PASSIVE OPTICAL NETWORK PLANNING IN LOCAL ACCESS NETWORKS - AN OPTIMIZATION APPROACH UTILIZING GENETIC ALGORITHMS

Optimal plant expansion in local access networks has strategic and fin ancial importance for public telecommunications providers. Probably th e most important future development in the area of digital communicati ons will be the deployment of fibre optics by the telephone companies. This paper describes a genetic algorithm-based planning system develo ped to support the deployment of passive optical network (PON) technol ogy in the local access network. Current planning practice is examined and the rationale behind the use of an optimisation method is explain ed. This paper presents a short introduction to genetic algorithms and outlines how they can be applied to PON planning. The system structur e and optimisation framework associated with a genetic algorithm-based PON planning optimiser is discussed. The planning system developed by the authors can be used to determine the location, type, quantity and connection arrangements of PON components in networks. Results for so me test problems are given; the optimiser is shown to produce low-cost solutions in an acceptable time.