江苏省科技期刊可视化阅读平台

导航条

《无线互联科技》杂志社 ›› 2025, Vol. 22 ›› Issue (1): 15-18.

• 电子通信 • 上一篇    下一篇

基于Merkle哈希树的通信网络异常数据流自动识别方法

孙树垚   

  1. 国家广播电视总局信息中心,北京 100866
  • 出版日期:2025-01-10 发布日期:2025-03-21
  • 作者简介:孙树垚(1992— ),男,工程师,硕士;研究方向:网络安全,政务信息化。

Automatic identification method for abnormal data flow in communication networks based on Merkle hash tree

SUN Shuyao   

  1. Information Center of National Radio and Television Administration, Beijing 100866,China
  • Online:2025-01-10 Published:2025-03-21

PDF

220

摘要： 为解决通信网络在高异常数据比例下识别模型性能退化、F1值偏低的问题,文章设计了一种基于Merkle哈希树的通信网络异常数据流自动识别方法。对通信网络数据流进行预处理,以消除噪声和冗余信息;通过设定合理的阈值对异常评分进行比较,从而提取出关键的异常数据流特征;利用Merkle哈希树,建立一个高效的异常数据识别模型,该模型通过哈希树节点的连接机制,在层级间有效传递信息,实现对通信网络中异常数据流的自动识别。实验结果表明,相较于传统方法,基于Merkle哈希树的通信网络异常数据流自动识别方法在各个数据集上的表现更为优越,特别是在数据集4中,取得了0.93的F1值,证明该方法识别通信网络异常数据流的准确性和可靠性,几乎无遗漏。

关键词: Merkle哈希树, 通信网络, 哈希树节点, 异常数据流, 自动识别

Abstract: In order to solve the problem that the performance of the identification model of communication network is degraded and the F1 value is low under the high proportion of abnormal data, an automatic identification method of abnormal data flow in communication network based on Merkle hash tree is designed. The data stream of communication network is preprocessed to eliminate noise and redundant information. By setting a reasonable threshold to compare the abnormal scores, the key abnormal data flow characteristics are extracted. Using Merkle hash tree, an efficient abnormal data identification model is established. The model effectively transmits information between levels through the connection mechanism of hash tree nodes, and realizes the automatic identification of abnormal data flow in the communication network. The experimental results show that, compared with the traditional methods, the Merkle hash tree based method for automatic identification of abnormal data flow in communication network performs better on each data set, especially in data set 4, which achieves an F1 value as high as 0.93, which proves the accuracy and reliability of this method for identification of abnormal data flow in communication network, with almost no omission.

Key words: Merkle hash tree, communication network, hash tree nodes, abnormal data flow, automatic recognition

中图分类号: