Service Function Chain Deployment Algorithm Based on Deep Reinforcement Learning in Space–Air–Ground Integrated Network
SAGIN is formed by the fusion of ground networks and aircraft networks. It breaks through the limitation of communication, which cannot cover the whole world, bringing new opportunities for network communication in remote areas. However, many heterogeneous devices in SAGIN pose significant challenge...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Future Internet |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-5903/16/1/27 |
| _version_ | 1827369596883042304 |
|---|---|
| author | Xu Feng Mengyang He Lei Zhuang Yanrui Song Rumeng Peng |
| author_facet | Xu Feng Mengyang He Lei Zhuang Yanrui Song Rumeng Peng |
| author_sort | Xu Feng |
| collection | DOAJ |
| description | SAGIN is formed by the fusion of ground networks and aircraft networks. It breaks through the limitation of communication, which cannot cover the whole world, bringing new opportunities for network communication in remote areas. However, many heterogeneous devices in SAGIN pose significant challenges in terms of end-to-end resource management, and the limited regional heterogeneous resources also threaten the QoS for users. In this regard, this paper proposes a hierarchical resource management structure for SAGIN, named SAGIN-MEC, based on a SDN, NFV, and MEC, aiming to facilitate the systematic management of heterogeneous network resources. Furthermore, to minimize the operator deployment costs while ensuring the QoS, this paper formulates a resource scheduling optimization model tailored to SAGIN scenarios to minimize energy consumption. Additionally, we propose a deployment algorithm, named DRL-G, which is based on heuristics and DRL, aiming to allocate heterogeneous network resources within SAGIN effectively. Experimental results showed that SAGIN-MEC can reduce the end-to-end delay by 6–15 ms compared to the terrestrial edge network, and compared to other algorithms, the DRL-G algorithm can improve the service request reception rate by up to 20%. In terms of energy consumption, it reduces the average energy consumption by 4.4% compared to the PG algorithm. |
| first_indexed | 2024-03-08T09:54:43Z |
| format | Article |
| id | doaj.art-3bab2be18ccd4e1dbe6fb26745ce2b2b |
| institution | Directory Open Access Journal |
| issn | 1999-5903 |
| language | English |
| last_indexed | 2024-03-08T09:54:43Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Future Internet |
| spelling | doaj.art-3bab2be18ccd4e1dbe6fb26745ce2b2b2024-01-29T13:53:08ZengMDPI AGFuture Internet1999-59032024-01-011612710.3390/fi16010027Service Function Chain Deployment Algorithm Based on Deep Reinforcement Learning in Space–Air–Ground Integrated NetworkXu Feng0Mengyang He1Lei Zhuang2Yanrui Song3Rumeng Peng4The School of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450000, ChinaThe School of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450000, ChinaSchool of Computer and Artificial Intelligence, Zhengzhou University, Zhengzhou 450000, ChinaSchool of Computer and Artificial Intelligence, Zhengzhou University, Zhengzhou 450000, ChinaThe School of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450000, ChinaSAGIN is formed by the fusion of ground networks and aircraft networks. It breaks through the limitation of communication, which cannot cover the whole world, bringing new opportunities for network communication in remote areas. However, many heterogeneous devices in SAGIN pose significant challenges in terms of end-to-end resource management, and the limited regional heterogeneous resources also threaten the QoS for users. In this regard, this paper proposes a hierarchical resource management structure for SAGIN, named SAGIN-MEC, based on a SDN, NFV, and MEC, aiming to facilitate the systematic management of heterogeneous network resources. Furthermore, to minimize the operator deployment costs while ensuring the QoS, this paper formulates a resource scheduling optimization model tailored to SAGIN scenarios to minimize energy consumption. Additionally, we propose a deployment algorithm, named DRL-G, which is based on heuristics and DRL, aiming to allocate heterogeneous network resources within SAGIN effectively. Experimental results showed that SAGIN-MEC can reduce the end-to-end delay by 6–15 ms compared to the terrestrial edge network, and compared to other algorithms, the DRL-G algorithm can improve the service request reception rate by up to 20%. In terms of energy consumption, it reduces the average energy consumption by 4.4% compared to the PG algorithm.https://www.mdpi.com/1999-5903/16/1/27space–air–ground integrated networkDRLresource allocationNFV |
| spellingShingle | Xu Feng Mengyang He Lei Zhuang Yanrui Song Rumeng Peng Service Function Chain Deployment Algorithm Based on Deep Reinforcement Learning in Space–Air–Ground Integrated Network Future Internet space–air–ground integrated network DRL resource allocation NFV |
| title | Service Function Chain Deployment Algorithm Based on Deep Reinforcement Learning in Space–Air–Ground Integrated Network |
| title_full | Service Function Chain Deployment Algorithm Based on Deep Reinforcement Learning in Space–Air–Ground Integrated Network |
| title_fullStr | Service Function Chain Deployment Algorithm Based on Deep Reinforcement Learning in Space–Air–Ground Integrated Network |
| title_full_unstemmed | Service Function Chain Deployment Algorithm Based on Deep Reinforcement Learning in Space–Air–Ground Integrated Network |
| title_short | Service Function Chain Deployment Algorithm Based on Deep Reinforcement Learning in Space–Air–Ground Integrated Network |
| title_sort | service function chain deployment algorithm based on deep reinforcement learning in space air ground integrated network |
| topic | space–air–ground integrated network DRL resource allocation NFV |
| url | https://www.mdpi.com/1999-5903/16/1/27 |
| work_keys_str_mv | AT xufeng servicefunctionchaindeploymentalgorithmbasedondeepreinforcementlearninginspaceairgroundintegratednetwork AT mengyanghe servicefunctionchaindeploymentalgorithmbasedondeepreinforcementlearninginspaceairgroundintegratednetwork AT leizhuang servicefunctionchaindeploymentalgorithmbasedondeepreinforcementlearninginspaceairgroundintegratednetwork AT yanruisong servicefunctionchaindeploymentalgorithmbasedondeepreinforcementlearninginspaceairgroundintegratednetwork AT rumengpeng servicefunctionchaindeploymentalgorithmbasedondeepreinforcementlearninginspaceairgroundintegratednetwork |