基于单源分段编码的无线对等感知网数据收集模型

Partial Collection Model of Triggered Data in Large-Scale Wireless Peer-to-Peer Sensing Network

摘要:

针对大规模无线对等感知网络在触发型数据场景下的部分数据收集问题，提出单源分段编码模型。利用单个源节点的多个分段之间编码而成的数据来记录源数据，提高数据可靠性的同时使其适应大规模网络下的部分数据收集。通过使用游走包进行编码操作，避免源节点过多的能量消耗。同时提出针对节点存储空间提出动态划分编码单元策略，利用邻居间的信息交换动态的调整源数据切分的编码单元大小，实现节点存储空间与收集效率的动态调整。并针对灾难场景下的数据收集提出了危险感知编码冗余量动态调整策略，通过对邻居状态的感知动态调整发送的随机游走包的个数，自适应地提升编码冗余量，提高数据恢复率。

Abstract:

This paper proposes single source segmented coding model (SSSCM) in view of the data collection problem in sub areas of a large-scale wireless peer-to-peer sensing network in a triggered data scenario. The data encoded between multiple segments of a single source node is used to record the source data, which improves the reliability of the data and makes it suitable for partial data collection under large-scale networks. We use random walk coding packets to perform encoding operations to avoid excessive energy consumption of the source node. At the same time, dynamic segmentation of coding units (DSCU) is proposed, using the information exchange to dynamically adjust the size of the coding unit to achieve a dynamic balance between node storage space and collection efficiency. In addition, for data collection in disaster scenarios, dynamic adjustment of disaster sensing coding redundancy (DADSCR) is proposed, which dynamically adjusts the number of random walk packets sent through the perception of neighbor status, adaptively increases redundancy, and further improve the data recovery rate.