An ns-based Bluetooth Simulation Environment

• Main Authors: JiaRuey Sheu, 許家瑞
• Other Authors: YuhJzer Joung
• Format: Others
• Language: zh-TW
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/53702253514223028102

碩士 === 國立臺灣大學 === 資訊管理研究所 === 90 === While network technologies became more popular, the new issue, mobility, is drawing people’s attention because wired networks cannot be connective without physical links. This property limits information services on space and time dimensions. Consequently, most people expect that wireless networks will cause another huge revolution in the near future. When the time comes, human will interact with information technologies in a completely new way, and new applications will emerge. Therefore, technologies related to wirelesses networks are rapidly developed recently. One of such technologies is Bluetooth. Wireless networks have some properties different from existing wired networks, such as channel transmission quality and the interference and fading of signals. These properties make it impossible to port some technologies developed in wired networks to wireless environments. Hence, this field is yet to be explored both in academia and in industry. At this time, we need a good network analysis tool like computer simulation because it can handle complex and dynamic models. However, up to now, open source network simulation packages are less mature compared to commercial and proprietary products. For example, BlueHoc, which is a Bluetooth extension based on NS（Network Simulator）, is developed by Kumar, a researcher of IBM. It focuses on simulation in a fixed-topology network, e.g. connection establishment under fixed-topology, QoS negotiation, and performance of TCP/IP. Therefore, researches which focus on Bluetooth network topology cannot be well supported by BlueHoc. For this reason, we wish to construct a flexible and complete simulator to provide a more generic simulation environment for Bluetooth. The first and most important point is to break the constraint of fixed-topology. Besides, we also consider the behavior and mobility of Bluetooth devices, presentations of simulation results, and the limitation of the number of piconets. Finally, we examine our simulator in some aspects through two topology construction protocols and one simple scenario with mobile devices. In these real cases, we break the constraint of fixed-topology. Moreover, we separate the implementation of algorithms apart from the implementation of simulator so that it is easier for algorithm developers to simulate their algorithms in our simulator. As for the mobility of Bluetooth devices, our simulator also makes some improvement over the presentation of simulation results, and the limitation of the number of piconets.