采用符号动力学方法检测低速率拒绝服务攻击

Low-rate-denial-of-service attack detection by symbolic dynamics method

针对低速率拒绝服务攻击引起的网络安全问题,提出采用符号动力学方法对其进行检测,通过将采样得到的数据包数量信号转化为对应的符号序列,并计算符号序列的熵值,发现低速率拒绝服务攻击具有令数据包数量信号信息复杂度激增的特点,从而能够结合预先设定的熵阈值进行比较识别.分别在仿真环境和实际测试服务器上进行了算法验证,实验结果表明,文中所提算法对低速率拒绝服务攻击辨识平均准确度超过92%.

In terms of the network security problem caused by Low-rate-Denial-of-Service attack, this paper proposes a symbolic dynamics method to detect such attacks. The number of packets sampled signals is transferred into the symbol sequence and the calculate the entropy corresponding to the symbol sequence. There is a significant increase in packets number signal information complexity caused by Low-rate-Denial- of-Service attack arrival, which can be compared with a default entropy threshold for recognition. We verify the algorithm in the simulation environment and the actual test servers respectively and the experimental results show that the proposed algorithm has an average accuracy of 92% to achieve the Low-rate-Denial-of- Service attack recognition.