Efficient multicast design for wireless mesh networks.
详细信息
文摘
Wireless mesh networks WMN) have emerged as an efficient means to expand the wireless reach of metro broadband deployments at a variety of locations or scenarios. It provides high quality service to end users as the last-mile of the Internet. Multicast provides an efficient mechanism for distributing data among a group of nodes, such as online games and video conferences. With the increasing popularity of content distribution and multimedia applications, efficient multicast communication becomes essential for the wide deployment of WMNs. In this dissertation, we discuss several research topics related with multicast communication in WMNs. First, we investigate the problem of routing and channel assignment for multicast communication in link-homogeneous WMNs with the goal of maximizing throughput. In this dissertation, we consider WMNs equipped with multiple channels and multiple interfaces. Previous research work on multicast does not take the multi-channel characteristic into consideration, thus it cannot fully explore the network capacity of WMNs. By exploiting multi-channel and multi-interface, our proposed approach has two major steps: i) it builds an efficient multicast tree that minimizes the number of relay nodes； and ii) the dedicated channel assignment strategies are designed to reduce the interference to improve the network capacity. We demonstrate that our proposed protocols not only improve the throughput, but also reduce the delay. Second, we study the multicast problem in link-heterogeneous WMNs. Unlike previous work that focuses on link-homogeneous WMNs only, we consider one important form of link heterogeneity: different link loss ratios. Under this constraint, although minimizing relay nodes helps to decrease interference in the WMN, it is also important to choose high quality links to minimize the number of transmissions. This is because decreasing the number of transmissions helps to increase the throughput. Based on this consideration, we define a new graph theory problem: HW-SCDS to model link-heterogeneity. Maximizing WMN throughput is equivalent to computing a minimum HW-SCDS MHW-SCDS) in the graph. We prove that computing an MHW-SCDS is NP-hard and devise a greedy algorithm for it. We show that our approach outperforms previous work in terms of throughput and delay. Third, we investigate the problem of opportunistic multicast in WMNs. By exploiting the broadcast nature and spatial diversity of the wireless medium, opportunistic routing has emerged as a new routing paradigm to improve unicast throughput. However, its natural multicast extension does not build any efficient multicast structure, thus the explosion of unnecessary retransmission is unavoidable. To overcome this shortcoming, we propose a new opportunistic multicast protocol to improve throughput in WMNs. The key concept is a tree backbone in this protocol. Our tree backbone protocol represents a tradeoff between traditional structured multicast protocols and unstructured protocols. Therefore, our solution is able to improve throughput by both utilizing spatial diversity and reducing transmissions.