基于网络编码的P2P数据分发方案研究
发布时间：2018-09-17 16:46
【摘要】：现有的网络编码与P2P技术结合的数据分发方案对网络性能有所限制,若信源节点以网络的组播容量来确定数据的发送速率将会导致大量链路空闲浪费网络资源,然而源节点的发送速率过大又会使小流量节点无法获得足够解码的数据块,降低网络分发数据的成功率。针对以上问题,本文提出一种基于负反馈机制的重复网络编码数据分发方案。该数据传输方案主要涉及两个技术关键点:重复网络编码的具体方案以及建立负反馈模型对全网节点的编码次数进行调节。重复网络编码是对传统单次网络编码进行优化,根据网络具体性能在编码过程中生成多个线性无关数据块以增加下游网络可用数据量,使下游小流量宿点能收到足够解码的线性无关数据块,成功解码,提高数据分发成功率。负反馈模型以数据分发率为调节因子,调节网络中编码节点的编码次数,当数据分发率低于理想值时,超级节点选择部分编码节点增加其编码次数以提高数据分发率,当数据分发率高于理想值时,超级节点选择部分编码节点降低其编码次数以减小因过度编码造成的延迟,使网络在保持理想数据分发率的同时保持较低的传输延迟和编解码开销。本文采用NS-2模拟器仿真网络环境,通过模拟网络中数据传输过程验证该方案的有效性,通过与存储转发和其他网络编码方案的对比分析该方案的整体性能及优势,从实验的角度对重复编码方案的可行性进行验证。
[Abstract]:The existing data distribution scheme combining network coding with P2P technology limits the performance of the network. If the source node determines the data transmission rate by the multicast capacity of the network, it will lead to a large number of idle links wasting network resources. However, if the transmission rate of the source node is too large, the small traffic node will not get enough decoded data blocks, which will reduce the success rate of data distribution in the network. In order to solve the above problems, this paper proposes a data distribution scheme for repetitive network coding based on negative feedback mechanism. The data transmission scheme mainly involves two key points: the specific scheme of repetitive network coding and the establishment of a negative feedback model to adjust the coding times of the whole network node. Repetitive network coding is to optimize the traditional single network coding. According to the network performance, multiple linear independent data blocks are generated in the coding process to increase the amount of data available in the downstream network. It can receive enough decoded linear independent data blocks and improve the success rate of data distribution. The negative feedback model takes the data distribution rate as the adjustment factor and adjusts the coding times of the coding nodes in the network. When the data distribution rate is lower than the ideal value, the super node selects some coding nodes to increase the coding times to improve the data distribution rate. When the data distribution rate is higher than the ideal value, the super node selects part of the encoding node to reduce the coding times to reduce the delay caused by over-coding. It makes the network keep low transmission delay and codec overhead while keeping the ideal data distribution rate. In this paper, the NS-2 simulator is used to simulate the network environment, and the validity of the scheme is verified by simulating the data transmission process in the network. The overall performance and advantages of the scheme are analyzed by comparing with the storage, forwarding and other network coding schemes. The feasibility of the repetitive coding scheme is verified from the point of view of experiment.
【学位授予单位】：哈尔滨工程大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.02
[Abstract]:The existing data distribution scheme combining network coding with P2P technology limits the performance of the network. If the source node determines the data transmission rate by the multicast capacity of the network, it will lead to a large number of idle links wasting network resources. However, if the transmission rate of the source node is too large, the small traffic node will not get enough decoded data blocks, which will reduce the success rate of data distribution in the network. In order to solve the above problems, this paper proposes a data distribution scheme for repetitive network coding based on negative feedback mechanism. The data transmission scheme mainly involves two key points: the specific scheme of repetitive network coding and the establishment of a negative feedback model to adjust the coding times of the whole network node. Repetitive network coding is to optimize the traditional single network coding. According to the network performance, multiple linear independent data blocks are generated in the coding process to increase the amount of data available in the downstream network. It can receive enough decoded linear independent data blocks and improve the success rate of data distribution. The negative feedback model takes the data distribution rate as the adjustment factor and adjusts the coding times of the coding nodes in the network. When the data distribution rate is lower than the ideal value, the super node selects some coding nodes to increase the coding times to improve the data distribution rate. When the data distribution rate is higher than the ideal value, the super node selects part of the encoding node to reduce the coding times to reduce the delay caused by over-coding. It makes the network keep low transmission delay and codec overhead while keeping the ideal data distribution rate. In this paper, the NS-2 simulator is used to simulate the network environment, and the validity of the scheme is verified by simulating the data transmission process in the network. The overall performance and advantages of the scheme are analyzed by comparing with the storage, forwarding and other network coding schemes. The feasibility of the repetitive coding scheme is verified from the point of view of experiment.
【学位授予单位】：哈尔滨工程大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.02
【参考文献】
相关期刊论文 前10条
1 罗莉;覃团发;唐振华;蒋果生;;基于网络编码的无线传感器网络广播节能算法[J];广西大学学报(自然科学版);2011年06期
2 叶晓国;;基于NS-2的无线传感器网络仿真模块扩展方法的研究[J];计算机研究与发展;2011年S2期
3 蒲保兴;王伟平;;线性网络编码运算代价的估算与分析[J];通信学报;2011年05期
4 郝琨;金志刚;;一种最小化编码节点的网络编码优化算法[J];电子与信息学报;2011年02期
5 卢文伟;朱艺华;陈贵海;;无线传感器网络中基于线性网络编码的节能路由算法[J];电子学报;2010年10期
6 雷迎春;程实;吴产乐;龚奕利;康R，
本文编号：2246525
本文链接：https://www.wllwen.com/guanlilunwen/ydhl/2246525.html
