An IDS based on modified chaos Elman’s neural network approaches for securing mobile ad hoc networks against DDoS attack

Abstract

Abstract A mobile ad hoc network (MANETs) is a collection of moving nodes that combine into a network with no predefined infrastructure. There are many types of attacks that could target MANETS, one among them is Distributed Denial of service attacks (DDoS). DDoS is defined as attacking routing functions and taking down the entire operation of the mobile ad hoc network. The two primary victims of DDoS attacks are the functions of routing and battery capacity. The DDoS attack can cause routing table overflow which in turn can potentially cause the infected node floods. The routing overflow is followed by creating a fake route packet to consume the available resources of the participating active nodes. This cause disrupts the normal functioning of legitimate routes. In recent years, different approaches are implemented to improve the security level of MANET. In this work, the Cuckoo Search Algorithm-based Modified Elman’s Neural Network (CSA - MENN) approaches have been proposed to overcome DDoS attacks. The CSA - MENN approaches consists of three-part which are Cuckoo search algorithm clustering area to enhance the route from source to destination, chaos theory module is used to detect the abnormal nodes, then the Modified Elman Neural Network (MENN) is employed to prevent a malicious node from sending data to the destination by determining node that consumed more resources. Packets could be lost or the victim could reset the path between the attacker and itself. CICIDS dataset has been used to test and evaluate the performance of the proposed approach based on the criteria of accuracy, packet loss, and jitter. The data set, CICIDS 2017, used in this article divides the data into 7 groups: 5 for training, 1 for validation, and 1 for generalization. In summary, approximately 71.4 percent of data is used for training and 28.6 percent for validation and generalization.