马尔可夫模型于无线信道异常检测中的应用

Application of markov model in wireless channel anomaly detection

无线信道异常检测中,现有基于大尺度衰落建模的能量检测法简便、迅速,然而其在检测过程中忽略了阴影衰落的实时、随机变化的特性。马尔可夫模型在无线信道建模中具有良好的应用前景,能够有效地应用于阴影衰落的动态分析。通过统计分析先验马尔可夫模型矩阵的相似度变化阈值,计算先验与实时马尔可夫模型矩阵相似度,检测阴影衰落的变化规律是否发生变化,实现无线信道环境的异常检测。该方法作为大尺度衰落建模能量检测法的补充,能够完善检测覆盖面,提高检测的准确率。多次仿真实验结果表明,在高斯白噪声入侵时,该方法可实现准确的检测。

Among wireless channel abnormal detection methods,the way of constructing large scale fading model for energy detection is easy and fast.However,this method has ignored shadow fading which has characteristics of real-time and random changing.Using Markov model as a way to analyze random process has good application prospects in wireless channel modeling.It can be effectively used to analyze the change of shadow fading.First,the threshold of the prior Markov model matrix’s similarity should be statistically analyzed.Then,the similarity between the prior and the real-time Markov model matrix is calculated.Whether the regular pattern of shadow fading has changed can be found to compare the two similarities mentioned above,then the abnormality of wireless channel environment detection has finished.Plenty of experimental results based on the simulation show that this method can achieve accurate detection for the Gaussian white noise intrusions.