Cooperative Resource Scheduling for Environment Sensing in Satellite–Terrestrial Vehicular Networks
- 7 November 2024
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (6) , 6734-6748
Abstract
In this paper, we investigate infrastructure-assisted environment sensing in satellite-terrestrial vehicular networks (STVN) for connected autonomous vehicles (CAVs), where satellites and roadside units (RSUs) cooperate to provide CAVs with fresh sensing data. To support satellite-and RSU-assisted environment sensing for CAVs, we formulate a long-term resource scheduling problem in STVN to satisfy sensing data freshness requirements with efficient resource usage. To deal with the challenges posed by the dynamic network environment as well as stringent data freshness requirements, we propose a cooperative satellite-terrestrial resource scheduling (CSTRS) scheme. CSTRS is a model-data co-driven approach that can jointly optimize the sensing interval and resource allocation in STVN. Specifically, benefiting from the multicast feature of the LEO satellite, coalition game and particle swarm optimization-based algorithms are designed to partition CAVs into groups and optimize sensing intervals in large timescales. Then, a reinforcement learning-based algorithm is developed to make real-time computing and communication resource allocation decisions based on the CAV partition. Simulation results demonstrate that the proposed scheme outperforms benchmark methods in terms of resource usage and reliability performance.Keywords
Funding Information
- Natural Sciences and Engineering Research Council of Canada (RGPIN-2023-03759)
This publication has 30 references indexed in Scilit:
- Non-geostationary orbit constellation design for global connectivity2022
- Learning-based Cache Placement and Content Delivery for Satellite-Terrestrial Integrated Networks2021
- Adaptive Access Mode Selection in Space-Ground Integrated Vehicular Networks2021
- Sensing system of environmental perception technologies for driverless vehicle: A review of state of the art and challengesSensors and Actuators A: Physical, 2021
- Network Traffic Prediction Using Recurrent Neural Networks2018
- Stochastic Network Optimization with Application to Communication and Queueing Systems2010
- Exponential-time approximation of weighted set coverInformation Processing Letters, 2009
- The particle swarm - explosion, stability, and convergence in a multidimensional complex spaceIEEE Transactions on Evolutionary Computation, 2002
- The Stability of Hedonic Coalition StructuresGames and Economic Behavior, 2002
- Elementary Number Theory1998