基于网络编码的多跳无线网络可靠组播
doi: 10.3724/SP.J.1146.2012.00364
An Efficient Reliable Multicast Framework Based on Network Coding in Multi-hop Wireless Networks
-
摘要: 多跳无线网络中实现可靠组播面临许多挑战,数据丢失恢复是其中的核心问题之一。该文提出一种基于域的随机线性网络编码的多跳无线网络中高效可靠组播(Network Coding Reliable Multicast, NCRM)算法,克服了XOR编码方式的局限性,将原始数据包划分成不同代(generation)进行发送,恢复节点采用随机线性网络编码方式发送编码包,发生丢包的组播组成员发送携带丢包比特向量的NACK(Negative ACKnowledgement),经过邻居恢复、多跳恢复或源端恢复,完成可靠组播过程。该文建立了节点丢失恢复过程的齐次马尔科夫链数学模型,给出理论平均时延和重传跳数。NS2仿真结果验证了理论分析模型的准确性。数值结果表明,与PGM(Pragmatic General Multicast)和CoreRM可靠组播协议相比,NCRM算法显著改善了网络吞吐量和丢失恢复延时等性能。Abstract: Providing reliable multicast service in multi-hop wireless network faces many technique challenges, among which packet loss recovery is one of most important issues. In this paper, an efficient packet loss recovery framework is proposed based on random linear network coding in field for reliable multicast in multi-hop wireless networks, called Network Coding Reliable Multicast (NCRM). NCRM overcomes the limitations of the XOR encoding scheme. The original packets are classified into different generations and sent. Repair nodes employ random linear network coding to send encoded packets. A group member who experiences packet loss sends a Negative ACKnowledgement (NACK) with loss bit vector. Through neighborhood recovery, multi-hop recovery and/or source recovery, loss recovery is efficiently fulfilled. A mathematical model of homogeneous Markov chain is constucted to analyze the theoretical delay and average hops of retransmission. Simulation experiments based on NS2 and the numerical results demonstrate that NCRM significantly outperforms Pragmatic General Multicast (PGM) and CoreRM, in terms of network throughput and loss recovery latency.
计量
- 文章访问数: 2367
- HTML全文浏览量: 128
- PDF下载量: 900
- 被引次数: 0