Name
Abstract | ||
|---|---|---|
Low-rate denial of service (LDoS) attack sends high-intensity burst data streams to victims, in order to reduce TCP traffic and cut down on network service capabilities. Since the LDoS attack is highly concealed, traditional DoS detection methods are inappropriate for detecting LDoS attacks. Although many detection methods for LDoS attacks have been proposed, these methods have the disadvantages of low efficiency, high overhead, and weak real-time performance. In view of the above problems, in this paper, a method based on an improved logistic regression model for detecting LDoS attacks is proposed. Based on the fact that the TCP traffic under the LDoS attack is lower than the normal average value and its distribution is more discrete, this method uses the network traffic to extract the eigenvalues such as average TCP, variance and sample entropy as the basis to classify the traffic data. In order to make the obtained attack-detection model more accurate and real-time, an improved logistic regression algorithm is used for data training. Finally, according to the obtained classifier, the regression analysis method is used to detect whether there is abnormal traffic, so as to determine whether LDoS attack has occurred in the network. Experiments on NS-2 and test-bed show that the method in this paper can detect LDoS attacks effectively and in real time with high accuracy, low false negative rate and false positive rate. Also, its complexity is reduced. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1109/HPCC/SmartCity/DSS.2019.00076 | 2019 IEEE 21st International Conference on High Performance Computing and Communications; IEEE 17th International Conference on Smart City; IEEE 5th International Conference on Data Science and Systems (HPCC/SmartCity/DSS) |
Keywords | Field | DocType |
low rate denial of service,attack detection,improved logistic regression,classifier | Network service,Data mining,False positive rate,Data stream mining,Sample entropy,Denial-of-service attack,Computer science,Regression analysis,Real-time computing,Classifier (linguistics),Logistic regression | Conference |
ISBN | Citations | PageRank |
978-1-7281-2059-1 | 0 | 0.34 |
References | Authors | |
0 | 6 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yudong Yan | 1 | 0 | 0.68 |
| Dan Tang | 2 | 13 | 6.67 |
| Sijia Zhan | 3 | 1 | 1.71 |
| Rui Dai | 4 | 49 | 14.71 |
| Jingwen Chen | 5 | 0 | 0.34 |
| Ningbo Zhu | 6 | 0 | 0.34 |