基于双约束滑动时间窗口的告警预处理方法研究

摘要

在通信网告警相关性分析中,针对传统时间窗口提取告警数据效率低的问题,提出了一种基于双约束滑动时间窗口的告警预处理方法.在双约束的条件下,滑动时间窗的窗口宽度和滑动步长能够根据告警序列的实际情况自动地调整,并选取时间段的中点作为参照点.实验证明,采用双约束滑动时间窗划分算法可以获得比传统方法更高的数据提取效率,并且能够有效去除噪声数据,非常适合于通信网告警数据的预处理.%In communication alarm correlation analysis, traditional time window has low efficiency on extracting data. This paper first proposed a novel double constraint based sliding window method. With double constraint condition, the window width and sliding step of the time window could automatically adjust according to the actual situation of alarm sequences, and the mid-point of the time period was selected as a reference point. The performance testing of the algorithm indicates that this method has higher extract efficiency than traditional methods, which is suitable for alarm data in telecommunication networks.