林永松2006-07-262018-06-292006-07-262018-06-292002http://ntur.lib.ntu.edu.tw//handle/246246/18827In this report, the survivable wireless communications network design problem is considered, where smart antennas may be adopted for sectorization. An optimization-based algorithm is then proposed to provide wireless network planners and managers with both decision and operation support, so that the total network deployment and operation cost can be minimized under QoS (Quality-of-Service) and survivability constraints. In the integrated model presented in this paper, the following five major problems are jointly considered: (i) the MTSO (Mobile Telephone Switching Office) configuration and capacity assignment problem; (ii) the wired backbone network design problem; (iii) the base station allocation, sectorization and power control problem; (iv) the channel assignment problem; and (v) the system survivability problem. These five problems are inherently coupled. The joint problem is formulated as a combinatorial optimization problem and the basic approach to the algorithm development is Lagrangean relaxation. In computational experiments the proposed algorithm achieves up to 23% and 21% cost reduction, respectively, over two sensible heuristics.application/pdf220584 bytesapplication/pdfzh-TW國立臺灣大學資訊管理學系暨研究所otherhttp://ntur.lib.ntu.edu.tw/bitstream/246246/18827/1/902213E002075.pdf