Classification of Digital Modulated COVID-19 Images in the Presence of Channel Noise Using 2D Convolutional Neural Networks

The wireless environment poses a significant challenge to the propagation of signals. Different effects such as multipath scattering, noise, degradation, distortion, attenuation, and fading affect distribution signals adversely. Deep learning techniques can be used differentiate among different modulated for reliable detection in communication system. This study aims at distinguishing COVID-19 disease images that have been by digital modulation schemes are then passed through noise channels classified using deep models. We proposed comprehensive evaluation 2D Convolutional Neural Network (CNN) architectures task multiclass (24-classes) classification presence fading. It is between Binary Phase Shift Keying, Quadrature 16- 64-Quadrature Amplitude Modulation Additive White Gaussian Noise, Rayleigh, Rician channels. obtained mixed results under settings data augmentation, disharmony batch normalization (BN), dropout (DO), well lack BN network. In this study, we found best performing model 2D-CNN DO trained 10-fold cross-validation (CV) with small value before softmax after every convolution fully connected layer along layers while least 5-fold CV without augmentation.