How to combine multiple data-streams for transmission in aggregate is a very interesting problem, especially for the emerging software-defined networks nowadays. The conventional packet-based protocols cannot provide the flexibility for combining data-streams in the ad hoc nature. If the number of data-streams changes over time, the existing packet formats cannot handle the transmission of multiple data-streams effectively. The effectiveness is measured by two performance metrics, namely coding efficiency and data-transmission intermittency. We propose a new software-defined multiplexing code (SDMC) approach, which can combine (multiplex) multiple data-streams easily and is much more effective than the conventional packet-based method. Three SDMC schemes (distributed, hierarchical, and hybrid) are compared theoretically and by simulation. A trade-off between these two performance metrics can be found when one selects one of the three SDMC schemes for combining multiple data-streams. The hierarchical SDMC scheme brings about the highest coding efficiency while the hybrid SDMC scheme suffers from the smallest overall intermittency.

中文翻译：

---

各种软件定义的复用码的理论分析

如何组合多个数据流以进行汇总传输是一个非常有趣的问题，尤其对于当今新兴的软件定义网络而言。常规的基于分组的协议不能提供自组织性质的用于组合数据流的灵活性。如果数据流的数量随时间变化，则现有的数据包格式将无法有效地处理多个数据流的传输。有效性通过两个性能指标来衡量，即编码效率和数据传输的间歇性。我们提出了一种新的软件定义的多路复用代码（SDMC）方法，该方法可以轻松地组合（多路复用）多个数据流，并且比传统的基于数据包的方法有效得多。从理论上和通过仿真比较了三种SDMC方案（分布式，分层和混合）。这两种性能度量之间的权衡，可以发现，当一个选择三个SDMC方案用于组合多个数据流中的一个。分层SDMC方案带来了最高的编码效率，而混合SDMC方案则遭受了最小的整体间歇性。