[关键词]
[摘要]
随着远程会议、远程教育和交互式仿真等分布式多媒体应用的兴起,组播技术受到网络研究人员的重视.而这些应用的QoS(quality of service)需求又进一步推动了QoS敏感的组播路由协议的发展.在已提出的各种QoS组播路由协议中,如何提高呼叫成功率、增强规模伸缩性、降低控制报文开销,仍然是一个有待探索的问题.提出了一个新的QoS组播路由协议,其基本思想是使路由器只存储其两层邻居节点的可达性信息以及链路的QoS状态信息,以减少路由器存储开销,提高协议的规模伸缩性(scalability).协议采用受限的泛播技术,构造了一个接受节点发起的、采用多路径技术的、分布式路由算法.描述了协议的数据结构、组播树的构造算法,并给出了模拟实验结果.分析表明,基于受限泛播技术的组播路由协议具有节点存储开销小、呼叫接收成功率高等特点.虽然该协议付出了泛播引起的额外带宽开销较大的代价,但是由于协议所需要的控制数据总量不大,加上两层存储结构在一定程度上限制了泛播通信量,因此该方案具有很好的性能.
[Key word]
[Abstract]
The emergence of distributed multimedia applications such as teleconferencing, remote education and distributed interactive simulation etc. prompts the importance of multicast services. The QoS (quality of service) requirements of these applications drive the development of QoS-aware multicast routing. Recently, many QoS-aware multicast protocols have been proposed to meet these requirements. However, few of them can achieve high success ratio, high scalability, and low control message overhead. A new QoS-ware multicast routing protocol is proposed based on bounded flooding technique. It aims at alleviating the memory overhead of routers for setting up multicast trees and improving scalability of the protocol. In this scheme, every node has a two-level forwarding table which contains information about its immediate neighbors (routers reachable in one hop) and its second-degree neighbors (neighbors of an immediate one). By the information about the second-degree neighbors, a router can forward Join_Probe messages intelligently instead of blindly flooding them. This protocol also utilizes multi-path searching to increase the probability of finding feasible branches while connecting a new node to the multicast tree. The details of the data structures in the protocol and the algorithm of building a distribution tree are described. It demonstrates the effectiveness of the proposed protocol through simulation which evaluates its performance in terms of ACAR (average call acceptance ratio) and ACMO (average control message overhead).
[中图分类号]
[基金项目]
Supported by the National High-Tech Research and Development Plan of China under Grant No.2001AA112051 (国家高技术研究发展计划(863))