Memory‐based deep reinforcement learning for cognitive radar target tracking waveform resource management
Abstract A cognitive radar (CR) system can offer enhanced target tracking performance due to its intelligence on the perception‐action cycle, wherein a CR adaptively allocates the limited transmitting resources based on its perception of surrounding environments. To effectively manage the transmit w...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-12-01
|
| Series: | IET Radar, Sonar & Navigation |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/rsn2.12469 |
| _version_ | 1827587751338311680 |
|---|---|
| author | Jiahao Qin Mengtao Zhu Zesi Pan Yunjie Li Yan Li |
| author_facet | Jiahao Qin Mengtao Zhu Zesi Pan Yunjie Li Yan Li |
| author_sort | Jiahao Qin |
| collection | DOAJ |
| description | Abstract A cognitive radar (CR) system can offer enhanced target tracking performance due to its intelligence on the perception‐action cycle, wherein a CR adaptively allocates the limited transmitting resources based on its perception of surrounding environments. To effectively manage the transmit waveform resource for the target tracking task, CR resource management problem is formulated under the partially observable Markov decision process framework. The sequential decision‐making and the inherent partial observability for target tracking problem are considered. In the proposed method, a long short‐term memory (LSTM)‐based twin delayed deep deterministic policy gradient (TD3) algorithm is developed to effectively solve the problem. A reward function is designed considering Haykin's cognitive executive attention mechanism for radar systems such that the CR resource management policy has stability in the decision of transmit waveform, which follows the principle of minimum disturbance. Simulation results demonstrate the superiority of the proposed LSTM memory‐based TD3 with improved target tracking performance and increased mean rewards for CR. |
| first_indexed | 2024-03-09T00:23:28Z |
| format | Article |
| id | doaj.art-94948ff6faf8416595d1712b860ba445 |
| institution | Directory Open Access Journal |
| issn | 1751-8784 1751-8792 |
| language | English |
| last_indexed | 2024-03-09T00:23:28Z |
| publishDate | 2023-12-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Radar, Sonar & Navigation |
| spelling | doaj.art-94948ff6faf8416595d1712b860ba4452023-12-12T05:20:22ZengWileyIET Radar, Sonar & Navigation1751-87841751-87922023-12-0117121822183610.1049/rsn2.12469Memory‐based deep reinforcement learning for cognitive radar target tracking waveform resource managementJiahao Qin0Mengtao Zhu1Zesi Pan2Yunjie Li3Yan Li4School of Cyberspace Science and Technology Beijing Institute of Technology Beijing ChinaSchool of Cyberspace Science and Technology Beijing Institute of Technology Beijing ChinaSchool of Information and Electronics Beijing Institute of Technology Beijing ChinaLaboratory of Electromagnetic Apace Cognition and Intelligent Control Beijing ChinaSchool of Cyberspace Science and Technology Beijing Institute of Technology Beijing ChinaAbstract A cognitive radar (CR) system can offer enhanced target tracking performance due to its intelligence on the perception‐action cycle, wherein a CR adaptively allocates the limited transmitting resources based on its perception of surrounding environments. To effectively manage the transmit waveform resource for the target tracking task, CR resource management problem is formulated under the partially observable Markov decision process framework. The sequential decision‐making and the inherent partial observability for target tracking problem are considered. In the proposed method, a long short‐term memory (LSTM)‐based twin delayed deep deterministic policy gradient (TD3) algorithm is developed to effectively solve the problem. A reward function is designed considering Haykin's cognitive executive attention mechanism for radar systems such that the CR resource management policy has stability in the decision of transmit waveform, which follows the principle of minimum disturbance. Simulation results demonstrate the superiority of the proposed LSTM memory‐based TD3 with improved target tracking performance and increased mean rewards for CR.https://doi.org/10.1049/rsn2.12469adaptive radardecision makingintelligent networks |
| spellingShingle | Jiahao Qin Mengtao Zhu Zesi Pan Yunjie Li Yan Li Memory‐based deep reinforcement learning for cognitive radar target tracking waveform resource management IET Radar, Sonar & Navigation adaptive radar decision making intelligent networks |
| title | Memory‐based deep reinforcement learning for cognitive radar target tracking waveform resource management |
| title_full | Memory‐based deep reinforcement learning for cognitive radar target tracking waveform resource management |
| title_fullStr | Memory‐based deep reinforcement learning for cognitive radar target tracking waveform resource management |
| title_full_unstemmed | Memory‐based deep reinforcement learning for cognitive radar target tracking waveform resource management |
| title_short | Memory‐based deep reinforcement learning for cognitive radar target tracking waveform resource management |
| title_sort | memory based deep reinforcement learning for cognitive radar target tracking waveform resource management |
| topic | adaptive radar decision making intelligent networks |
| url | https://doi.org/10.1049/rsn2.12469 |
| work_keys_str_mv | AT jiahaoqin memorybaseddeepreinforcementlearningforcognitiveradartargettrackingwaveformresourcemanagement AT mengtaozhu memorybaseddeepreinforcementlearningforcognitiveradartargettrackingwaveformresourcemanagement AT zesipan memorybaseddeepreinforcementlearningforcognitiveradartargettrackingwaveformresourcemanagement AT yunjieli memorybaseddeepreinforcementlearningforcognitiveradartargettrackingwaveformresourcemanagement AT yanli memorybaseddeepreinforcementlearningforcognitiveradartargettrackingwaveformresourcemanagement |