Litcius/Paper detail

Mission-Driven Resource Scheduling in Satellite-Terrestrial Networks: From Perspective of Collaboration and Reconfiguration

Di Zhou, Min Sheng, Chenxi Bao, Yixin Wang, Jiandong Li, Zhu Han

2025IEEE Transactions on Communications34 citationsDOI

Abstract

Satellite-terrestrial networks (STNs) are emerging as a promising solution for provisioning comprehensive services, such as the Internet of Remote Things (IoRT) and remote sensing, within the realm of 6G wireless networks. Nonetheless, resource failures and the exigencies of diverse mission urgencies exacerbate the intricacies of resource scheduling in STNs, thus impeding the effective alignment of distinct mission requirements with dynamic resources. In light of these challenges, we first mathematically formulate the complex resource scheduling problem in STNs as a stochastic optimization paradigm, endeavoring to maximize the number of successfully accomplished missions. Subsequently, we conceptualize the resource evolution to delineate scheduling dynamics, encompassing potential contingencies of resource discontinuities. Next, we propose an innovative hierarchical deep learning-based mission-driven resource scheduling (HDL-MDRS) algorithm, aimed at optimizing resource collaboration and reconfiguration to amplify network performance within the dynamic ambits characterized by resource disruptions. The HDL-MDRS algorithm achieves a coarse-grained alignment of diverse mission requirements with multidimensional resources. It enhances overall mission fulfillment and network resource utilization efficiency through fine-grained collaboration and reconfiguration among satellites, both within and across different clusters. Notably, the simulation findings substantiate the effectiveness of the HDL-MDRS algorithm, effectively ensuring the requirements of different types of missions in case of the unforeseen resource failures, orchestrated through efficient resource collaboration and on-demand reconfiguration.

Topics & Concepts

Control reconfigurationScheduling (production processes)Communications satelliteComputer scienceSatellitePerspective (graphical)Distributed computingTelecommunicationsComputer networkEngineeringEmbedded systemAerospace engineeringOperations managementArtificial intelligenceSatellite Communication SystemsDistributed and Parallel Computing Systems
Mission-Driven Resource Scheduling in Satellite-Terrestrial Networks: From Perspective of Collaboration and Reconfiguration | Litcius