Denial of Service (DoS) Attack Detection Using Feed Forward Neural Network in Cloud Environment
Ponugoti Kalpana, P. Srilatha, Gudepu Sai Krishna, Ahmad Alkhayyat, Debarshi Mazumder
Abstract
In the recent years, the Denial of Service (DoS) developed as a big security threats for networks. Preventing as well as identifying DoS attacks is a big task within the industries. There are IoT of existing methods are there to identify DoS attacks and minimize the damage. However, many of these methods do not efficiently distinguish noise from a signal. This research focuses on addressing the DoS problem and introduces a method called Feed forward Neural Network (FNN). First, the KDD cup 99 data set goes through a preprocessing step where data normalization is processed using Min-Max normalization which standardizes the data points values. This normalized data is then passed to the whale optimization method which selects the relevant features making the classification system easier. The selected features are then used by an FNN classifier to differentiate between normal and attacked data. The proposed method is implemented and simulated on MATLAB and tested experimentally resulting to a detection accuracy of 0.99. The proposed FNN classifier is compared with the existing classifiers such as Auto-encoder and LSTM where it outperformed in detecting DoS attacks efficiently.