Blind signal-type classification using a novel robust feature subset selection method and neural network classifier

annals of telecommunications - annales des télécommunications (Impact Factor: 0.7). 05/2010; DOI: 10.1007/s12243-010-0180-4
1 Follower
11 Reads
  • EJC Supplements 11/2010; 8(7):181-181. DOI:10.1016/S1359-6349(10)72280-1 · 9.39 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Automatic recognition of the communication signals plays an important role for various applications. Most of the existing techniques require high levels of signal to noise ratio (SNR). In this paper, we propose a high efficient technique for classification of the digital modulations that requires a low level of SNRs. This technique includes two main modules: feature extraction module and the classifier module. In the feature extraction module we use the auto-regressive modeling together other useful features. These features are a combination set of the entropy and energy of the signal, variance of the coefficients wavelet packet transform, fourth order of moment and zero-crossing rate. In the classifier module we have used the two structures of the neural networks: multi-layer perceptron (MLP) neural network and radial basis neural networks. Simulation results show the proposed technique has very high recognition accuracy for identification of the considered digital modulations even at very low SNRs.
    Measurement 10/2011; 44(8):1475-1481. DOI:10.1016/j.measurement.2011.05.019 · 1.48 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents the recognition of digital signal modulation by combining four cumulants as signal features and a threshold discriminator assigned for each cumulant. In this paper, the problem of recognition has been solved by finding a suitable relationship among these features and their classifier. The decision of each classifier has been compared with other classifier outputs. Then, the classifier with the highest priority has been considered as the final efficient decision. The priority of each classifier is based on how its cumulant is robust against a wide range of signal to noise ratio (SNR), for different types of digital modulations under investigation. Many types of quadrature amplitude modulation (16, 32, 64, 128, and 256) have been recognized with low SNR down to −5 dB. Using such combination, the average recognition percentage was increased with low SNR. The recognition gained 81.166%–100% for 4096 samples while the percentages are from 86.9% to 98.6% with a variable number of samples (256, 512, 1024, and 2048).
    IETE Journal of Research 01/2015; 61(1):65 - 71. DOI:10.1080/03772063.2014.985263 · 0.19 Impact Factor