Abstract

Long distance communication links may severely affect the cyber-physical systems (CPSs) in 5G (and future 6G) networks and degrade its reliability and resilience by disrupting the quality index of network latency. Further, centralized network architectures have low fault tolerance and are prone to security threats. Virtualized software defined network (vSDN)-enabled 5G networks closely monitor these facts and redefine the existing network topology to find potential locations for deploying controller and hypervisor instances. In this article, we propose an approach of dynamically deploying controller-hypervisor (C-H) pair(s) to provide a variety of network functions like differentiation between control and data signals, various translation functions, etc., with ultra low latency (ULL). The system model deals with real network topology and four well-defined network latency matrices with a mixed integer linear programming model to optimize latency objectives. A reverse path-flow mechanism (RPFM) has been proposed to provide feasible solutions by keeping the network load, and controller capacity under a tolerance limit. We have further minimized the H-plane load by distributing the network resources based on the arrival time of SERVICE_IN requests from the users. Simulation results show that our proposed technique achieves significant reduction in latency and an evolved-ULL (e-ULL) experience, where all real-time user demands are handled efficiently. The proposed approach can also be used for similar critical localization problems like service chain mapping in 5G-NR, baseband unit deployment in 5G C-RAN and firewall deployment in distributed CPS.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call