摘要
通过检测代表性能降质的异常点来实现故障的提前发现和快速恢复是提高通信网的可靠性的重要手段.采用基于统计假设检验的网络异常点检测方法,提出一种综合运用季节累积自回归滑动平均模型时间序列预测和置信区间计算来动态获取性能指标阈值的方法.利用累积自回归滑动平均模型在训练集上的拟合残差白噪声符合正态分布的假设,给出了一种通过构造满足t分布的随机变量来计算预测值在任意置信度1-α下置信区间的新算法.理论分析和实验结果表明,该阈值动态确定方法有效.
Fault correction by detecting anomalies designating performance degradation is an important approach to improving the reliability of communication network. Statistical hypotheses testing approach is employed to detect network anomaly. A new approach to acquiring the fluctuation threshold is proposed comprehensively when taking advantage of time series prediction confidence interval computation based on multiplieative autoregressive integrated moving average. Furthermore, under the assumption that the training residual which is a white noise process follows normal distribution, the associated confidence interval of predietion can be figured out under any given confidence degree by constructing random variable satisfying t distribution. Experiments verify the effectiveness of anomaly detection mechanism and the accuracy of the algorithm.
出处
《北京邮电大学学报》
EI
CAS
CSCD
北大核心
2011年第2期45-49,共5页
Journal of Beijing University of Posts and Telecommunications
基金
国家"十一五"科技支撑项目(2008BAH24B04)
国家自然科学基金项目(61072060)
中央高校基本科研业务费专项资金项目
关键词
异常点检测
时间序列预测
季节累积自回归滑动平均模型
置信区间
anomaly detection
time series prediction
seasonal autorgressive integrated moving average
confidence interval
