An Attack-Feedback-Based Approach for Verifying the Success of Intrusion Attempts

It is well-known that current Intrusion Detection Systems produce large numbers of false alerts. Those low quality alerts make it very hard for administrators to understand and take appropriate actions. To deal with false positive, in this paper, an attack-feedback-based approach is introduced to verify the success of attacks. This method processes each packet as soon as it is received. When a suspect packet is indicative of an attack on an existing network service, the effects of that packet on the host will be further tracked by following the causal dependencies. The experimental results have shown that the proposed technique is highly effective in reducing the alert volume and verifying the success of intrusion attempts.