Resource allocation of distributed MIMO radar based on the hybrid action space reinforcement learning

Abstract The distributed multiple-input multiple-output (MIMO) radar system exhibits superior target localization capability by jointly processing target information from multiple radars under different observation angles. To improve the resource utilization of the distributed MIMO radar system, thi...

Full description

Saved in:
Bibliographic Details
Main Authors: Jin Zhu, Wenxu Liu, Feifei Lyu, Siwei Li, Tianyang Zhang
Format: Article
Language:English
Published: Nature Portfolio 2025-06-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-02698-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850136589323206656
author Jin Zhu
Wenxu Liu
Feifei Lyu
Siwei Li
Tianyang Zhang
author_facet Jin Zhu
Wenxu Liu
Feifei Lyu
Siwei Li
Tianyang Zhang
author_sort Jin Zhu
collection DOAJ
description Abstract The distributed multiple-input multiple-output (MIMO) radar system exhibits superior target localization capability by jointly processing target information from multiple radars under different observation angles. To improve the resource utilization of the distributed MIMO radar system, this paper proposes a hybrid action space reinforcement learning (HAS-RL) method, aiming to maximize the target localization performance under the radar resource constraints. Specifically, the Cramer–Rao Lower Bound (CRLB) incorporating the transmit radar power and receive radar selection is first derived and employed as the target localization performance metric of the distributed MIMO radar system. Subsequently, the radar resource allocation problem is modeled as a constrained optimization problem with continuous and discrete variables, and a hybrid action space reinforcement learning is proposed to solve the above optimization problem. Simulation results demonstrate that the proposed HAS-RL method can obtain better target localization performance under the given radar resource constraints.
format Article
id doaj-art-e12b828282ca4c5db0117c5e21687f8b
institution OA Journals
issn 2045-2322
language English
publishDate 2025-06-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-e12b828282ca4c5db0117c5e21687f8b2025-08-20T02:31:04ZengNature PortfolioScientific Reports2045-23222025-06-0115111110.1038/s41598-025-02698-1Resource allocation of distributed MIMO radar based on the hybrid action space reinforcement learningJin Zhu0Wenxu Liu1Feifei Lyu2Siwei Li3Tianyang Zhang4School of Artificial Intelligence, Xidian UniversityCETC Key Laboratory of Aerospace Information Applications, The 54th Research Institute of China Electronics Technology Group CorporationCETC Key Laboratory of Aerospace Information Applications, The 54th Research Institute of China Electronics Technology Group CorporationSchool of Artificial Intelligence, Xidian UniversitySchool of Artificial Intelligence, Xidian UniversityAbstract The distributed multiple-input multiple-output (MIMO) radar system exhibits superior target localization capability by jointly processing target information from multiple radars under different observation angles. To improve the resource utilization of the distributed MIMO radar system, this paper proposes a hybrid action space reinforcement learning (HAS-RL) method, aiming to maximize the target localization performance under the radar resource constraints. Specifically, the Cramer–Rao Lower Bound (CRLB) incorporating the transmit radar power and receive radar selection is first derived and employed as the target localization performance metric of the distributed MIMO radar system. Subsequently, the radar resource allocation problem is modeled as a constrained optimization problem with continuous and discrete variables, and a hybrid action space reinforcement learning is proposed to solve the above optimization problem. Simulation results demonstrate that the proposed HAS-RL method can obtain better target localization performance under the given radar resource constraints.https://doi.org/10.1038/s41598-025-02698-1Resource allocationRadar signal processingReinforcement learning
spellingShingle Jin Zhu
Wenxu Liu
Feifei Lyu
Siwei Li
Tianyang Zhang
Resource allocation of distributed MIMO radar based on the hybrid action space reinforcement learning
Scientific Reports
Resource allocation
Radar signal processing
Reinforcement learning
title Resource allocation of distributed MIMO radar based on the hybrid action space reinforcement learning
title_full Resource allocation of distributed MIMO radar based on the hybrid action space reinforcement learning
title_fullStr Resource allocation of distributed MIMO radar based on the hybrid action space reinforcement learning
title_full_unstemmed Resource allocation of distributed MIMO radar based on the hybrid action space reinforcement learning
title_short Resource allocation of distributed MIMO radar based on the hybrid action space reinforcement learning
title_sort resource allocation of distributed mimo radar based on the hybrid action space reinforcement learning
topic Resource allocation
Radar signal processing
Reinforcement learning
url https://doi.org/10.1038/s41598-025-02698-1
work_keys_str_mv AT jinzhu resourceallocationofdistributedmimoradarbasedonthehybridactionspacereinforcementlearning
AT wenxuliu resourceallocationofdistributedmimoradarbasedonthehybridactionspacereinforcementlearning
AT feifeilyu resourceallocationofdistributedmimoradarbasedonthehybridactionspacereinforcementlearning
AT siweili resourceallocationofdistributedmimoradarbasedonthehybridactionspacereinforcementlearning
AT tianyangzhang resourceallocationofdistributedmimoradarbasedonthehybridactionspacereinforcementlearning