A novel approach based on ant colony system for IP traceback

作者：Liu, Menglin^*; Jin, Zhengping

来源：International Conference on Artificial Intelligence and Engineering Applications (AIEA), Hong Kong, PEOPLES R CHINA, 2016-11-12 To 2016-11-13.

• DDoS
• IP traceback
• Ant colony optimization (ACO)
• Louvain algorithm

Ant colony algorithm is often used to solve the IP traceback problem without the entire network routing information. However, such algorithm is easy to converge to a local suboptimal solution. Especially with the increase of network topology size, the real attacker is more difficult to find. In this paper, a novel approach based on colony system is proposed to identify the source of the DDoS attack. This approach brings in a new concept of the network topology division before tracing the attack by the ant colony algorithm. The division based on flow information can enhance the ability of the ants to search a more globally optimal solution for the attack path, even if the network topology is large-scale. The performance of the novel approach in reconstruction the attack path is evaluated through a series of ns2 simulations. The simulations results show that the proposed scheme has better performance than the conventional ACS algorithm, e.g. the performance of the novel ACO approach has been about 10% higher than the improvement ACO scheme, when the network topology is p=800.

• 出版日期2016
• 单位网络与交换技术国家重点实验室; 北京邮电大学