多层实时网络加密数据流频繁项集挖掘方法  

Mining method for frequent item sets of encrypted data streamin multi-layer real-time network

作　　者：蔡中民 CAI Zhong-min(Information Engineering College,Henan University of Animal Husbandry and Economy,Zhengzhou 450044,China)

机构地区：[1]河南牧业经济学院信息工程学院,郑州450044

出　　处：《沈阳工业大学学报》2021年第3期301-306,共6页Journal of Shenyang University of Technology

基　　金：河南省科技厅公关项目(172102310554).

摘　　要：针对多层实时网络加密数据流频繁项集常受码间干扰,现有挖掘方法缺少对干扰的抑制而导致挖掘输出效果不好、传输误码率偏高、滤波效果差的问题,提出一种基于集对分析的挖掘方法.构建频繁项集的传输信道模型,对频繁项集的输出进行聚簇性设计和跟踪识别,根据空频结构在簇首节点完成集成处理;在近场源中提取频繁项集的平均集对特征量,通过自适应滤波器进行码间干扰抑制;对经干扰抑制的加密大数据流频繁项集进行集对分析,提取频繁项集的平均集对特征量,优化挖掘函数.结果表明,该方法的抗干扰能力强,传输误码率较低,滤波效果好.In order to solve the problem that the frequent item sets of encrypted data stream in multi-layer real-time network are often subjected to inter-symbol interference,while the existing mining methods lack the interference suppression,resulting in poor mining output effect,high transmission bit error rate(BER)and poor filtering effect,a mining method based on set pair analysis was proposed.A transmission channel model for the frequent item sets was constructed,the clustering design and tracking recognition for the output of frequent item sets were carried out,and the integration treatment of frequent item sets was completed at the cluster head nodes according to space-frequency structure.The feature quantity of average set pairs of frequent item sets was extracted from near-field sources,and the inter-symbol interference was suppressed by an adaptive filter.The set pair analysis was carried out for the frequent item sets of encrypted large data streams with suppressed interference,the feature quantity of average set pairs of frequent item sets was extracted,and the mining function was optimized.The results show that the as-proposed method has strong anti-interference ability,low transmission BER and good filtering effect.

关 键 词：集对分析 多层实时网络 加密数据流 频繁项集 挖掘 干扰滤波 关联规则 最小支持度 

分 类 号：TN911.7[电子电信—通信与信息系统]