Neural Computing and Applications

, Volume 29, Issue 9, pp 351–360 | Cite as

Performance analysis and improvement of machine learning algorithms for automatic modulation recognition over Rayleigh fading channels

  • M. A. Hazar
  • N. Odabasioglu
  • T. Ensari
  • Y. Kavurucu
  • O. F. Sayan


Automatic modulation recognition (AMR) is becoming more important because it is usable in advanced general-purpose communication such as, cognitive radio, as well as, specific applications. Therefore, developments should be made for widely used modulation types; machine learning techniques should be employed for this problem. In this study, we have evaluated performances of different machine learning algorithms for AMR. Specifically, we have evaluated performances of artificial neural networks, support vector machines, random forest tree, k-nearest neighbor, Hoeffding tree, logistic regression, Naive Bayes and Gradient Boosted Regression Tree methods to obtain comparative results. The most preferred feature extraction methods in the literature have been used for a set of modulation types for general-purpose communication. We have considered AWGN and Rayleigh channel models evaluating their recognition performance as well as having made recognition performance improvement over Rayleigh for low SNR values using the reception diversity technique. We have compared their recognition performance in the accuracy metric, and plotted them as well. Furthermore, we have served confusion matrices for some particular experiments.


Automatic modulation recognition Artificial neural networks Support vector machines Random forest tree k-Nearest neighbor Hoeffding tree Naive Bayes Logistic regression Gradient Boosted Regression Tree 


Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.


  1. 1.
    Azzouz EE, Nandi AK (1995) Automatic identification of digital modulation rypes. Signal Process 47:55–69CrossRefzbMATHGoogle Scholar
  2. 2.
    Wong MLD, Nandi AK (2004) Automatic digital modulation recognition using artificial neural network and genetic algorithm. Signal Process 84:351–365CrossRefzbMATHGoogle Scholar
  3. 3.
    Valipour MH, Homayounpour MM, Mehralian MA (2012) Automatic digital modulation recognition in presence of noise using SVM and PSO. In: IEEE international symposium on telecommunications, pp 378–382Google Scholar
  4. 4.
    Roganovic MM, Neskovic AM, Neskovic NJ (2009) Application of artificial neural networks in classification of digital modulations for software defined radio. In: IEEE, EUROCON, pp 1700–1706Google Scholar
  5. 5.
    Ozen A, Ozturk C (2013) A novel modulation recognition technique based on artificial bee colony algorithm in the presence of multipath fading channels. In: 36th international conference on telecommunications and signal processing (TSP), pp 239–243Google Scholar
  6. 6.
    Vladimir DO, Miroslav LD (2009) Automatic modulation classification algorithm using higher-order cumulants under real-world channel conditions. IEEE Commun Lett 13(12):917–919CrossRefGoogle Scholar
  7. 7.
    Ghasemi S, Gangal A (2014) An effective algorithm for automatic modulation recognition. In: IEEE, signal processing and communications applications conference (SIU), pp 903–906Google Scholar
  8. 8.
    Hassan K, Dayoub I, Hamouda W, Berbineau M (2009) Automatic modulation recognition using wavelet transform and neural network. In: 9th international conference on intelligent transport systems telecommunications, pp 234–238Google Scholar
  9. 9.
    Hassan K, Dayoub I, Hamouda W (2010) Automatic modulation recognition using wavelet transform and neural networks in wireless systems. EURASIP J Adv Signal Process 2010(1):532898CrossRefzbMATHGoogle Scholar
  10. 10.
    Dubey HC, Nandita, Tiwari AK (2012) Blind modulation classification based on MLP and PNN. In: Students conference on engineering systems (SCES), pp 1–6Google Scholar
  11. 11.
    Keshk M, Elrabie E, El-Samie F, El-Naby M (2013) Blind modulation recognition in wireless MC-CDMA systems using a support vector machine classifier. Wirel Eng Technol 4(3):145–153CrossRefGoogle Scholar
  12. 12.
    Mobasseri BG (2000) Digital modulation classification using constellation shape. Signal Process 80:251–277CrossRefzbMATHGoogle Scholar
  13. 13.
    Ghauri SA, Qureshi IM, Shah I (2014) Modulation classification using cyclostationary fatures on fading channels. Res J Appl Sci Eng Technol 7(24):5331–5339Google Scholar
  14. 14.
    Al-Makhlasawy RM, Elnaby MMA, El-Khobby HA, El-Samie FEA (2012) Automatic modulation recognition in OFDM systems using cepstral analysis and A fuzzy logic interface. In: International conference on informatics and systems (INFOS), pp CC56–CC62Google Scholar
  15. 15.
    El Rube IAH, El-Madany NE.-d (2010) Cognitive digital modulation classifier using artificial neural networks for NGNs. In: International conference on wireless and optical communications, pp 1–5Google Scholar
  16. 16.
    Sobolewski S, Adams WL, Sankar R (2012) Automatic modulation recognition techniques based on cyclostationary and multifractal features for distinguishing LFM, PWM and PPM waveforms used in radar systems as example of artificial intelligence implementation in test. In: IEEE AUTOTESTCON, pp 335–340Google Scholar
  17. 17.
    Hazar MA, Odabaiolu N, Ensari T, Kavurucu Y (2015) Evaluation of machine learning algorithms for automatic modulation recognition. In: The 22nd international conference on neural information processing (ICONIP), pp 208–215Google Scholar
  18. 18.
    Platt J (1998) Fast training of support vector machines using sequential minimal optimization. Advances in kernel methods—support vector learning, MIT PressGoogle Scholar
  19. 19.
    Breiman L (2001) Random forests. Mach Learn 45(1):5–32CrossRefzbMATHGoogle Scholar
  20. 20.
    Aha D, Kibler D (1991) Instance-based learning algorithms. Mach Learn 6:37–66zbMATHGoogle Scholar
  21. 21.
    Hulten G, Domingos P (2000) Mining high-speed data streams. In: Proceedings of the 6th ACM SIGKDD international conference on knowledge discovery and data mining, pp 71–80Google Scholar
  22. 22.
    Le Cessie S, Van Houwelingen JC (1992) Ridge estimators in logistic regression. Appl Stat 41(1):191–201CrossRefzbMATHGoogle Scholar
  23. 23.
    John GH, Langley P (1995) Estimating continuous distributions in Bayesian classifiers. In: Proceedings of the 11th conference on uncertainty in artificial intelligence. Morgan Kauffman Publisher Inc., San Mateo, pp 338–345Google Scholar
  24. 24.
    Swami A, Sadler BM (2000) Hierachical digital modulation classification using cumulants. IEEE Trans Commun 48(3):416–429CrossRefGoogle Scholar
  25. 25.
    Friedman J, Hastie T, Tibshirani R (2000) Additive logistic regression: a statistical view of boosting. Ann Stat 28(2):337–407MathSciNetCrossRefzbMATHGoogle Scholar

Copyright information

© The Natural Computing Applications Forum 2017

Authors and Affiliations

  1. 1.Electrical and Electronics EngineeringIstanbul Medeniyet UniversityIstanbulTurkey
  2. 2.Electrical and Electronics EngineeringIstanbul UniversityIstanbulTurkey
  3. 3.Computer EngineeringIstanbul UniversityIstanbulTurkey
  4. 4.Computer EngineeringTurkish Naval AcademyIstanbulTurkey
  5. 5.Information Technologies and Communication AuthorityAnkaraTurkey

Personalised recommendations