论文标题
充分利用零星的反馈：在物联网中进行数据恢复的低复杂性应用程序层编码
Making the Most of Sporadic Feedback: Low-Complexity Application-Layer Coding for Data Recovery in the Internet of Things
论文作者
论文摘要
我们建议在物联网中以延迟敏感的上行链路（传感器到门户）通信恢复丢失的数据。拟议的计划的显着特征是基于结合了重传和正向擦除校正的方法，包括低计算复杂性以及利用零星接收器反馈以进行有效的数据恢复的能力。通过保持尽可能低的编码传输数量并设计一种机制来计算O（1）中编码的数据包的最佳程度来实现降低的复杂性。我们的主要贡献是：（a）对称为窗口编码的现有方案的增强，该方案的复杂性大大降低，并且通过我们提出的方法提高了数据恢复性能。 （b）将窗口编码元素与新的反馈结构相结合的技术，以进一步降低编码复杂性并改善数据恢复。 （c）一种编码的转发方案，其中继电器节点通过听到源到目的地通信并根据听到的信息做出转发决策，从而为数据包丢失提供了进一步的弹性。
We propose application-layer coding schemes to recover lost data in delay-sensitive uplink (sensor-to-gateway) communications in the Internet of Things. Built on an approach that combines retransmissions and forward erasure correction, the proposed schemes' salient features include low computational complexity and the ability to exploit sporadic receiver feedback for efficient data recovery. Reduced complexity is achieved by keeping the number of coded transmissions as low as possible and by devising a mechanism to compute the optimal degree of a coded packet in O(1). Our major contributions are: (a) An enhancement to an existing scheme called windowed coding, whose complexity is greatly reduced and data recovery performance is improved by our proposed approach. (b) A technique that combines elements of windowed coding with a new feedback structure to further reduce the coding complexity and improve data recovery. (c) A coded forwarding scheme in which a relay node provides further resilience against packet loss by overhearing source-to-destination communications and making forwarding decisions based on overheard information.
