Grant: Discovery Launch Supplement

Mobility management in future software-defined satellite networks Internet service has become one of the essential utilities. The 2021 report of the International Telecommunication Union, the United Nations specialized agency for information and communication technologies, reveals that 2.9 billion people around the world still do not have Internet services. Building and deploying terrestrial infrastructure communication networks in remote and rural areas is not always possible due to the place geographic nature and cost inefficiency. To connect the unconnected communities, huge investments are being made to provide global Internet services via satellite networks. For example, the Starlink project, operated by SpaceX, aims to create a global network of low Earth orbit (LEO) satellites to provide global Internet services. LEO satellites are closer to Earth than geosyncorouns (GEO) satellites. This reduces communication delays significantly and makes LEO satellites suitable for providing Internet services to users on Earth. Since LEO satellites are located on low orbits, they must orbit the Earth at high speeds. LEO satellite network topology is very dynamic and varies over time. For an Internet user on Earth, the high speed of LEO satellites results in a handover (switching of Internet connection from one satellite to another) every 5-10 minutes. To maintain continuous satellite services for users, implementing mobility management is necessary for future satellite networks (SatNets). Mobility management has two components, location management and handover management. In the future, there will be thousands of LEO satellites orbiting Earth. Managing such a large-scale and highly dynamic network of satellites using conventional mobility management techniques is inefficient. Software-defined network (SDN) concept supports softwarized and flexible network management by separating the control plane from the data plane. Several studies considered SDN a promising technology for creating adaptive mobility management solutions. By coupling SDN with artificial intelligence or machine learning algorithms, intelligent and autonomous mobility management can be achieved. However, implementing the concept of SDN in satellite networks is not a straightforward process due to the large-scale and dynamic nature of satellite networks. This research aims to develop an intelligent mobility management framework for software-defined LEO satellite networks. To address the scalability problem, an adaptive network clustering algorithm will be developed to support cluster-based mobility management. Novel and intelligent algorithms will be developed for both SDN controller selection and SDN switch-to-controller association in future SatNets. To provide continuous Internet service for users on Earth, a novel handover management algorithm will be developed. The performance of the proposed algorithms will be evaluated through numerical analysis and simulation results.