An Intelligent Model for Botnet Attack Detection in Cloud Computing

Abstract

During the 1940s and 1950s, personal computers filled large rooms and consumed a significant amount of power. Over time, these large and costly rooms were replaced by smaller, more powerful computers. With the advent of computer systems and the internet, distributed systems also emerged as a means to aggregate computational resources to support highly complex tasks that require multiple processing sources. This is where the importance of cloud computing comes in. Cloud computing is a collection of resources and services provided over the internet. Cloud computing is offered from data centers located around the world, making it easier for users to access virtual resources over the internet. However, it faces several security concerns, with security gaps being a significant barrier to the widespread adoption of cloud computing. Among the major security concerns threatening cloud computing is botnets, which are currently considered the most dangerous threat to internet security. They have a significant impact on criminal activities such as Distributed Denial of Service (DDoS) attacks, phishing, click fraud, and more. This raises questions about how cloud computing users can ensure the security of their information. In case of exposure to these risks, how can they be detected and prevented? In this context and after studying the fundamental concepts of cloud computing and botnet activities, this thesis primarily attempts to find a solution to this problem: detecting botnet attacks in cloud computing to prevent them from harming user data and maintaining information security. We have relied on designing a deep learning model for botnet detection in cloud computing to achieve this goal.