基于二维图像特征的网络时间隐通道检测方法

Detection Method of Covert Timing Channels Based on Two-dimensional Image Features

网络时间隐通道以违反安全策略的形式传递信息且以难以检测,为了减少不必要的安全隐患,网络时间隐通道的检测已成为网络安全领域亟需解决的问题。针对现有检测方法仅反映一维特征的缺陷,提出一种基于二维图像特征的网络时间隐通道的检测方法,通过将网络流隐蔽通道在时间轴上的特性关系反映到二维图像的纹理特性上,进而通过基于灰度共生矩阵的纹理特征描述方法实现网络时间隐通道检测。研究结果表明该方法可以显著提高网络隐通道的检测效率和准确率。

Information is transmitted in a way to disobey security strategy in covert timing channels,in order to reduce unnecessary security risks,the detection of covert timing channel has become the main focus remaining to be solved. Since the novel approach can only reflect the flaw in one-dimensional features,a detection approach of covert timing channels based on two-dimensional image features is proposed. By reflecting the feature relationship of network flow covert channel on the time axis to the texture features of two-dimensional image,the network covert timing channel detection is realized by the texture feature description method based on the gray level co-occurrence matrix. The results show that this method can improve the detection efficiency and accuracy of covert timing channel significantly.