Classification of chaos-based digital modulation techniques using wavelet neural networks and performance comparison of wavelet families

This paper presents a comparative study of implementation of feature extraction and classification algorithms based on wavelet neural networks (WNN) for chaos-based digital modulation (CBDM) classification. Thirteen different feature extraction methods are generated by separately using Daubechies, Biorthogonal, Coiflets, and Symlets wavelet filters. WNN model is used, which consists of two layers: wavelet entropy and multi-layer perceptron (MLP) neural networks for expert CBDM classification. The chaos-based digital modulated signals used in this experimental study are Chaos Shift Keying (CSK), Chaotic On–Off Keying (COOK), Differential Chaos Shift Keying (DCSK), Correlation Delay Shift Keying (CDSK), Symmetric Chaos Shift Keying (SCSK) and Frequency-Modulated Differential Chaos Shift Keying (FM-DCSK). The performance of this comparison system is evaluated by using total 1806 CBDM signals for each of these feature extraction methods. Mean correct classification rate is about 98.76% for the sample CBDM signals.