Published 2023 | Version v1
Publication

Energy-Aware Satellite Handover Based on Deep Reinforcement Learning

Description

Multiple Low Earth Orbit (LEO) satellites have been deployed in constellations to provide User Equipments (UE)s with direct Internet connection at all times and from any location. UEs experience several handovers (HO)s during their service period due to the high speed of LEO satellites, which has a bad influence on UEs' Quality of Service (QoS) if occurred extensively. Furthermore, next-generation communication technologies are intended to serve a broad range of applications each with unique performance needs, thus distinguishing UEs with diverse and varied Traffic-Profiles (TP) has become necessary. Moreover, LEO satellites have limited onboard resources (e.g., energy and channel resources), and the deployed constellations ensure that each UE is covered by more than one LEO satellite at any time, making it difficult to pick the best satellite at each time to provide the best QoS. Therefore, a satellite HO strategy has to effectively use the limited available satellite resources and prevent network congestion while respecting the various TPs per UE. To address the aforementioned challenges, we propose a Load Balancing Energy Aware Satellite Handover (LBEASH) strategy, that is the first in the state of the art to address the limited energy resource of LEO satellites and the variety of UEs' performance requirements. The proposed LBEASH showcases significant achievements by avoiding unnecessary HOs and achieving a zero blocking rate while balancing the load among the satellites.

Additional details

Created:
July 3, 2024
Modified:
July 3, 2024