Cloud Intrusion Detection Using Hybrid Convolutional Neural Networks

Abstract

Instead of storing data on a hard drive, cloud computing is seen as the best option. The Internet is used to deliver three different kinds of computing services to users all over the world. One advantage that cloud computing provides to its customers is greater access to resources and higher performance while at the same time increasing the risk of an attack. Intrusion detection systems that can handle a large volume of data packets, analyse them, and generate reports based on knowledge and behaviour analysis were developed as part of this research. As an added layer of protection, the Convolution Neural Network Algorithm is used to encrypt data during end-to-end transmission and to store it in the cloud. Intrusion detection increases the safety of data in the cloud. In this paper demonstrates the data is encrypted and decrypted using a model of an algorithm and explains how it is protected from attackers. It's important to take into account the amount of time and memory required to encrypt and decrypt large text files when evaluating the proposed system's performance. The security of the cloud has also been examined and compared to other existing encoding methods