摘要
针对广播电视业务系统中的疑似攻击快速检测难题,利用大数据分析思路,采用先怀疑后识别的技术方案,首先利用聚类算法对疑似攻击下的系统数据进行模糊数据分离,组建对分析有价值的疑似攻击数据集合;然后利用这个数据集合,进行基于模糊数据分离的联合评分偏离度判别来分离疑似攻击行为。本文给出了系统的总体设计框架和功能结构,原型系统运行结果表明该方法无需对数据进行复杂的预处理且具有较好的检测性能,能够准确分析疑似攻击行为,具有较高的精确度和较强的适用性。
Aiming at the rapid detection of suspected attacks in TV broadcasting service, through big data analysis and the technical scheme of doubt first and identification thereafter, this paper first uses clustering algorithm to carry out a fuzzy data separation of system data under suspected attacks, and composes a data set of suspected attacks that is valuable for analysis; and then uses the data set to separate the suspected attacking behaviors by carrying out the joint score deviation judgment based on the fuzzy data separation. The overall design framework and functional structure of the system are provided, and the results show that this method does not need a complicated data preprocessing and has a good detection performance. It accurately analyzes suspected attacking behaviors, and is relatively high precise and strong applicable.
作者
王欣刚
Wang Xingang(Administrative Center for the DTH Service, SAPPRFT, Beijing, 100866, China)
出处
《广播与电视技术》
2018年第6期123-126,共4页
Radio & TV Broadcast Engineering
关键词
广播电视
疑似攻击
检测方案
大数据
Radio ＆ TV
Suspected attack
Detective scheme
Big data
