Conference Paper

Digital modulation recognition using support vector machine classifier

Dept. of Electr. & Comput. Eng., George Washington Univ., DC, USA
DOI: 10.1109/ACSSC.2004.1399565 Conference: Signals, Systems and Computers, 2004. Conference Record of the Thirty-Eighth Asilomar Conference on, Volume: 2
Source: IEEE Xplore

ABSTRACT We propose four features to classify amplitude shift keying with two levels and four levels, binary phase shift keying, quadrature phase keying, frequency shift keying with two carriers and four carriers. After that we present a new method of classification based on support vector machine (SVM) that uses the four proposed features. We study the performance of SVM classifier and compare it to the previous work done in the literature on the digital modulation classification problem.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Automatic digital modulation recognition in intelligent communication systems is one of the most important issues in software radio and cognitive radio. In this paper a new method will be presented for automatic digital modulation classification in presence of additive white Gaussian noise (AWGN). In this method a set of three different types of features is extracted to be employed in recognition process. Classification is based on support vector machine (SVM) as a powerful method for pattern recognition, and particle swarm optimization (PSO) to configure kernel parameters. Computer simulations of 16 different types of digitally modulated signals corrupted by AWGN are carried out to measure the performance of the method. Employing multiple SVMs in a hi-erarchical structure as inter-class and intra-class classifiers and also our proposed method for feature selection based on features impact on severance, presents good results in simulations. The results show that with infinite SNR, accuracy tends to 99.9%. Also this method shows eligible robustness in presence of noise as we can see in experiments conducted using low SNR data.
    6'th International Symposium on Telecommunications (IST'2012), Tehran; 01/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, a new classification method based on Kernel Fisher Discriminant Analysis(KFDA) is brought forward in the MPSK signals modulation classification. The fourth order cumulants of the received signals are used as the classification vector firstly, then the kernel thought is used to map the feature vector impliedly to the high dimensional feature space and linear fisher discriminant analysis is applied to signal classification. The two classifiers based on kernel function - Support Vector Machine and Kernel Fisher Discriminant Analysis are introduced in detail. In order to build effective and robust SVM and KFDA classifiers and compared with each other, the radial basis kernel function is selected, one against one or one against rest of multi-class classifier is designed, and method of parameter selection using cross-validating grid is adopted. Through the experiments it can be concluded that compared with SVM classifier, KFDA can get almost the same classification accuracy and requires less time.
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we propose a feature to distinguish frequency from amplitude-phase digital modulations. We compare the performance of the feature where every symbol is sampled more than once to that where every symbol is sampled only once. The feature is based on the product of two consecutive signal values and on time averaging of the imaginary part of the product if a symbol is sampled more than once. First, the conditional probability density functions of the feature given the present modulation are determined. The central limit theorem for strictly stationary m-dependent sequences is used to obtain Gaussian approximations. Then thresholds are determined based on the minimization of the total probability of misclassification. Following that, effects of fast and slow fading, and of the symbol period and delay being non-integer multiples of sampling period on the performance are studied. In the course of doing that, the proposed classifier is compared to the maximum likelihood classifier and the wavelet based classifier using support vector machine.
    Information Sciences and Systems (CISS), 2013 47th Annual Conference on; 01/2013


1 Download
Available from