李秀惠臺灣大學：資訊工程學研究所吳青坡Wu, Ching-PoChing-PoWu2007-11-262018-07-052007-11-262018-07-052006http://ntur.lib.ntu.edu.tw//handle/246246/53776近年來因為無線通訊技術的快速發展，使得許多的無線網路架構也接續被提出與討論，行動隨意網路便為其中之一。由於行動隨意網路的所有節點都可隨意移動至任何地方，所以該網路拓樸時常改變，也因此延伸出許多繞徑機制的議題。例如：資料傳輸的安全性、繞徑的穩定性、節省電源…等等。 在繞徑穩定性為主的議題中，許多的相關研究皆提出以評價機制來達到穩定性要求，這些相關研究大都著重於小範圍的路徑局部穩定。然而只考量局部穩定並不能保證被選定的路徑在所有可用路徑中是最佳的，因此我們將以繞徑的“整體”穩定度為此篇論文的主要探討方向。 此篇論文將以評價機制來判斷哪一條路徑是“整體”最佳，我們提出下列幾項機制： 1.利用評價值來決定下一個最佳(更穩定與更近)的跳點: 我們以目前經過的所有跳點的評價乘積來代表該條路徑的穩 定程度，數值越大代表越穩定。 2.加入漣漪作用的概念來擴大受影響的範圍: 當路徑因某一跳點的影響而斷裂，我們將以此跳點為中心， 通知該跳點附近的所有鄰點進行評價值的調整，以儘量避免 再次挑選此節點做為新路徑的跳點。Mobile ad-hoc networking (MANET) has been proposed for many years. There are several important issues in ad-hoc network like routing, trust mechanism, etc. However, most of routing protocols using reputation mechanism only pay attention to one hop nodes. Compared with other related systems, our protocol not only focuses on reputation mechanism but also adopts globalization concept. In this study, we assume that the network layer is based on the Ad-hoc On-demand Distance Vector (AODV) routing. This paper uses reputation to decide what the route is the best in global area. In order to achieve this, we propose some mechanisms for routing in MANET. First, the notion of reputation is used to decide which neighbor node is better (more stable and faster) for next hop. Second, the ripple concept is adopted for reputation adjustment when the route is broken. Finally, the performance evaluation results show that our system not only decreases the accumulative frequency of route broken but also improves the goodput ratio.中文摘要....................................I Abstract...................................II Chapter 1.Introduction......................1 1.1 Motivation.......................1 1.2 Background.......................5 1.3 Organization of This Thesis......9 Chapter 2.Related Works....................10 2.1 Reputation system...............10 2.2 Credit mechanism ................12 2.3 Multi-path routing protocol.....13 Chapter 3.System Implementation............15 3.1 Our Assumption..................15 3.2 Reputation Evaluation...........16 3.2.1 Reputation Setup..............16 3.2.2 Non-linear Evaluation.........18 3.3 Global Route....................21 3.3.1 Global Success Rate of Route..22 3.3.2 Function Process..............24 3.3.3 Extended Issues...............29 3.4 Ripple Effect...................34 Chapter 4.Simulations and Results..........38 4.1 Simulation Environment..........38 4.2 Results and Analysis............40 Chapter 5.Conclusions and Future Works.....47 5.1 Conclusions.....................47 5.2 Future Works....................48 References.................................492120222 bytesapplication/pdfen-US行動隨意網路繞徑評價機制MANETreputationrouting protocolthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/53776/1/ntu-95-R93922083-1.pdf