Applications of Random Walk Model on Mobile Networks

無線網路技術的成熟以及存取設備的普及吸引更多的行動使用者使用，使用者可在任何的無線網路範圍內於任何時間與地點存取網際網路，因此，在無線環境當中如何管理使用者的行動性是一重要的議題，並許多行動性模型已提出以研究行動性管理。在本論文中，我們著重由林一平教授所提出的隨機漫步模型的相關應用，以研究無線異質網路上不同行動管理機制的效能。我們尤其會研究如何修改隨機漫步模型以在不同的行動網路上作效能分析，我們考慮下面三個例子： 在第一個例子中，我們應用隨機漫步至模擬實驗中來分析802.16網狀網路中的效能。本論文提出結合分散與集中方案來將802.16網狀網路所提出的集中排程與分散排程整合，使得minislot配置可以更有彈性並且效能可以更高，我們應用隨機漫步模型所建議的拓樸到模型實驗中以分析提出方案的效能。 在第二個例子中，我們採用隨機漫步模型來分析行動IP網路AAA架構中的效能。AAA伺服器會產生並且分配金鑰給網路節點與使用者以保護信令訊息不被惡意攻擊者竊聽。我們提出一金鑰快取機制可當地認證使用者以減少信令消耗與延遲。我們也提出一金鑰大小選擇機制使得頻寬消耗可以更低。我們設計了隨機漫步模型以及模擬實驗來分析金鑰快取與金鑰大小選擇機制的效能。 在第三個例子當中，我們應用隨機漫步於分析模型以檢視異質網路整合支援節點 (HISH) 的效能，其在應用層整合異質網路。此節點會針對初始使用者執行內容格式轉換，針對交手使用者執行資料通道。HISN的服務範圍的大小會影響初始使用者以及交手使用者的數量，並會影響內容格式轉換以及資料通道的耗費。我們設計了隨機漫步模型以針對HISN的服務範圍大小來分析。The mature of wireless network technologies and the spread of access devices have attracted numerous mobile users. Through wireless access technologies, user can access Internet services anytime and anywhere in any wireless signal covered area. Therefore, in the wireless environment, how to manage user mobility is a critical issue and several mobility models were proposed to study the mobility management. In this dissertation, we focus on the applications of the random walk model proposed by Akyildiz and Lin to study the performance of wireless access technologies and heterogeneous networks with different kinds of mobility management approaches. Specially, we study how to modify the random walk model for the performance evaluations on different kinds of mobile networks. The following three case studies are considered. In the first case, we apply the random walk model to the simulation experiments to investigate the performance of the 802.16 mesh network. This dissertation proposes the Combined Distributed and Centralized (CDC) and the CDC with Queue capability (CDCQ) schemes to combine the centralized scheduling and distributed scheduling proposed by 802.16 mesh network so that the minislot allocation can be more flexible, and the utilization is increased. We apply the topology suggested by the random walk model to the simulation experiments to investigate the performance for the proposed schemes. In the second case, we adopt the random walk model to analyze the performance of mobile IP network with the Authentication, Authorization, and Accounting (AAA) framework. The AAA server in the home network (i.e., AAAH) generates and distributes session keys to network nodes and users to protect signaling messages from eavesdropping by malicious attackers. We propose an AAA Key Caching (AKC) mechanism to locally authenticate with smaller signaling cost and delay. We also propose an automatic K-selection mechanism to select a proper cache size with minimal bandwidth consumption. The random walk model and simulation experiments are conducted to investigate the performance of the AKC and the K-selection mechanisms. In the third case, we apply the random walk to the analytical model to investigate the performance of the Heterogeneous network Integration Support Node (HISN), which was proposed in our previous work to integrate heterogeneous networks in the application level. The major tasks of an HISN include the content format translation for original users and data tunneling for handoff users. The service area of an HISN may affect the number of original and handoff users, and content format translation and data tunneling costs are affected as well. We conduct random walk model to address the scaling concern of the HISN network.Acknowledgement i Chinese Abstract ii English Abstract iii Contents v List of Figures viii List of Tables xi 1 Introduction 1 2 Random Walk Model 5 3 Minislot Allocation in 802.16 Mesh Mode 10 3.1 Notations 10 3.2 Background 12 3.3 The Partition and CDC Schemes 15 3.4 The RR and Greedy Algorithms 23 3.5 Performance Evaluation 26 3.6 Summary 41 4 Performance Modeling and Evaluation on Key Caching for Mobile IP AAA 42 4.1 Notations 42 4.2 Background 45 4.3 Mobile IP AAA Framework 47 4.4 The AKC Algorithm 53 4.5 The Analytic Model for AKC with Fixed Cache Size 56 4.6 Performance of AKC with Fixed Cache Size 60 4.7 The Automatic K-Selection Mechanism 66 4.8 Summary 69 5 A Performance Study for Application-level Integration of Heterogeneous Wireless Networks 72 5.1 Background 72 5.2 The HISN Network 75 5.3 The Analytical Model 80 5.3.1 Simulation Validation 87 5.4 Performance Evaluation 87 5.5 Summary 95 6 Conclusions and Future Works 97 6.1 Key Management in MBMS 100 6.2 Modeling the Sleep Mode Operation in 802.16e 101 6.3 Modeling Link Layer Protocols for Intelligent Transportation Systems 102 A State Transitions for the Centralized Scheduling Markov Process in Partition RR and CDC RR in 802.16 Mesh Mode 103 B Software Architecture of HISN 107 Bibliography 110 Curriculum Vita 119 Publication Lists 12