The use of forward error correction (FEC) codes in future deployments of access networks is quite foreseeable and requires an analysis of the proper code for this purpose. Therefore, an overview of the most important classes of FEC
codes is presented, assessing the most suitable to be used in time-division multiple access networks. Concerning the traffic characteristics and the properties of such networks, the Reed–Solomon codes were found to be the most appropriate, and an analysis of the performance improvement obtained with these codes is reported. The upstream performance of 10 Gbits/s Ethernet passive optical network systems employing the selected FEC code is evaluated and the benefits of introducing FEC are assessed, aiming for the minimization of characteristic impairments of such networks.