Feature-based automatic modulation recognition of chaotic and conventional schemes in the AWGN and Rayleigh fading channels
AEU - International Journal of Electronics and Communications, cilt.216, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 216
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.aeue.2026.156464
- Dergi Adı: AEU - International Journal of Electronics and Communications
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Applied Science & Technology Source, Compendex, INSPEC, Academic Search Ultimate (EBSCO), Engineering Source (EBSCO), Technology Collection (ProQuest)
- Anahtar Kelimeler: Automatic modulation recognition (AMR), Chaotic modulation, Higher-order statistics (HOS), Rayleigh fading channel, Spectral feature extraction, Support vector machine (SVM)
- İstanbul Üniversitesi-Cerrahpaşa Adresli: Evet
Özet
Automatic modulation recognition (AMR) is a fundamental component of adaptive and cognitive wireless communication systems. This study proposes a support vector machine (SVM)-based AMR framework that classifies both conventional and chaotic modulation schemes using Higher-Order Statistics (HOS) and spectral features. A total of 13 modulation types are considered, including chaos-based CSK, DCSK, and CBPSK generated using Chua, Rössler, and Lorenz systems, as well as conventional QPSK, 8PSK, 16QAM, and 2FSK modulations. The performance of the proposed system is evaluated under additive white Gaussian noise (AWGN) and Rayleigh fading channel conditions across various signal-to-noise ratio (SNR) levels. To ensure statistical reliability, the reported results are obtained from 20 independent Monte Carlo trials. The simulation results demonstrate that the proposed method achieves 100% classification accuracy at 18 dB and 99% at 5 dB under AWGN conditions. Under Rayleigh fading, the accuracy reached 97% at 18 dB and decreased to 74% at 5 dB. Additional multipath and Doppler analyses confirmed the robustness of the proposed framework under challenging wireless channel conditions. Overall, the proposed feature based SVM framework offers an effective, interpretable, and computationally efficient solution for multiclass modulation recognition in noisy and fading environments.