Automatic Digital Modulation Recognition Based on Genetic-Algorithm-Optimized Machine Learning Models

Recognition of the modulation scheme is intermediate step between signal detection and demodulation received in communication networks. Automatic recognition (AMR) plays a central role many applications, especially military security sectors. In general, several properties are extracted employed for AMR. Selecting appropriate features has significant impact on increasing efficiency this paper, we implement compare digital via multi-layer perceptrons (MLP), radial basis function (RBF), adaptive neuro-fuzzy inference system (ANFIS), decision tree (DT), naïve Bayes (NB) algorithms. addition, optimal parameters each model obtained by utilizing genetic algorithm (GA). A series studies conducted work order to determine different algorithms identifying modulated signals with commonly used modulations. Numerous computer simulations performed presence additive white Gaussian noise (AWGN) signal-to-noise ratio (SNR) ranging from -10 dB 30 dB. The simulation results comparisons previous demonstrate that applying proposed along selected leads enhancement accuracy speed automatic determination types at low SNRs. convergence rates models enhanced.