Keywords :
controller placement problem (CPP); multi-orbit; satellite network; Software-defined networking (SDN); Controller placement problem; Controller placements; Management IS; Multi-orbit; Networks management; Placement problems; Remote areas; Satellite network; Software-defined networking; Software-defined networkings; Artificial Intelligence; Computer Networks and Communications; Computer Science Applications; Signal Processing; Information Systems and Management; Renewable Energy, Sustainability and the Environment
Abstract :
[en] Satellite networks provide connectivity to remote areas, but efficient network management is crucial for seamless operations and resource optimization. Software-defined networks (SDN) offer flexibility, making them appealing for satellite network management. The dynamic placement of SDN controllers is crucial for an SDN-enabled satellite network, as it must adapt to dynamic topology changes, potential disruptions, limited capacity, and considerable latency. Existing literature lacks dynamism for controller placement in satellite networks considering multi-orbit scenarios, time-varying load and topology and facilitating swift adaptation to network variations. This work optimizes network performance by strategically positioning SDN controllers in multi-orbit satellite networks using a genetic algorithm based on network topology, traffic distribution, and link stability constraints considering low-earth and medium-earth satellites. The proposed scheme is evaluated through simulations using operational constellation parameters, showing 33.13% and 6.17% reduction in controller count than benchmarks static placement with dynamic assignment (SPDA) and fixed placement (FP).
Scopus citations®
without self-citations
0