舰船通信网络受到外界环境干扰会导致通信数据产生异常，影响通信质量。针对该问题，设计基于云计算平台的舰船通信数据异常智能检测方法。在基于Hadoop的云计算平台内，利用HDFS分布式文件系统存储舰船通信数据，MapReduce计算框架针对存储的通信数据，在Map任务环节中利用Master节点搜索非工作状态下的Worker节点执行检测任务；Reduce任务环节中利用主成分分析方法提取通信数据特征，并将其作为聚类中心，利用欧几里得距离确定待检测舰船通信数据与聚类中心间的距离值，距离越小说明两者相似度越高；设定距离阈值，当距离值小于距离阈值时，即可将其定义为正常数据，相反为异常数据，由此实现异常数据检测。实验结果显示该方法可有效获取异常数据检测结果，且检测结果的AUC值达到0.91，能够有效提升舰船通信服务质量。
External environmental interference on ship communication networks can cause abnormal communication data and affect communication quality. To address this issue, a cloud computing platform based intelligent detection method for ship communication data anomalies is designed. In the Hadoop based cloud computing platform, the HDFS distributed file system is used to store ship communication data. The MapReduce computing framework targets the stored communication data and uses the Master node to search for non working Worker nodes in the Map task stage to perform detection tasks; In the Reduce task, principal component analysis is used to extract communication data features and use them as clustering centers. The Euclidean distance is used to determine the distance value between the communication data of the ship to be detected and the clustering center. The shorter the distance, the higher the similarity between the two; Set a distance threshold. When the distance value is less than the distance threshold, it can be defined as normal data and vice versa, thus achieving abnormal data detection. The experimental results show that this method can effectively obtain abnormal data detection results, and the AUC value of the detection results reaches 0.91, which can effectively improve the quality of ship communication services.
2024,46(16): 162-165 收稿日期：2024-02-15
DOI：10.3404/j.issn.1672-7649.2024.16.027
分类号：TP181
基金项目：河南省2020年重点研发与推广专项（科技攻关）项目（212102210081）
作者简介：仇丹丹（1984 – ），女，硕士，副教授，研究方向为人工智能、云计算及大数据技术
参考文献：
[1] 张艺严, 马巍, 李彬. 能量受限无人机与移动舰船通信中的轨迹优化[J]. 中国舰船研究, 2022, 17(4): 47-56.
ZHANG Yiyan, MA Wei, LI Bin. Trajectory optimization for communication between energy-constrained UAV and moving ship[J]. Chinese Journal of Ship Research, 2022, 17(4): 47-56.
[2] 王晓湘, 刘洞天, 刘南江, 等. 基于LSTM的震后通信数据异常检测分析[J]. 中国地震, 2022, 38(2): 270-279.
[3] 段雪源, 付钰, 王坤. 基于VAE-WGAN的多维时间序列异常检测方法[J]. 通信学报, 2022, 43(3): 1-13.
[4] RAMIREZ J M, DIEZ F, ROJO P, et al. Explainable machine learning for performance anomaly detection and classification in mobile networks[J]. Computer communications, 2023, 200(4): 113-131.
[5] 杨静, 吴成茂, 周流平. 基于全局-局部自注意力网络的视频异常检测方法[J]. 通信学报, 2023, 44(8): 241-250.
[6] 李红映, 张天荣. 移动无线传感网络通信异常行为识别方法研究[J]. 传感技术学报, 2022, 35(2): 240-245.
