A Method for Modulation Recognition Based on Entropy Features and Random Forest

Since the low SNR environment, generally the modulation recognition rate of signal modulation type is not very high. In this paper, we studied an automatic recognition method of communication signal modulation type in the low SNR. According to analyze the signal entropy as the feature, three characteristics are selected, and the random forest is as the classifier, finally we get a high recognition-rate of several communication signal modulation types in low SNR. Via simulation, it is demonstrated that the method achieves excellent performance to recognize signal type under different signal-to-noise ratio (SNR). The signal recognition rate is more than 95% when SNR is higher than 5 dB except QPSK signal. In a word, it is simple to design the recognition system, and it will have important application value.

[1]  J. Reichert,et al.  Automatic classification of communication signals using higher order statistics , 1992, [Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  Zhendong Yin,et al.  Research of Communication Signal Modulation Scheme Recognition Based on One-Class SVM Bayesian Algorithm , 2009, 2009 5th International Conference on Wireless Communications, Networking and Mobile Computing.

[3]  Haibo Yang,et al.  Noise Reduction for Chaotic Time Series Based on Singular Entropy , 2013, 2013 Sixth International Symposium on Computational Intelligence and Design.

[4]  Zhao Zhijin,et al.  A new method for modulation type recognition based on the time frequency representations , 2002, 6th International Conference on Signal Processing, 2002..

[5]  Li Shi-ping,et al.  Modulation recognition algorithm of digital signal based on support vector machine , 2012, 2012 24th Chinese Control and Decision Conference (CCDC).

[6]  Jason H. Moore,et al.  Feature Selection using a Random Forests Classifier for the Integrated Analysis of Multiple Data Types , 2006, 2006 IEEE Symposium on Computational Intelligence and Bioinformatics and Computational Biology.

[7]  Shuliang Liu,et al.  The application of entroy power method improved on wind power generation project , 2011, 2011 International Conference on Consumer Electronics, Communications and Networks (CECNet).

[8]  Zhang Hui,et al.  Terrain classification in field environment based on Random Forest for the mobile robot , 2016, 2016 35th Chinese Control Conference (CCC).

[9]  Hong Man,et al.  Distributed Automatic Modulation Classification Based on Cyclic Feature via Compressive Sensing , 2013, MILCOM 2013 - 2013 IEEE Military Communications Conference.

[10]  Ali Abdi,et al.  Survey of automatic modulation classification techniques: classical approaches and new trends , 2007, IET Commun..

[11]  Arturo Garcia-Perez,et al.  FPGA-Based Online Detection of Multiple Combined Faults in Induction Motors Through Information Entropy and Fuzzy Inference , 2011, IEEE Transactions on Industrial Electronics.