Using Hybrid Generative Adversarial Network with Cognitive Routing Cryptography to Improve Network Security and Quality of Service

Abstract

Data communications can assist multimedia applications with complicated QoS (Quality of Service) specifications. Different varieties of networks, such as wired and wireless are to be had to ensure the quality of multimedia applications. These networks demonstrate the exceptional QoS characteristics and heterogeneity, in addition to varying QoS parameters including bandwidth, delay, and jitter. Several network design properties can result in congestion in networks with unregulated bandwidth. This paper primarily aims to explore network scalability and security concerning precision, end-to-end delay, scalability, accuracy, and throughput. To achieve this, a novel machine learning method called Hybrid General Adversarial Network-Cybersecurity Risk Profile (HGAN-CRP) has been implemented along routing protocols and included with authentication structures. In this method, various packet sizes were utilized to assess the performance of network scalability and safety. The results demonstrate improved network scalability and protection when compared to existing algorithms such as Self-Organizing Maps (SOM) and Double P-value of Transductive Confidence Machines for K-Nearest Neighbors (DPTCM-KNN). Besides, to enhance security, an authenticated cryptographic intrusion detection mechanism has been implemented.