Multilayer Self-defense System to Protect Enterprise Cloud

Abstract

A data breach can seriously impact organizational intellectual property, resources, time, and product value. The risk of system intrusion is augmented by the intrinsic openness of commonly utilized technologies like TCP/IP protocols. As TCP relies on IP addresses, an attacker may easily trace the IP address of the organization. Given that many organizations run the risk of data breach and cyber-attacks at a certain point, a repeatable and well-developed incident response framework is critical to shield them. Enterprise cloud possesses the challenges of security, lack of transparency, trust and loss of controls. Technology eases quickens the processing of information but holds numerous risks including hacking and confidentiality problems. The risk increases when the organization outsources the cloud storage services through the vendor and suffers from security breaches and need to create security systems to prevent data networks from being compromised. The business model also leads to insecurity issues which derail its popularity. An attack mitigation system is the best solution to protect online services from emerging cyber-attacks. This research focuses on cloud computing security, cyber threats, machine learning-based attack detection, and mitigation system. The proposed SDN-based multilayer machine learning-based self-defense system effectively detects and mitigates the cyber-attack and protects cloud-based enterprise solutions. The results show the accuracy of the proposed machine learning techniques and the effectiveness of attack detection and the mitigation system.

Blockchain-Based SQKD and IDS in Edge Enabled Smart Grid Network

“…Machine learning (ML) approaches are used for intrusion detection and mitigation to protect cloud-based enterprise solutions [22]. Using machine learning algorithms to predict, and distinguish the DDoS attack from normal traffic prediction, detection, and mitigation are fast and effective.…”

Web Security: Emerging Threats and Defense

“…Similar to white box testing, black box testing can be done either manually or with tools. In [13], the authors explain a system that automatically detects and mitigates the attacks. Web applications are inefficient in checking the correctness of the data request.…”

Cyber-Attack Detection and Mitigation Using SVM for 5G Network

“…SVM has been used for a very simple but obvious reason, which is the fact that it has its development mechanism when it comes to algorithms, and its static prediction techniques lift it above most frameworks, not to mention how quickly and intelligently it can train and learn algorithms in both linear and nonlinear capacities. SVM is popular among all other techniques because it can very quickly separate the linear space from the supposedly nonlinear space [5].…”

“…The distributed denial-of-service (DDoS) detection rates are reported in [5][6][7][8] using the SVM algorithm are in the range of 93% to 98%. Therefore, the high efficiency of DDOS attacks and their ability to disrupt normal network functions is the main motivation for starting this research using SVM.…”