Signal Detection and Classification in Communication Network Using Quadrature Amplitude Modulation with Signal to Noise Ratio

In recent years, signal detection in the telecommunication networks automated dependent radio frequency for aircraft identification and verification with emotion recognition, radar and biomedical systems. However, the existing Deep Learning (DL) based techniques failed to interact and detect noise and dynamic environments. To overcome the problem, a Quadrature Amplitude Modulation and Signal-to-Noise Ratio (QAM-SNR) method is proposed for digital communication and signal detection of both the amplitude and phase of carrier signal. The proposed QAM-SNR method is developed on Dataset for Emotion Analysis using Psychological signals (DEAP) dataset, which consists of electromyogram and electrooculogram for signal detection in stimuli using Low-Pass Filter for detection. Then, the preprocessed data of detection in signal is extracted using Generative Adversarial Network and Spiking Neural Network (GAN-SNN) for enhanced detection with improved robustness to adversarial attacks. Finally, QAM-SNR is employed for efficient use of bandwidth in detection of the signals for communication networks with high quality of results. Experimental results of the proposed QAM-SNR method attained accuracy (99.70%), precision (99.78%), F1-score (99.87%) are higher than existing detection techniques such as Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory (Bi-LSTM).