一种编码感知路由低时延数据传输算法

Low-delay data transmission algorithm for coding-aware routing

当前的编码感知路由算法在数据包编码时采用基于机会的网络编码策略,不会推迟数据包的转发来等待未来的编码机会,这样会降低网络编码对时延的贡献.为克服以上问题,提出了一种基于缓存管理的编码感知路由低时延数据传输算法.在编码节点,该算法采用基于队列长度的数据包决策策略来替代现有编码感知路由算法中的基于机会的网络编码策略.该算法在数据传输阶段之前引入了网络时延训练阶段,使编码节点获得了基于队列长度策略的最优阈值.仿真结果表明,在网络拥塞的情况下,此算法比传统的基于机会的网络编码策略具有更低的数据包传递时延和数据包丢失率,并且具有更高的吞吐量.

It is significant to reduce packet delivery delay for real-time applications in a wireless network. Existing coding aware routing algorithms use the opportunistic coding scheme in the packet coding algorithm. They never delay packets to wait for the arrival of a future coding opportunity which results in the degradation of the contribution of network coding to delay performance. To overcome the above problem, for coding-aware routing, this paper presents a low-delay data transmission algorithm based buffer management. In the coding node, this algorithm decides packets according to the queue-length based threshold policy instead of the regular opportunistic coding policy as used in existing coding-aware routing algorithms. This algorithm introduces the network delay training phase before the data transmission phase to make the coding node obtain the optimal threshold for the queue-length based threshold policy. Simulation results show that our algorithm can achieve a lower packet delivery delay, a lower packet loss ratio and a higher throughput than the traditional opportunistic coding policy in network congestion.