Convolutional Neural Network for DDoS Detection

A DoS attack is a sort of cyber assault in which an attacker attempts to overload or disable a network, server, or application by flooding it with traffic or overloading it with requests. In addition, the implementation of intrusion detection systems based on convolutional neural network help to detect and respond to attacks in real time before it is too late by knowing the importance of critical services and infrastructures present in a smart city. A CNN may detect malign traffic patterns and warn by monitoring the frequency and amount of incoming traffic. Therefore, this paper implements a denial of service attack detection system under the architecture of a convolutional neural network that demonstrates the effectiveness of detection of the Dos attacks with an accuracy of 98.39% f1 score of 98.43%, and Precision of 98.39%, which is very important to provide security to critical infrastructures in smart cities.