SNR and Fading Channel Conditions. In Proceedings of the 2017 IEEE AFRICON, Cape Town, South Africa, 18–20 September 2017; pp. 198–203. [Google Scholar]
 Zhang, X.; Ge, T. Automatic Modulation Recognition of Communication Signals Based on Instantaneous Statistical Characteristics and SVM Classifier. In Proceedings of the IEEE AsiaPacific Conference on Antennas and Propagation (APCAP), Auckland, New Zealand, 5–8 August 2018; IEEE: Piscataway, NJ, USA, 2018; pp. 344–346. [Google Scholar]
 Dobre, O.A. Signal identification for emerging intelligent radios: Classical problems and new challenges. IEEE Instrum. Meas. Mag. 2015, 18, 11–18. [Google Scholar] [CrossRef]
 Ahn, W.H.; Choi, J.; Park, C.; Seo, B.; Lee, M. Automatic Modulation Classification of Digital Modulation Signals based on Gaussian Mixture Model. In Proceedings of the UBICOMM 2014: The Eighth International Conference on Mobile Ubiquitous Computing, Systems, Services and Technologies, Rome, Italy, 24–28 August 2014. [Google Scholar]
 Ghauri, S.A. KNN based classification of digital modulated signals. IIUM Eng. J. 2016, 17, 71–82. [Google Scholar] [CrossRef]
 Meng, F.; Chen, P.; Wu, L.; Wang, X. Automatic modulation classification: A deep learning enabled approach. IEEE Trans. Veh. Technol. 2018, 67, 10760–10772. [Google Scholar] [CrossRef]
 Zhu, Z.; Aslam, M.W.; Nandi, A.K. Genetic algorithm optimized distribution sampling test for MQAM modulation classification. Signal Process. 2014, 94, 264–277. [Google Scholar] [CrossRef]
 Wang, F.; Xu, R.; Zhong, Z. Low complexity KolmogorovSmirnov modulation classification. In Proceedings of the 2011 IEEE Wireless Communications and Networking Conference, Cancun, Mexico, 28–31 March 2011; pp. 1607–1611. [Google Scholar]
 Wang, F.; Wang, X. Fast and robust modulation classification via KolmogorovSmirnov test. IEEE Trans. Commun. 2010, 58, 2324–2332. [Google Scholar] [CrossRef]
 Lathi, B.P. Modern Digital and Analog Communication Systems; Oxford University Press: New York, NY, USA, 1998. [Google Scholar]
 Sadeque, G. Bit Error Rate (BER) Comparison of AWGN Channels for Different Type’s Digital Modulation Using MATLAB Simulink. Am. Sci. Res. J. Eng. Technol. Sci. 2015, 13, 61–71. [Google Scholar]
 Peng, S.; Jiang, H.; Wang, H.; Alwageed, H.; Yao, Y.D. Modulation classification using convolutional neural network based deep learning model. In Proceedings of the 2017 26th Wireless and Optical Communication Conference (WOCC), Newark, NJ, USA, 7–8 April 2017; pp. 1–5. [Google Scholar]
 Agilent Technologies. Digital Modulation in Communications Systems: An introduction. 2001. Available online: http://www.iikt.ovgu.de/iesk_media/Downloads/HF/Digital+Modulations+_+Communication+Systems+by+Agilentp1252.pdf (accessed on 21 November 2019).
 Hosany, M.A. The Matlab Communication Toolbox; University of Mauritius: Moka, Mauritius, May 2005. [Google Scholar]
 Popescu, S.O.; Budura, G.; Gontean, A.S. Review of PSK and QAM: Digital Modulation Techniques on FPGA. In Proceedings of the 2010 International Joint Conference on Computational Cybernetics and Technical Informatics, Timisoara, Romania, 27–29 May 2010. [Google Scholar]
 White, C. Data Communications and Computer Networks: A Business User’s Approach, 8th ed.; Cengage Learning: Boston, UK, 2015. [Google Scholar]
 Basar, E. Index modulation techniques for 5G wireless networks. IEEE Commun. Mag. 2016, 54, 168–175. [Google Scholar] [CrossRef]
 Tanawade, N.; Sudhansu, S. FPGA implementation of digital modulation techniques BPSK and QPSK using HDL Verilog. Int. J. Comput. Appl. 2017, 165, 44–50. [Google Scholar] [CrossRef]
 Kassam, S.A. (Modulation and Demodulation) Modems for Data Transmission over Analog Links. In TCOM 370 Principles of Data Communication; notes 99–5; University of Pennsylvania: Philadelphia, PA, USA, 1999. [Google Scholar]
 Patial, A.; Narayana, B.D.R.; Kumar, B.N.; Kurup, D.G.; Jha, B.B. Simulation of coastal line surveillance system and comparative study of digital modulation techniques using GNU radio. Int. J. Adv. Res. Comput. Commun. Eng. 2014, 3, 6714–6718. [Google Scholar]
 Zhang, W. Automatic Modulation Classification Based on Statistical Features and Support Vector Machine. In Proceedings of the General Assembly and Scientific Symposium (URSI GASS), 31th URSI, IEEE, Beijing, China, 16–23 August 2014; pp. 1–4. [Google Scholar]
 Valipour, M.H.; Homayounpour, M.M.; Mehralian, M.A. Automatic Digital Modulation Recognition in Presence of Noise using SVM and PSO. In Proceedings of the Telecommunications (IST), Sixth International Symposium, IEEE, Tehran, Iran, 6–8 November 2012; pp. 378–382. [Google Scholar]
 Hassanpour, S.; Pezeshk, A.M.; Behnia, F. Automatic Digital Modulation Recognition Based on Novel Features and Support Vector Machine. In Proceedings of the SignalImage Technology & InternetBased Systems (SITIS), 12th International Conference, IEEE, Naples, Italy, 28 November–1 December 2016; pp. 172–177. [Google Scholar]
 Bagga, J.; Tripathi, N. Automatic modulation classification using statistical features in fading environment. Int. J. Adv. Elec. Electron. Ins. Eng. 2013, 2, 3701–3709. [Google Scholar]
 Azzouz, E.E.; Nandi, A.K. Automatic identification of digital modulation types. Signal Process. 1995, 47, 55–69. [Google Scholar] [CrossRef]
 Zhu, Z.; Nandi, A.K. Automatic Modulation Classification: Principles, Algorithms and Applications; John Wiley & Sons: London, UK, 2014. [Google Scholar]
 Wu, Z.; Zhou, S.; Yin, Z.; Ma, B.; Yang, Z. Robust automatic modulation classification under varying noise conditions. IEEE Access 2017, 5, 19733–19741. [Google Scholar] [CrossRef]
 Moser, E.; Moran, M.K.; Hillen, E.; Li, D.; Wu, Z. Automatic Modulation Classification via Instantaneous Features. In Proceedings of the Aerospace and Electronics Conference (NAECON), IEEE, Dayton, OH, USA, 15–19 June 2015; pp. 218–223. [Google Scholar]
 Liu, Y.; Liang, G.; Xu, X.; Li, X. The Methods of Recognition for Common Used MAry Digital Modulations. In Proceedings of the Wireless Communications, Networking and Mobile Computing, WiCOM’08, 4th International Conference, IEEE, Dalian, China, 12–14 October 2008; pp. 1–4. [Google Scholar]
 Hazza, A.; Shoaib, M.; Saleh, A.; Fahd, A. Classification of Digitally Modulated Signals in Presence of NonGaussian HF Noise. In Proceedings of the Wireless Communication Systems (ISWCS), 7th International Symposium, IEEE, York, UK, 19–22 September 2010; pp. 815–819. [Google Scholar]
 Wang, H.K.; Zhang, B.; Wu, J.P.; Han, Y.Z.; Wu, X.W.; Jia, R.S. A Research on Automatic Modulation Recognition with the Combination of the Rough Sets and Neural Network. In Proceedings of the Pervasive Computing Signal Processing and Applications (PCSPA), 2010 First International Conference, IEEE, Harbin, China, 17–19 September 2010; pp. 807–810. [Google Scholar]
 Benedetto, F.; Tedeschi, A.; Giunta, G. Automatic Blind Modulation Recognition of Analog and Digital Signals in Cognitive Radios. In Proceedings of the Vehicular Technology Conference (VTCFall), IEEE 84th, Montreal, QC, Canada, 18–21 September 2016; pp. 1–5. [Google Scholar]
 Jagannath, J.; O’Connor, D.; Polosky, N.; Sheaffer, B.; Foulke, S.; Theagarajan, L.N.; Varshney, P.K. Design and Evaluation of Hierarchical Hybrid Automatic Modulation Classifier using Software Defined Radios. In Proceedings of the Computing and Communication Workshop and Conference (CCWC), 2017 IEEE 7th Annual. IEEE, Las Vegas, NV, USA, 9–11 January 2017; pp. 1–7. [Google Scholar]
 ElKhoribi, R.A.; Shoman, M.A.I.; Mohammed, A.A. Automatic digital modulation recognition using artificial neural network in cognitive radio. Int. J. Emerg. Trends Technol. Comput. Sci. 2014, 3, 132–136. [Google Scholar]
 Liu, D.; Liu, J. A Novel Signal Recognition Algorithm based on SVM in Cognitive Networks. In Proceedings of the Communication Technology (ICCT), 2010 12th IEEE International Conference, IEEE, Nanjing, China, 11–14 November 2010; pp. 1264–1267. [Google Scholar]
 Popoola, J.J. Automatic recognition of both inter and intra classes of digital modulated signals using artificial neural network. J. Eng. Sci. Technol. 2014, 9, 273–285. [Google Scholar]
 Huang, F.Q.; Zhong, Z.M.; Xu, Y.T.; Ren, G.C. Modulation Recognition of Symbol Shaped Digital Signals. In Proceedings of the Communications, Circuits and Systems, ICCCAS International Conference, IEEE, Fujian, China, 25–27 May 2008; pp. 328–332. [Google Scholar]
 Hazza, A.; Shoaib, M.; Alshebeili, S.A.; Fahad, A. An Overview of FeatureBased Methods for Digital Modulation Classification. In Proceedings of the 1st International Conference on Communications, Signal Processing, and their Applications (ICCSPA), IEEE, Sharjah, United Arab Emirates, 12–14 February 2013; pp. 1–6. [Google Scholar]
 Soliman, S.S.; Hsue, S.Z. Signal classification using statistical moments. IEEE Trans. Commun. 1992, 40, 908–916. [Google Scholar] [CrossRef]
 Swami, A.; Sadler, B.M. Hierarchical digital modulation classification using cumulants. IEEE Trans. Commun. 2000, 48, 416–429. [Google Scholar] [CrossRef]
 Liu, M.Z.; Zhao, Y.; Shi, L.; Dong, J.W. Research on Recognition Algorithm of Digital Modulation by Higher Order Cumulants. In Proceedings of the Fourth International Conference on Instrumentation and Measurement, Computer, Communication and Control (IMCCC), IEEE, Harbin, China, 18–20 September 2014; pp. 686–690. [Google Scholar]
 Zhao, Y.; Xu, Y.T.; Jiang, H.; Luo, Y.J.; Wang, Z.W. Recognition of Digital Modulation Signals based on Highorder Cumulants. In Proceedings of the International Conference on Wireless Communications & Signal Processing (WCSP), IEEE, Nanjing, China, 15–17 October 2015; pp. 1–5. [Google Scholar]
 Dobre, O.A.; Abdi, A.; BarNess, Y.; Su, W. Survey of automatic modulation classification techniques: Classical approaches and new trends. IEEE Trans. Commun. 2007, 1, 137–156. [Google Scholar] [CrossRef]
 Abdelmutalab, A.; Assaleh, K.; ElTarhuni, M. Automatic modulation classification based on high order cumulants and hierarchical polynomial classifiers. Phys. Commun. 2016, 21, 10–18. [Google Scholar] [CrossRef]
 Shen, L.; Li, S.; Song, C.; Chen, F. Automatic Modulation Classification of MPSK Signals using High Order Cumulants. In Proceedings of the 8th International Conference in Signal Processing, IEEE, Beijing, China, 16–20 November 2006. [Google Scholar]
 Das, D.; Bora, P.K.; Bhattacharjee, R. Cumulant based Automatic Modulation Classification of QPSK, OQPSK, 8PSK and 16PSK. In Proceedings of the 8th International Conference in Communication Systems and Networks (COMSNETS), IEEE, Bangalore, India, 5–10 January 2016; pp. 1–5. [Google Scholar]
 Hussain, A.; Sohail, M.F.; Alam, S.; Ghauri, S.A.; Qureshi, I.M. Classification of MQAM and MPSK signals using genetic programming (GP). Neural. Comput. Appl. 2019, 31, 6141–6149. [Google Scholar] [CrossRef]
 Ghauri, S.A.; Qureshi, I.M.; Malik, A.N.; Cheema, T.A. Higher order cummulants based digital modulation recognition scheme. Res. J. Appl. Sci. Eng. 2013, 6, 3910–3915. [Google Scholar] [CrossRef]
 Chang, D.C.; Shih, P.K. Cumulantsbased modulation classification technique in multipath fading channels. IEEE Trans. Commun. 2015, 9, 828–835. [Google Scholar] [CrossRef]
 Ozen, A.; Ozturk, C. A Novel Modulation Recognition Technique Based On Artificial Bee Colony Algorithm In The Presence Of Multipath Fading Channels. In Proceedings of the 36th International Conference in Telecommunications and Signal Processing (TSP), IEEE, Rome, Italy, 2–4 July 2013; pp. 239–243. [Google Scholar]
 Mirarab, M.R.; Sobhani, M.A. Robust Modulation Classification for PSK/QAM/ASK using Higherorder Cumulants. In Proceedings of the 6th International Conference on Information, Communications & Signal Processing, IEEE, Singapore, 10–13 December 2007; pp. 1–4. [Google Scholar]
 Ali, A.; Yangyu, F. Higherorder Statistics based Modulation Classification using Hierarchical Approach. In Proceedings of the Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC), IEEE, Xi’an, China, 3–5 October 2016; pp. 370–374. [Google Scholar]
 Chaithanya, V.; Reddy, V.U. Blind Modulation Classification in the Presence of Carrier Frequency Offset. In Proceedings of the International Conference in Signal Processing and Communications (SPCOM), IEEE, Bangalore, India, 18–21 July 2010; pp. 1–5. [Google Scholar]
 Geisinger, N.P. Classification of digital modulation schemes using linear and nonlinear classifiers. Master’s Thesis, Naval Postgraduate School, Monterey, CA, USA, 2010. [Google Scholar]
 Yu, Z.; Shi, Y.Q.; Su, W. Mary frequency shift keying signal classification basedon discrete Fourier transform. In Proceedings of the Military Communications Conference, 2003, MILCOM’03, IEEE, Boston, MA, USA, 13–16 October 2003; Volume 2, pp. 1167–1172. [Google Scholar]
 Kumar, P.H.L.; Shrinivasan, L. Automatic Digital Modulation Recognition using Minimum Feature Extraction. In Proceedings of the 2nd International Conference in Computing for Sustainable Global Development (INDIACom), IEEE, New Delhi, India, 11–13 March 2015; pp. 772–775. [Google Scholar]
 Grajal, J.; YesteOjeda, O.; Sanchez, M.A.; Garrido, M.; LópezVallejo, M. Real Time FPGA Implementation of an Automatic Modulation Classifier for Electronic Warfare Applications. In Proceedings of the 19th European in Signal Processing Conference, IEEE, Barcelona, Spain, 29 August–2 September 2011; pp. 1514–1518. [Google Scholar]
 Hassanpour, S.; Pezeshk, A.M.; Behnia, F. A Robust Algorithm based on Wavelet Transform for Recognition of Binary Digital Modulations. In Proceedings of the 38th International Conference in Telecommunications and Signal Processing (TSP), IEEE, Prague, Czech Republic, 9–11 July 2015; pp. 508–512. [Google Scholar]
 Maliatsos, K.; Vassaki, S.; Constantinou, P. Interclass and Intraclass Modulation Recognition using the Wavelet Transform. In Proceedings of the IEEE 18th International Symposium in Personal, Indoor and Mobile Radio Communications PIMRC, Athens, Greece, 3–7 September 2007; pp. 1–5. [Google Scholar]
 Ho, K.C.; Prokopiw, W.; Chan, Y.T. Modulation Identification by the Wavelet Transform. In Proceedings of the Military Communications Conference, MILCOM’95, IEEE, San Diego, CA, USA, 5–8 November 1995; Volume 2, pp. 886–890. [Google Scholar]
 Prakasam, P.; Madheswaran, M. Automatic Modulation Identification of QPSK and GMSK using Wavelet Transform for Adaptive Demodulator in SDR. In Proceedings of the International Conference in Signal Processing, Communications and Networking, ICSCN’07, IEEE, Chennai, India, 22–24 February 2007; pp. 507–511. [Google Scholar]
 Kumar, S.; Bohara, V.A.; Darak, S.J. Automatic Modulation Classification by Exploiting Cyclostationary Features in Wavelet Domain. In Proceedings of the Twentythird National Conference in Communications (NCC), Chennai, India, 2–4 March 2017; pp. 1–6. [Google Scholar]
 Hong, L. Lowcomplexity Identifier for Mary QAM Signals. In Proceedings of the Southeastcon, SOUTHEASTCON’09, IEEE, Atlanta, GA, USA, 5–8 March 2009; pp. 164–168. [Google Scholar]
 Prakasam, P.; Madheswaran, M. Digital modulation identification model using wavelet transform and statistical parameters. J. Comput. Syst. Netw. Commun. 2008, 2008, 8. [Google Scholar] [CrossRef]
 Walenczykowska, M.; Kawalec, A. Type of modulation identification using wavelet transform and neural network. Bull. Pol. Acad. Sci. Tech. 2016, 64, 257–261. [Google Scholar] [CrossRef]
 Prakasam, P.; Madheswaran, M. Intelligent decision making system for digital modulation scheme classification in software radio using wavelet transform and higher order statistical moments. Kluw. Commun. 2009, 50, 509–528. [Google Scholar] [CrossRef]
 Wang, L.; Guo, S.; Jia, C. Recognition of Digital Modulation Signals based on Wavelet Amplitude Difference. In Proceedings of the 7th IEEE International Conference in Software Engineering and Service Science (ICSESS), IEEE, Beijing, China, 26–28 August 2016; pp. 627–630. [Google Scholar]
 Wilfred, A.; Okonkwo, O.R. A review of cyclostationary feature detection based spectrum sensing technique in cognitive radio networks. E3 J. Sci. Res. 2016, 4, 41–47. [Google Scholar] [CrossRef]
 Zhu, X.; Lin, Y.; Dou, Z. Automatic Recognition of Communication Signal Modulation based on Neural Network. In Proceedings of the IEEE International Conference in Electronic Information and Communication Technology (ICEICT), IEEE, Harbin, China, 20–22 August 2016; pp. 223–226. [Google Scholar]
 Dobre, O.A.; BarNess, Y.; Su, W. Higherorder Cyclic Cumulants for High Order Modulation Classification. In Proceedings of the Military Communications Conference, MILCOM’03, IEEE, Boston, MA, USA, 13–16 October 2003; Volume 1, pp. 112–117. [Google Scholar]
 Wang, H.; Dobre, O.A.; Li, C.; Inkol, R. MFSK Signal Recognition in Fading Channels for Cognitive Radio. In Proceedings of the Radio and Wireless Symposium (RWS), IEEE, Santa Clara, CA, USA, 15–18 January 2012; pp. 375–378. [Google Scholar]
 Wang, H.; Dobre, O.A.; Li, C.; Inkol, R. Experimental Results for MFSK Signal Classification and Parameter Estimation. In Proceedings of the Instrumentation and Measurement Technology Conference (I2MTC), IEEE, Minneapolis, MN, USA, 6–9 May 2013; pp. 1786–1789. [Google Scholar]
 Wang, H.; Dobre, O.A.; Li, C.; Popescu, D.C. Blind cyclostationaritybased symbol period estimation for FSK signals. IEEE Commun. Lett. 2015, 19, 1149–1152. [Google Scholar] [CrossRef]
 Dobre, O.A.; Oner, M.; Rajan, S.; Inkol, R. Cyclostationaritybased robust algorithms for QAM signal identification. IEEE Commun. Lett. 2012, 16, 12–15. [Google Scholar] [CrossRef]
 Karami, E.; Dobre, O.A. Identification of SMOFDM and ALOFDM signals based on their secondorder cyclostationarity. IEEE Trans. Veh. Technol. 2015, 64, 942–953. [Google Scholar] [CrossRef]
 Mobasseri, B.G. Digital modulation classification using constellation shape. Signal Process 2000, 80, 251–277. [Google Scholar] [CrossRef]
 Wang, L.; Li, Y. Constellation based Signal Modulation Recognition for MQAM. In Proceedings of the IEEE 9th International Conference in Communication Software and Networks (ICCSN), Guangzhou, China, 6–8 May 2017; pp. 826–829. [Google Scholar]
 Zhendong, C.; Weining, J.; Changbo, X.; Min, L. Modulation Recognition based on Constellation Diagram for MQAM Signals. In Proceedings of the 11th International Conference in Electronic Measurement & Instruments (ICEMI), Harbin, China, 16–19 August 2013; Volume 1, pp. 70–74. [Google Scholar]
 Cai, T.; Wang, C.; Cui, G.; Wang, W. Constellationwavelet Transform Automatic Modulation Identifier for Mary QAM Signals. In Proceedings of the 26th Annual International Symposium in Personal, Indoor, and Mobile Radio Communications (PIMRC), IEEE, Hong Kong, China, 30 August–2 September 2015; pp. 212–216. [Google Scholar]
 Cutno, P.; Cheng, C.H. A Softwaredefined Radio based Automatic Modulation Classifier. In Proceedings of the Wireless Telecommunications Symposium (WTS), IEEE, Chicago, IL, USA, 26–28 April 2017; pp. 1–6. [Google Scholar]
 Han, Y.; Wei, G.; Song, C.; Lai, L. Hierarchical Digital Modulation Recognition based on HigherOrder Cumulants. In Proceedings of the Second International Conference in Instrumentation, Measurement, Computer, Communication and Control (IMCCC), IEEE, Harbin, China, 8–10 December 2012; pp. 1645–1648. [Google Scholar]
 Lv, Y.; Liu, Y.; Liu, F.; Gao, J.; Liu, K.; Xie, G. Automatic Modulation Recognition of Digital Signals Using CWT Based on Optimal Scales. In Proceedings of the International Conference in Computer and Information Technology (CIT), IEEE, Xi’an, China, 11–13 September 2014; pp. 430–434. [Google Scholar]
 Satija, U.; Manikandan, M.S.; Ramkumar, B. Performance Study of Cyclostationary Based Digital Modulation Classification Schemes. In Proceedings of the 9th International Conference in Industrial and Information Systems (ICIIS), IEEE, Gwalior, India, 15–17 December 2014; pp. 1–5. [Google Scholar]
 Athira, S.; Mohan, R.; Poornachandran, P.; Soman, K.P. Automatic Modulation Classification using Convolutional Neural Network. Int. J. Comput. Technol. Appl. 2016, 9, 7733–7742. [Google Scholar]
 Zhang, G.P. Neural networks for classification: A survey. IEEE Trans. Syst. 2000, 30, 451–462. [Google Scholar] [CrossRef]
 Chaozhu, Z.; Lianbai, Y.; Xin, W. Discrete Wavelet Neural Network Group System for Digital Modulation Recognition. In Proceedings of the IEEE 3rd International Conference in Communication Software and Networks (ICCSN), IEEE, Xi’an, China, 27–29 May 2011; pp. 603–606. [Google Scholar]
 Wang, H.; Guo, L. A New Method of Automatic Modulation Recognition Based on Dimension Reduction. In Proceedings of the Cooperative Positioning and Service (CPGP), IEEE, Harbin, China, 19–21 May 2017; pp. 316–320. [Google Scholar]
 Kim, B.; Kim, J.; Chae, H.; Yoon, D.; Choi, J.W. Deep Neural Networkbased Automatic Modulation Classification Technique. In Proceedings of the International Conference on Information and Communication Technology Convergence (ICTC), IEEE, Jeju, SouthKorea, 19–21 October 2016; pp. 579–582. [Google Scholar]
 Khan, F.N.; Zhong, K.; AlArashi, W.H.; Yu, C.; Lu, C.; Lau, A.P.T. Modulation format identification in coherent receivers using deep machine learning. IEEE Photonics Technol. Lett. 2016, 28, 1886–1889. [Google Scholar] [CrossRef]
 Wang, D.; Zhang, M.; Li, J.; Li, Z.; Li, J.; Song, C.; Chen, X. Intelligent constellation diagram analyzer using convolutional neural networkbased deep learning. Opt. Express 2017, 25, 17150–17166. [Google Scholar] [CrossRef]
 Wang, L.X.; Ren, Y.J.; Zhang, R.H. Algorithm of Digital Modulation Recognition based on Support Vector Machines. In Proceedings of the International Conference in Machine Learning and Cybernetics, IEEE, Baoding, China, 12–15 July 2009; Volume 2, pp. 980–983. [Google Scholar]
 Zhu, Z.; Aslam, M.W.; Nandi, A.K. Support Vector Machine Assisted Genetic Programming for MQAM Classification. In Proceedings of the 10th International Symposium in Signals, Circuits and Systems (ISSCS), IEEE, lasi, Romania, 30 June–1 July 2011; pp. 1–6. [Google Scholar]
 Thrane, J.; Wass, J.; Piels, M.; Diniz, J.C.; Jones, R.; Zibar, D. Machine learning techniques for optical performance monitoring from directly detected PDMQAM signals. J. Lightw. Technol. 2016, 35, 868–875. [Google Scholar] [CrossRef] [Green Version]
 Lin, X.; Dobre, O.A.; Ngatched, T.M.; Eldemerdash, Y.A.; Li, C. Joint modulation classification and OSNR estimation enabled by support vector machine. IEEE Photonics Technol. Lett. 2018, 30, 2127–2130. [Google Scholar] [CrossRef]
 Ahn, W.H.; Nah, S.P.; Seo, B.S. Automatic classification of digitally modulated signals based on knearest neighbour. Lect. Notes Electr. Eng. 2015, 329, 63–69. [Google Scholar] [CrossRef]
 Aslam, M.W.; Zhu, Z.; Nandi, A.K. Automatic Modulation Classification using Combination of Genetic Programming and KNN. IEEE Trans. Wirel. Commun. 2012, 11, 2742–2750. [Google Scholar]
 Guldemır, H.; Sengur, A. Comparison of clustering algorithms for analog modulation classification. Expert Syst. Appl. 2006, 30, 642–649. [Google Scholar] [CrossRef]
 Zhinan, X.; Wenle, B. A Novel Blind Recognition Algorithm for modulated MQAM Signals. In Proceedings of the WRI International Conference in Communications and Mobile Computing, CMC’09, IEEE, Kunming, China, 6–8 January 2009; Volume 1, pp. 461–465. [Google Scholar]
 Yin, C.; Li, B.; Li, Y. Modulation Classification of MQAM Signals from Their Constellation using Clustering. In Proceedings of the Second International Conference on Communication Software and Networks, IEEE, Singapore, 26–28 February 2010; pp. 303–306. [Google Scholar]
 Ahmadi, N. Using fuzzy clustering and TTSAS algorithm for modulation classification based on constellation diagram. Eng. Appl. Artif. Intell. 2010, 23, 357–370. [Google Scholar] [CrossRef]
 Satija, U.; Mohanty, M.; Ramkumar, B. Automatic modulation classification using Stransform based features. In Proceedings of the 2015 2nd International Conference on Signal Processing and Integrated Networks (SPIN) 2015, IEEE, Noida, India, 19–20 February 2015; pp. 708–712. [Google Scholar]
Features  Mathematical Equation 

Maximum value of PSD${\mathit{\gamma}}_{\mathit{max}}$of the normalized centered instantaneous amplitude  ${\gamma}_{\mathit{max}}=\frac{\mathit{max}DFT({A}_{cn}\left(i\right){}^{2}}{{N}_{s}}$, where DFT is the discrete Fourier transform of the modulated signal, ${N}_{s}$ is the sample number, ${A}_{cn}=\raisebox{1ex}{${A}_{i}$}\!\left/ \!\raisebox{1ex}{${\mu}_{A}$}\right.1$, ${A}_{i}$ is the ${i}^{th}$ instantaneous amplitude and ${\mu}_{A}$ is the sample mean 
Standard deviation of the absolute values of the centered nonlinear components of instantaneous phase${\mathit{\sigma}}_{\mathit{a}\mathit{p}}$  ${\sigma}_{ap}=\sqrt{\frac{1}{{N}_{c}}\left({\displaystyle {\displaystyle \sum}_{{A}_{n}\left(i\right)>{A}_{t}}}{\phi}_{NL}^{2}\left(i\right)\right)\frac{1}{{N}_{c}}{\left({\displaystyle {\displaystyle \sum}_{{A}_{n}\left(i\right)>{A}_{t}}}{\phi}_{NL}\left(i\right)\right)}^{2}}$, where ${N}_{c}$ is the number of sample(s) in $\left\{{\varphi}_{NL}\right\}$ for ${A}_{n}\left(i\right){A}_{t}$, where ${A}_{t}$ is the threshold value of ${A}_{n}\left(i\right)$ when the filter provides the minimum amplitude of the signal sample due to high noise sensitivity and ${\varphi}_{NL}\left(i\right)$ is the nonlinear component of the i^{th} instantaneous phase of the sample. 
Standard deviation of the absolute value of the normalized centered instantaneous amplitude in the nonweak segment of signal${\mathit{\sigma}}_{\mathit{a}}$  ${\sigma}_{a}=\sqrt{\frac{1}{L}\left({\displaystyle {\displaystyle \sum}_{{A}_{n}\left(i\right)>{t}_{th}}}{a}_{cn}^{2}\left(i\right)\right)\frac{1}{L}{\left({\displaystyle {\displaystyle \sum}_{{A}_{n}\left(i\right)>{t}_{th}}}{\phi}_{cn}\left(i\right)\right)}^{2}}$, where L is the length of the nonweak value and ${t}_{th}$ is the threshold value of the nonweak signal. 
Standard deviation of the direct value of the centered nonlinear component of the direct instantaneous phase in nonweak segment${\mathit{\sigma}}_{\mathit{d}\mathit{p}}$  ${\sigma}_{dp}=\sqrt{\frac{1}{{N}_{c}}\left({\displaystyle {\displaystyle \sum}_{{A}_{n}\left(i\right)>{A}_{t}}}{\phi}_{NL}^{2}\left(i\right)\right)\frac{1}{{N}_{c}}{\left({\displaystyle {\displaystyle \sum}_{{A}_{n}\left(i\right)>{A}_{t}}}{\phi}_{NL}\left(i\right))\right)}^{2}}$, where all parameters are similar to σ_{ap} but differs in the absence of the absolute operator in the nonlinear component of the instantaneous phase. 
Standard deviation of the absolute value of the normalized centered instantaneous amplitude of signal segment${\mathit{\sigma}}_{\mathit{a}\mathit{a}}$  ${\sigma}_{aa}=\sqrt{\frac{1}{{N}_{c}}\left({\displaystyle {\displaystyle \sum}_{i=1}^{N}}{A}_{cn}^{2}\left(i\right)\right)\frac{1}{{N}_{c}}{\left({\displaystyle {\displaystyle \sum}_{i=1}^{N}}{A}_{cn}^{2}\left(i\right)\right)}^{2}}$, where ${A}_{cn}$ is the normalized and centered instantaneous amplitude of the incoming signal at the time instant. 
Standard deviation of the absolute value of the normalized centered instantaneous frequency of signal segment${\mathit{\sigma}}_{\mathit{a}\mathit{f}}$  ${\sigma}_{af}=\sqrt{\frac{1}{{N}_{c}}\left({\displaystyle {\displaystyle \sum}_{{A}_{n}\left(t\right)>{A}_{t}}}{f}_{N}^{2}\left(i\right)\right)\frac{1}{{N}_{c}}{\left({\displaystyle {\displaystyle \sum}_{{A}_{n}\left(t\right)>{A}_{t}}}{f}_{N}\left(i\right)\right)}^{2}}$, where ${f}_{N}$ is the normalized frequency. 
Kurtosis of the normalized centered instantaneous amplitude${\mathit{\mu}}_{\mathbf{42}}^{\mathit{a}}$  ${\mu}_{42}^{a}=\frac{E\left\{{A}_{cn}^{4}\left[n\right]\right\}}{{\left\{E\left\{{A}_{cn}^{2}\left[n\right]\right\}\right\}}^{2}}$. 
Kurtosis of the normalized centered instantaneous frequency${\mathit{\mu}}_{\mathbf{42}}^{\mathit{f}}$  ${\mu}_{42}^{f}=\frac{E\left\{{f}_{N}^{4}\left[n\right]\right\}}{{\left\{E\left\{{f}_{N}^{2}\left[n\right]\right\}\right\}}^{2}}$. 
Reference(s)  Key Features  Modulation Set  Advantages  Disadvantages 

[35]  Spectral features  2ASK, 4ASK, 2FSK, 4FSK, 2PSK, 4PSK, 8ASK, 8FSK, and 8PSK. 


[38]  Spectral features  2ASK, 2PSK, 2FSK, 4ASK, 4PSK,16QAM, and 4FSK 


[43]  Spectral features  2ASK, 4ASK, 2FSK, BPSK, and QPSK 


[12]  Sixth and fourthorder cumulants  BPSK, QPSK, QPSK, 16QAM, 64QAM 


[31]  Eighthorder moment and eightorder cumulant  16QAM, 64QAM, and 256QAM 


[48]  Second and fourth cumulants  2ASK, 4ASK, 4PSK, 2FSK, 4FSK, and 16QAM 


[39]  2nd, 4th, and 8thorder cumulants  BPSK, QPSK, QAM, 16QAM, and 64QAM 


[51]  HOCs  BPSK, QPSK, 8PSK, 16QAM, 64QAM, and 256QAM 


[53]  HOCs  QPSK, OQPSK, 8PSK, and 16PSK 


[54]  HOCs  BPSK, QPSK, QAM, 16QAM, and 64QAM 


[76]  Cyclic frequency domain  2FSK, 4FSK, 8FSK, BPSK, QPSK, MSK, and 2ASK. 


[78,79]  Cyclic frequency domain  2FSK, 4FSK, 8FSK 


[81]  Cyclostationary  MQAM 


[62]  FFT  2FSK, 4FSK, 8FSK, 16FSK, 32FSK 


[63]  FFT  2FSK and 4FSK 


[65]  WT  BASK, BFSK, and BPSK 


[66]  WT  PSK, QAM, FSK, and ASK 


[70]  WT  4QAM, 16QAM, and 64QAM 


[84]  Constellation  4QAM, 16QAM, 32QAM, 64QAM, 128QAM, 256QAM 


[85]  Constellation  16QAM, 32QAM, and 64QAM 


Reference(s)  Classifier(s)  Merits  Demerits 

[35,52,53,58,59,63,71,74,88,89]  DT  Simple implementation. This method can accommodate many modulations by adding additional decision branches  Sensitive to noise when the threshold changes 
[43]  ANN  High classification rate. Reduced computational complexity  Ineffective when a large number of input features are used. Small data set 
[76]  ANN  Improved accuracy  Requires long training time for computation. Smallorder modulation types 
[72,94]  ANN  Reduced computational complexity through PCA. Improved accuracy  Smallorder modulation types 
[99]  GP with SVM  Improved classification performance  Two modulation types, namely, 16QAM and 64QAM, are used 
[29]  PSO with SVM  Robust to noise. Good classification  Computationally complex 
[34]  Rough set theory with SVM  Reduced computational complexity  Poor discrimination of MASK at lower SNR 
[54]  GP with KNN  Reduced complexity  Small set of modulation schemes 
[84,85]  Clustering  A priori information about signal is not required.  Requires high SNR 
Reference and Year  Classifier Types  Key Features  Modulation Set  SNR (dB)  PCC %  Advantages  Disadvantages 

[99] 2011  GPSVM  HOC  16QAM, 64QAM  10  99.8 


[29] 2012  SVM with PSO  Combines features: spectral, higherorder statistical, and wavelet  2ASK, 4ASK, 8AASK, 16ASK, 2FSK, 4FSK, 8FSK, 2PSK, 4PSK, 8PSK, 16PSK, 16QAM, 32QAM, 64QAM, ASKPSK4, ASKPSAK16  0  96 


[28] 2014  SVM  Spectral features  2ASK, 4ASK, 2FSK, 4FSK, 2PSK, 4PSK  0  81 


[34] 2017  Rough set with SVM  Instantaneous, HOC, cyclostationarity, wavelet  2ASK, 4ASK, 8ASK, 2FSK, 4FSSK, 8FSK, 2PSK, 4PSK, 8PSK, and 16QAM  5⁓20  95 


[93] 2011  ANN  DWT  2ASK, 4ASK, 8ASK, 2FSK, 4FSK, 8ASK, 2PSK, 4PSK, 8PSK, 4QAM, 16QAM  95.7 

 
[57] 2013  ABC+ANN  HOC  2PSK, 4PSK, 8PSK, 16BPSK, 4QAM, 16QAM, 64QAM  0 20  57.15 76.87 


[43] 2014  ANN  Instantaneous features  2ASK, 4ASK, 2PSK, 4PSK, 2FSK, 4FSK  0  98 


[76] 2016  ANN  Cyclic frequency  2ASK, 2FSK, 4FSK, 8FSK, BPSK, QPSK, MSK  0  95 


[72] 2016  PCA+ANN  Mean value, variance, and central moments. Up to five CWTs  4ASK, 8ASK, 16ASK, 2PSK, 4PSk, 8PSK, 16PSK, 4FSK, 8FSK, 16FSK, 8QAM, 16QAMMSK, OOK  20  100 


[94] 2017  PCA+ANN  Cyclic frequency  2ASK, 2FSK, 4FSK, 8FSK, BPSK, QPSK, MSK  0  95 


[35] 2015  DT  Instantaneous; amplitude, phase, and frequency  2ASK, 4ASK, 8ASK, 2FSK, 4FSK, 8FSK, 2PSK, 4PSK, 8PSK  10  100 


[52] 2006  DT  Fourthorder cumulant  BPSK, QPSK, 8PSK, and π/4 DQPSK  15  100 


[53] 2016  DT  Fourthorder zeroconjugate cumulant  QPSK, OQPSK, 8PSK, and 16PSK  10  100 


[58] 2007  DT  HOCs  4ASK, BPSK, QPSK, OQPSK, 8PSK, 16PSK, 8QAM, 16QAM, 64QAM  15  96 


[88] 2012  DT  Fourthorder and sixthorder cumulants  BPSK, QPSK, OQPSK, 8PSK, π/4DQPSK, 16APK, 16QAM,64QAM  >10  90 


[59] 2016  DT  Instantaneous amplitude, HOC  2ASK, 4ASK, 8ASK, BPSK, QPSK, 8PSK  10  96.6 


[89] 2014  DT  CWT  2FSK, 4FSK, 8FSK, BPSK, QPSK, 2ASK, 4ASK, 8ASK, 16QAM, 64QAM  >2  90 


[71] 2008  DT  Histogram peaks in WT magnitude and mean and variance of normalized histogram  BPSK, QPSK, 8PSK, 16PSK, 2QAM, 4QAM, 8QAM, 16QAM, GMSK, MFSK  5  96.8 


[74] 2016  DT  Wavelet variation coefficient difference and similarity feature  2FSK, 4FSK, 8FSK, 2ASK, 4AASK, 2PSK, 4PSK, 8PSK, 16QAM  >2  92.39 


[63] 2015  DT  Instantaneous amplitude, kurtosis, sumFFT  2ASK, 4ASK, 2FSK, 4FSK, 2PSK, 8PSK  4  98.8 


[103] 2012  GP+KNN  HOC  BPSK, QPSK, 16QAM, and 64QAM  10  98 


[54] 2019  GP+KNN  HOC  BPSK, QPSK, QAM, 16QAM, and 64QAM  0  99.4 


[90] 2014  KNN  Cyclostationarity  BPSK, QPSK, FSK, MSK  10  100 


[84] 2017  Clustering  Constellation  4QAM, 16QAM, 32QAM, 64QAM, 128QAM, 256QAM.  >15  100 


[85] 2013  Clustering  Constellation  16QAM, 32QAM, 64QAM  >15  100 


[106] 2009  Clustering  Constellation  4QAM, 16QAM, 32QAM, 64QAM  5  100 


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
AlNuaimi, D.H.; Hashim, I.A.; Zainal Abidin, I.S.; Salman, L.B.; Mat Isa, N.A. Performance of FeatureBased Techniques for Automatic Digital Modulation Recognition and Classification—A Review. Electronics 2019, 8, 1407. https://doi.org/10.3390/electronics8121407
AlNuaimi DH, Hashim IA, Zainal Abidin IS, Salman LB, Mat Isa NA. Performance of FeatureBased Techniques for Automatic Digital Modulation Recognition and Classification—A Review. Electronics. 2019; 8(12):1407. https://doi.org/10.3390/electronics8121407
Chicago/Turabian StyleAlNuaimi, Dhamyaa H., Ivan A. Hashim, Intan S. Zainal Abidin, Laith B. Salman, and Nor Ashidi Mat Isa. 2019. "Performance of FeatureBased Techniques for Automatic Digital Modulation Recognition and Classification—A Review" Electronics 8, no. 12: 1407. https://doi.org/10.3390/electronics8121407