CNN-Based Learning System in a Generalized Fading Environment

In this paper, we investigate the Block Error Rate (BLER) performance of a Convolutional Neural Network (CNN)- based autoencoder as a self-learning communication system under a generalized fading condition. To this end, the α-µ model is chosen, both for its flexibility and applicability to practical fading scenarios. For comparison and consistency purposes, the analytical BLER is also explored under the same environment for different classical modulation schemes. Our simulation results show that the studied architecture is able to converge quickly maintaining its generalization capability under various fading environments.