A nested genetic algorithm for topologyoptimization in LRPON

This paper researches and formulates a topology optimization model in long-reach passive optical network (LRPON). The objective is to minimize the deployment cost with certain physical constraints. A nested genetic algorithm (NGA) is proposed to solve the topology optimization problem. The NGA consists of two loops, and the outer-loop GA (OLG) is a simple genetic algorithm (SGA) to solve the location problem of splitters, while the inner-loop GA (ILG) is a partheno-genetic algorithm (PGA) to complete the construction of the spanning tree. Additionally, three scenarios with different network scales are simulated to examine the efficiency of the proposed algorithm.