Performance analysis of Mixed RF/FSO DF relay based on Low Density Parity Check (LDPC) codes using deep learning techniques

This Cooperative optical communication allows a free-space optical network to coordinate antenna distribution and achieve dramatic performance gains similar to those of conventional MIMO systems. It enables significant improvements in terms of network capacity and coverage. Cooperative Free Space Optics (FSO) communication is one of the most promising technologies for the next generation of wireless optical communication networks. FSO systems are particularly attractive for problem last mile access by bridging fiber optic-based backbone connectivity and RF access networks. To address this practical deployment scenario, there has been increasing attention on so-called hybrid RF-FSO (dual-hop) systems where RF transmission is used at one hop followed by FSO transmission at the other. . In this article, we study the performance of hybrid RF/FSO cooperative systems based on error-correcting codes, particularly LDPC codes using an LDPC decoder based on deep learning.The use of mixed RF/FSO cooperative systems improves the reliability and transmission of the network. The results demonstrate improved performance of the LDPC code-based RF/FSO DF cooperative relay system using a deep learning-based LDPC decoder compared to LDPC code-based RF/FSO systems without the use of the deep learning, but also to the DF systems proposed in the existing literature.