LAUSR

Security issue in cloud environment is one of the major obstacle in cloud implementation. Network attacks make use of the vulnerability in the network and the protocol to damage the data and application. Cloud follows distributed technology; hence it is vulnerable for intrusions by malicious entities. Intrusion detection systems (IDS) has become a basic component in network protection infrastructure and a necessary method to defend systems from various attacks. Distributed denial of service (DDoS) attacks are a great problem for a user of computers linked to the Internet. Data mining techniques are widely used in IDS to identify attacks using the network traffic. This paper presents and evaluates a Radial basis function neural network (RBF-NN) detector to identify DDoS attacks. Many of the training algorithms for RBF-NNs start with a predetermined structure of the network that is selected either by means of a priori knowledge or depending on prior experience. The resultant network is frequently inadequate or needlessly intricate and a suitable network structure could be configured only by trial and error method. This paper proposes Bat algorithm (BA) to configure RBF-NN automatically. Simulation results demonstrate the effectiveness of the proposed method.