随机网络中分布式节点高精度自定位算法

发布时间：2018-04-24 19:02

  本文选题：随机网络 + 分布式　； 参考：《科学技术与工程》2017年26期

【摘要】：随机网络不确定性高,较为复杂,当前节点定位算法无法准确对随机网络中分布式节点进行准确定位,且适应性差。为此,提出一种新的基于标记传递的随机网络中分布式节点高精度自定位算法,标记传递算法将随机网络中的分布式节点用有标记数据和无标记数据进行描述,依据节点间的相似度将节点标记传递至其相邻节点。对节点定位问题进行描述,介绍了标记传递算法,在此基础上通过标记传递算法对随机网络中分布式节点进行高精度自定位,定位过程主要分为离线训练阶段与在线自定位阶段,给出定位详细实现过程,对定位结果进行滤波处理。实验结果表明,所用所提算法对随机网络中分布式节点进行自定位,定位精度高,适应性强。
[Abstract]:Because of the high uncertainty and complexity of stochastic networks, the current node localization algorithms can not accurately locate distributed nodes in random networks, and the adaptability is poor. In this paper, a new high-precision self-localization algorithm for distributed nodes in random networks based on label transfer is proposed, which describes the distributed nodes in random networks with labeled data and unmarked data. According to the similarity between nodes, the node markers are passed to its adjacent nodes. In this paper, the problem of node localization is described, and the label transfer algorithm is introduced. The localization process is mainly divided into off-line training stage and on-line self-positioning stage. The detailed realization process of location is given, and the location results are filtered. The experimental results show that the proposed algorithm can locate the distributed nodes in random networks with high accuracy and adaptability.
【作者单位】： 郑州工业应用技术学院机电工程学院;
【基金】：河南省科技厅重点科研项目(162102210315)资助
【分类号】：O157.5
，

本文编号：1797839