Capacity Optimization of the Next-Generation Passive Optical Networks Based on Genetic Algorithm

This paper proposes an analytical and a numerical models for Next Generation Passive Optical Network (NGPON) by combining Gigabit-PON (GPON) and 10 Gigabit asymmetrical PON (XGPON) technologies. This allows for a larger number of subscribers to be accommodated on the network, up to 256. A capacity optimization procedure based on Genetic Algorithm (GA) techniques is then proposed and analyzed. The uplink and downlink capacities are maximized by optimizing various link parameters, such as the number of Optical Network Units (ONUs), the average optical transmitter power, the receiver sensitivity and the network operating margins. Simulation results shown that, among others, the Distributed-feedback laser (DBF) average power, the Positive-Intrinsic-Negative (PIN) sensitivity and the network margin are the key parameters entrusted to the GA technique in order to maximize the downlink capacity. In contrast, the Fabry Perot (FP) average power, the Avalanche Photodiode (APD) sensitivity and network margin are found to be the most influential parameters optimized by GA to maximize uplink capacity. Significantly, it is demonstrated that the GA technique, when used in optimizing NGPON capacity, enabling 256 subscribers and offering a data rates up to 71.02 Gb/s and 390.56 Gb/s in the upstream and downstream directions, respectively.