Enhanced Security of Software-defined Network and Network Slice Through Hybrid Quantum Key Distribution Protocol

Abstract

Software-defined networking (SDN) has revolutionized the world of technology as networks have become more flexible, dynamic and programmable. The ability to conduct network slicing in 5G networks is one of the most crucial features of SDN implementation. Although network programming provides new security solutions of traditional networks, SDN and network slicing also have security issues, an important one being the weaknesses related to openflow channel between the data plane and controller as the network can be attacked via the openflow channel and exploit communications with the control plane. Our work proposes a solution to provide adequate security for openflow messages through using a hybrid key consisting of classical and quantum key distribution protocols to provide double security depending on the computational complexity and physical properties of quantum. To achieve this goal, the hybrid key used with transport layer security protocol to provide confidentiality, integrity and quantum authentication to secure openflow channel. We experimentally based on the SDN-testbed and network slicing to show the workflow of exchanging quantum and classical keys between the control plane and data plane and our results showed the effectiveness of the hybrid key to enhance the security of the transport layer security protocol. Thereby achieving adequate security for openflow channel against classical and quantum computer attacks.

Performance Analysis of Big Model Transmission under Double Rayleigh Fading

“…It began with the inception of neural networks, where early attempts to simulate the human brain's interconnected neurons laid the foundation. However, it was the advent of deep neural networks that truly revolutionized the landscape [4][5][6]. With the introduction of deep learning, multiple layers of interconnected neurons enabled the extraction of intricate features from complex data, paving the way for more accurate and powerful models [7][8][9].…”