ABAP: Anchor Node Based DDoS Attack Detection Using Adaptive Neuro-Fuzzy Inference System

Wireless Sensor Network is utilized in many applications so it is prone to various attacks. Among which the flooding attacks interrupt the communication between the sensor nodes while exchanging information. It is the most common type of Distributed Denial of Service, it overloads the legitimate nodes with huge amounts of traffic and makes it slow down. Therefore, it is necessary to protect the wireless sensor networks from the flooding attacks. In this paper, two approaches have been used: Fuzzy Inference System and Adaptive Neuro-Fuzzy Inference System (ANFIS) based on detection of flooding attack. In which two parameters namely, Energy Consumption of Node and Packet Transfer Rate were used to detect the presence of flooding attack. Moreover, Cluster Head also elected very effectively by using the Fuzzy method by calculating three metrics—mobility factor, residual energy and trust factor. ANFIS method merges Fuzzy logic and learning capability of neural networks, so it detects flooding attacks more efficiently. The proposed method is also compared with the existing methods. The simulation studies show that the proposed system detects the flooding attack very effectively and improves the throughput and lifetime of the sensor nodes.