Asma Harbi Alashjaee et al., International Journal of Advanced Trends in Computer Science and Engineering, 11(6), November - December 2022, 218 - 226 218 ABSTRACT Electronic networks and cloud servers face many security problems since they are vulnerable to DDOS attacks. As cloud servers are exposed to attacks similar to legitimate requests on servers, it turns out that they are healthy, but they carry an attack on the data on the servers. DDOS attacks not only affect networks, but also the data carried by servers. In this paper, we proposed a new scheme to discover DDOS attacks. By using a Principal Component Analysis (PCA) scheme for network state analysis on traffic packet data, we divide and segment the network for reducing the overall computation. Comparing the results to the sample entropy, we were able to detect DDOS attacks more accurately. Key words : DDOS attacks, Principal Component Analysis, Electronic networks . 1. INTRODUCTION Cloud computing technology offers an all-inclusive collection of computer resources. It is accessible anytime, anywhere via the Internet at a minimal or no cost. Numerous business owners have increased the performance of their businesses and reduced IT costs by utilizing cloud computing. While cloud computing has numerous advantages over those on-premise However, they are also susceptible to internal as well as external attacks [ 1 ]. Therefore cloud developers have to implement security measures to guard their users' personal data from cyber attacks. This paper is written for cloud service providers and developers who want to increase the security of their cloud-based offerings. The following are the contributions of this paper: • We present a source-based DDoS defense technique used to reduce DDoS assaults in both fog and cloud environments. • We employ SDN technology. It includes the DDoS defense module installed to stop network and transport-level DDoS attacks. • The proposed system provides the deep-learning (DL)-based detection technique that is able to detect DDoS-infected traffic and can keep the same packet from reaching the cloud. This work incorporates various machine learning algorithms like Random Forest, and Naïve Bayes for classification. In this paper, we look at solutions for detecting DDOS attacks and distinguishing them from legitimate requests to servers. The remaining part of the paper is as follows: Section 2 we will pave the way for understanding what DDOS attacks are, the types of attacks that can penetrate cloud servers, what cloud servers are, and what is the harm of attacking them. In the 3 section, we will clarify some previous studies. In the 4 section, we will explain the methods of machine learning and deep learning to solve the problem of detecting DDOS attacks, and then we will clarify in the 5 section the results obtained. We will conclude with the 6 section of the paper. 2. DANGERS TO DATA Data centers keep a variety of data types, and a sizable amount of that data contains sensitive information about individuals or businesses. However, due to human mistakes, programming defects, and unforeseen circumstances, this data is prone to lose, hacking, or corruption [ 2 ]. Cloud developers must apply cutting-edge encryption techniques to protect the integrity of data carried from the user to the cloud, even if it is clear that the cloud service provider cannot prevent all data dangers. [ 3]. 2.1 Cloud computing attack vectors Cloud computing attacks focus on accessing user data and hindering users from accessing cloud computing services. Both could cause significant harm to cloud users and reduce confidence in the security of cloud computing services [4 ]. Hackers typically monitor the communications between cloud users and apps when they are preparing cloud-based attacks by: - Exploiting cloud computing vulnerabilities; Obtaining users' credentials elsewhere other than the cloud After the user passwords have been compromised, use the authentic pre-access that was provided to the cloud. There are a variety of ways to hack Cloud computing platforms. Additionally, hackers are always trying to develop more complex cloud computing services [5 ]. A Efficient Method to Detect DDos Attack in Cloud Computing Asma Harbi Alashjaee 1 , Dr Randa Ahmed Jabeur 2 1 Jouf University, SA, 421204012@ju.edu.sa 2 Jouf University, SA, rjabeur@ju.edu.sa Received Date: October 15, 2022 Accepted Date: November 20, 2022 Published Date: December 06, 2022 ISSN 2278-3091 Volume 11, No.6, November - December 2022 International Journal of Advanced Trends in Computer Science and Engineering Available Online at http://www.warse.org/IJATCSE/static/pdf/file/ijatcse011162022.pdf https://doi.org/10.30534/ijatcse/2022/011162022