AN INTRUSION DETECTION SYSTEM BASED ON TYPE-2 FUZZY NEURAL NETWORKS

Denial of Service (DoS) is one of the most popular attacks in networks. The goal behind this kind of attacks is to make network resources unavailable to legitimate users. Therefore, these resources must be protected against the DoS attacks. Indeed, intrusion detection is an important research topic in computer network security. There are various approaches to detect this attack. In this paper, we use existing soft computing techniques such as fuzzy logic and neural network for detection of DoS attack. Neural networks, type-1 and type-2 fuzzy logic systems are important methods in real-life applications. Recent researches show that the hybrid neuro-fuzzy systems can be very effective for a wide number of problems. In this paper, we compare Adaptive Neuro-Fuzzy Inference System (ANFIS) and Interval type-2 fuzzy neural networks (IT2FNN) as a classifier for our research. The simulation results show that IT2FNN achieves high detection accuracy