Computing networks that are capable of supporting multimedia applications with complex QoS performance specifications are emerging. In order to ensure that the requisite QoS multimedia applications are guaranteed, there are currently various types of networks; wired and wireless, which coexist with each other. These networks have basically different QoS characteristics and the degree of heterogeneity of different QoS parameters, such as bandwidth, delay and jitter, differs considerably. There are numerous network design properties that can lead to congestion with an unregulated bandwidth network. Here consider the separate security [encryption / authentication] algorithm and its efficiency assessment for diff. Packet size and output for both ESP and QoS. The goal of this paper is to put all of this into perspective and to provide a holistic view of the need and value of building a QoS-enabled network. The network scalability and security has been enhanced by machine learning techniques with routing protocol integrated with authentication system. The QoS of the network has been improved by hybrid General Adversarial network with cognitive routing protocol. Security enhancement has been carried out by authenticated cryptographic intrusion detection system. The experimental results has been obtained based on network scalability and network security in terms of accuracy, precision, end-to- end delay, scalability, network throughput.