Decoding 5G-NR Communications via Deep Learning
Upcoming modern communications are based on 5G specifications and aim at providing solutions for novel vertical industries. One of the major changes of the physical layer is the use of Low-Density Parity-Check (LDPC) code for channel coding. Although LDPC codes introduce additional computational complexity compared with the previous generation, where Turbocodes where used, LDPC codes provide a reasonable trade-off in terms of complexity-Bit Error Rate (BER). In parallel to this, Deep Learning algorithms are experiencing a new revolution, specially to image and video processing. In this context, there are some approaches that can be exploited in radio communications. In this paper we propose to use Autoencoding Neural Networks (ANN) jointly with a Deep Neural Network (DNN) to construct Autoencoding Deep Neural Networks (ADNN) for demapping and decoding. The results will unveil that, for a particular BER target, $3$ dB less of Signal to Noise Ratio (SNR) is required, in Additive White Gaussian Noise (AWGN) channels.