Abstract
This dissertation proposes an efficient approach to design the broadband transport network infrastructure taking cost and performance into account. In the network planning, this dissertation focuses on how to use ATM, SDH, and DCS technology characteristics to design broadband transport network to meet QoS requirements. It concentrate on the planning method, model, and system development, including virtual path (VP) capacity assignment, SDH network planning, and DCS network restoration planning models, and their applications.A heuristic method based on the tabu search has been developed for the problem of minimizing the maximum link load by selecting a minimum hop route for loading every bandwidth or circuit demand pair in a network. In this dissertation, We will show that it is NP-complete to solve this problem. However, an efficient algorithm has been found to solve the problem in polynomial time here. The VP-based ATM network is used in this study to illustrate the proposed algorithm. Computational results are given to show that the solution procedure is effective in obtaining the satisfactory solution.SDH has brought about fundamental changes in the way of transmission networks are planned and managed. The extra flexibility in an SDH network makes its optimization more complicated. In this dissertation, We first outline some characteristics of SDH network, including multiplexing scheme, system and configuration features and then discuss the various equipment and configurations to determine their application networks. We present certain models to support the planning of cost efficient and robust SDH networks. Bi-objective Hard C-Means (BHCM) cluster method is used in network hierarchy planning, and some heuristic algorithms are adopted for various SDH network configuration planning. Finally, some guidelines are summarized.This dissertation employs tabu search to solve the network spare capacity planning problem. The study involves the determination of the necessary spare capacity on each link of a network required to achieve a certain desired level of restoration at minimum cost during network failure. We have developed here a procedure based on tabu search. The computational results show that the proposed tabu search algorithm offers an effective and efficient way for handling this network restoration problem.SDHTOOL is developed here which is a decision-support system that applies a heuristic approach to SDH network architecture and evolution planning. SDH network planning is a highly complex process and in face of keen competition and rapid changes in telecommunications field, strategic decisions and evolution plans have to be made under pressing time. Thus, a flexible, easy-to-use and state-of-the-art planning tool is always in demand. SDHTOOL offers a simplified planning process in selecting the most appropriate network architecture in terms of low cost and high survivability. It helps to achieve global optimization ensuring that services and protection objectives are met economically as well as maintaining compatibility between new and old equipment in the transition. Compared to conventional labor intensive planning method, SDHTOOL not only reduces planning time significantly but also cuts cost 15 to 33%.