用于IoT的应用层编码：益处、限制和实现方面
Application Layer Coding for IoT: Benefits, Limitations, and Implementation Aspects
摘要：
未来IoT/M2M系统的关键技术之一是低功率广域网，其被设计为使用随机接入协议来支持通常在未授权共享频谱中的大量低端设备。然而，这些通常在没有集中控制的情况下操作，并且由于自动重传请求和确认机制由于严格的占空比限制和共享频带中的高干扰而不是非常有效，因此许多分组因冲突而丢失。在本文中，我们分析了最近提出的应用层编码方案，其目的是恢复丢失的数据包通过引入冗余的形式的喷泉码。我们展示了延迟和解码复杂度如何受到丢包率的影响，但也证明了通过引入更多冗余可以实现的效果是有限的。仿真结果验证了分析的正确性。
英文摘要：
One of the key technologies for future IoT/M2M systems are low power wide area networks, which are designed to support a massive number of low-end devices often in the unlicensed shared spectrum using random access protocols. However these usually operate without centralised control and since Automatic Repeat request and acknowledgement mechanisms are not very effective due to the strict duty cycles limits and high interference in the shared bands, many packets are lost from collisions. In this paper we analyse a recently proposed application layer coding scheme, which aims to recover lost packets by introducing redundancy in the form of a fountain code. We show how latency and decoding complexity is affected by the packet loss rate but also prove that there is a limit to what can be achieved by introducing more redundancy. The analysis is backed up by simulation results.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
