Work

Perceptron Based Deep Learning Technique to Enhance Quality of Service (QoS) and Security in Software Defined Network

Year: 2024
Type: article
Abstract: As network technology are always being improved, the Internet economy is quickly growing. Consequently, it is critical to pay attention to the reliability and safety of the network services offered by the ISP. A unified monitoring and control mechanism is available with state-of-the-art technologies such as Software-defined Network (SDN), however the SDN controller receives too much data to handle network traffic maintenance independently. Through the use of software-defined networking (SDN), networks are able to continuously monitor traffic, detect threats, adjust security policies, and include security services. Threats like man-in-the-middle attacks, DoS attacks, and saturation attacks are brought about by the SDN. So, the centralised controller can employ modern methods, like AI, to govern the flow of data across the network. Managing network congestion and detecting distributed denial-of-service (DDoS) assaults are the main concerns of this article. This study uses the Multilayer Perceptron (MLP) to detect DDoS attacks and connection congestion through packet loss using data acquired from the Open Flow Switch Table. Simulation results show that the proposed methodology out performs the status quo in terms of network performance.
Language: English
Open Access status: diamond
OpenAlex ID: W7217588042
FWCI: 0.107
Cites: 23
Cited by: 1
Related to: 10