Zhifeng Hu, Chong Han, Wolfgang Gerstacker, Ian F. Akyildiz
{"title":"Tera-SpaceCom:基于 GNN 的深度强化学习,用于 TeraHertz 频带空间网络中的联合资源分配和任务卸载","authors":"Zhifeng Hu, Chong Han, Wolfgang Gerstacker, Ian F. Akyildiz","doi":"arxiv-2409.07911","DOIUrl":null,"url":null,"abstract":"Terahertz (THz) space communications (Tera-SpaceCom) is envisioned as a\npromising technology to enable various space science and communication\napplications. Mainly, the realm of Tera-SpaceCom consists of THz sensing for\nspace exploration, data centers in space providing cloud services for space\nexploration tasks, and a low earth orbit (LEO) mega-constellation relaying\nthese tasks to ground stations (GSs) or data centers via THz links. Moreover,\nto reduce the computational burden on data centers as well as resource\nconsumption and latency in the relaying process, the LEO mega-constellation\nprovides satellite edge computing (SEC) services to directly compute space\nexploration tasks without relaying these tasks to data centers. The LEO\nsatellites that receive space exploration tasks offload (i.e., distribute)\npartial tasks to their neighboring LEO satellites, to further reduce their\ncomputational burden. However, efficient joint communication resource\nallocation and computing task offloading for the Tera-SpaceCom SEC network is\nan NP-hard mixed-integer nonlinear programming problem (MINLP), due to the\ndiscrete nature of space exploration tasks and sub-arrays as well as the\ncontinuous nature of transmit power. To tackle this challenge, a graph neural\nnetwork (GNN)-deep reinforcement learning (DRL)-based joint resource allocation\nand task offloading (GRANT) algorithm is proposed with the target of long-term\nresource efficiency (RE). Particularly, GNNs learn relationships among\ndifferent satellites from their connectivity information. Furthermore,\nmulti-agent and multi-task mechanisms cooperatively train task offloading and\nresource allocation. Compared with benchmark solutions, GRANT not only achieves\nthe highest RE with relatively low latency, but realizes the fewest trainable\nparameters and the shortest running time.","PeriodicalId":501301,"journal":{"name":"arXiv - CS - Machine Learning","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Tera-SpaceCom: GNN-based Deep Reinforcement Learning for Joint Resource Allocation and Task Offloading in TeraHertz Band Space Networks\",\"authors\":\"Zhifeng Hu, Chong Han, Wolfgang Gerstacker, Ian F. Akyildiz\",\"doi\":\"arxiv-2409.07911\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Terahertz (THz) space communications (Tera-SpaceCom) is envisioned as a\\npromising technology to enable various space science and communication\\napplications. Mainly, the realm of Tera-SpaceCom consists of THz sensing for\\nspace exploration, data centers in space providing cloud services for space\\nexploration tasks, and a low earth orbit (LEO) mega-constellation relaying\\nthese tasks to ground stations (GSs) or data centers via THz links. Moreover,\\nto reduce the computational burden on data centers as well as resource\\nconsumption and latency in the relaying process, the LEO mega-constellation\\nprovides satellite edge computing (SEC) services to directly compute space\\nexploration tasks without relaying these tasks to data centers. The LEO\\nsatellites that receive space exploration tasks offload (i.e., distribute)\\npartial tasks to their neighboring LEO satellites, to further reduce their\\ncomputational burden. However, efficient joint communication resource\\nallocation and computing task offloading for the Tera-SpaceCom SEC network is\\nan NP-hard mixed-integer nonlinear programming problem (MINLP), due to the\\ndiscrete nature of space exploration tasks and sub-arrays as well as the\\ncontinuous nature of transmit power. To tackle this challenge, a graph neural\\nnetwork (GNN)-deep reinforcement learning (DRL)-based joint resource allocation\\nand task offloading (GRANT) algorithm is proposed with the target of long-term\\nresource efficiency (RE). Particularly, GNNs learn relationships among\\ndifferent satellites from their connectivity information. Furthermore,\\nmulti-agent and multi-task mechanisms cooperatively train task offloading and\\nresource allocation. Compared with benchmark solutions, GRANT not only achieves\\nthe highest RE with relatively low latency, but realizes the fewest trainable\\nparameters and the shortest running time.\",\"PeriodicalId\":501301,\"journal\":{\"name\":\"arXiv - CS - Machine Learning\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Machine Learning\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.07911\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Machine Learning","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.07911","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Tera-SpaceCom: GNN-based Deep Reinforcement Learning for Joint Resource Allocation and Task Offloading in TeraHertz Band Space Networks
Terahertz (THz) space communications (Tera-SpaceCom) is envisioned as a
promising technology to enable various space science and communication
applications. Mainly, the realm of Tera-SpaceCom consists of THz sensing for
space exploration, data centers in space providing cloud services for space
exploration tasks, and a low earth orbit (LEO) mega-constellation relaying
these tasks to ground stations (GSs) or data centers via THz links. Moreover,
to reduce the computational burden on data centers as well as resource
consumption and latency in the relaying process, the LEO mega-constellation
provides satellite edge computing (SEC) services to directly compute space
exploration tasks without relaying these tasks to data centers. The LEO
satellites that receive space exploration tasks offload (i.e., distribute)
partial tasks to their neighboring LEO satellites, to further reduce their
computational burden. However, efficient joint communication resource
allocation and computing task offloading for the Tera-SpaceCom SEC network is
an NP-hard mixed-integer nonlinear programming problem (MINLP), due to the
discrete nature of space exploration tasks and sub-arrays as well as the
continuous nature of transmit power. To tackle this challenge, a graph neural
network (GNN)-deep reinforcement learning (DRL)-based joint resource allocation
and task offloading (GRANT) algorithm is proposed with the target of long-term
resource efficiency (RE). Particularly, GNNs learn relationships among
different satellites from their connectivity information. Furthermore,
multi-agent and multi-task mechanisms cooperatively train task offloading and
resource allocation. Compared with benchmark solutions, GRANT not only achieves
the highest RE with relatively low latency, but realizes the fewest trainable
parameters and the shortest running time.