|
|
|
| Intelligent Spectrum Sensing Technology in Unmanned Air Defense Systems |
| XUE Yujia1， ZHUANG Ruowang2， XU Yi2， RUAN Kaizhi2，3，
ZHONG Kai1， HU Jinfeng4 |
| 1. School of Information and Communication Engineering， University of Electronic Science and
Technology of China，Chengdu 611731， Sichuan， China；
2. Shanghai Electro-Mechanical Engineering Institute， Shanghai 201109， China；
3. Key Laboratory of Automatic Target Recognition ，Shanghai 201109， China；
4. Intelligent Terminal Sichuan Province Key Laboratory， Yibin 644000， Sichuan， China |
|
|
|
|
Abstract With modern warfare evolving toward informatization and intelligentization, unmanned air defense systems face severe challenges in complex and dynamic electromagnetic environments, rendering efficient and robust spectrum sensing capabilities an urgent requirement. Traditional methods relying on explicit statistical features are critically dependent on accurate channel and noise modeling, exhibiting insufficient robustness under low signal-to-noise ratio (SNR) conditions. Although existing deep learning approaches demonstrate powerful implicit feature extraction capabilities, they generally suffer from inadequate discriminative feature focusing and excessively high model complexity, making them difficult to deploy in resource-constrained unmanned systems. This paper observes that the heterogeneous features presented by signals at the temporal waveform and statistical feature levels exhibit strong complementarity, and is committed to addressing model lightweighting. To this end, a lightweight Dual-Stream Feature Fusion Network (DSFF-Net) is proposed: the network captures local patterns and global features of signals through a Temporal Feature Extraction Network (TF-Net) and a Statistical Feature Extraction Network (SF-Net), respectively; subsequently, a gated attention mechanism is employed to perform weighted selection and fusion of the dual-stream features, achieving end-to-end high-performance spectrum sensing. Experimental results demonstrate that at a fixed false alarm probability of 0.06, the detection probability of DSFF-Net improves significantly compared with the conventional energy detection method. Particularly in the low-SNR range of -16 dB to -12 dB, the detection probability increases by 5%-20% compared with deep learning methods such as CNN and CL-DNN. More importantly, the number of network parameters of DSFF-Net is approximately 0.5 orders of magnitude lower than that of the aforementioned deep learning methods, effectively validating that the proposed method ensures superior sensing performance while possessing the advantages of lightweight architecture and high deployability.
|
|
Received: 30 January 2026
Published: 10 July 2026
|
|
|
|
|
|
| [1] |
CHEN Liangwen, ZHU Yuxin, SHEN Tao, YU Yifan, LING Xiao, SHENG Qinghong. Deep Learning-Based Infrared Ship Target Wake Matching and Detection Algorithm[J]. Air & Space Defense, 2026, 9(1): 80-90. |
| [2] |
LUO Zhijun, WANG Jianrui, YIN Jiawei. A Survey of Task-Driven Intelligent Target Recognition Methods in Complex Battlefield Environments[J]. Air & Space Defense, 2026, 9(1): 1-11. |
| [3] |
RONG Guang, ZHANG Yexin, TANG Chao, CHEN Jinbao, ZHOU Yiling, WANG Jianyuan. Study on Simulation Data-Driven Fault Diagnosis Technology for Unmanned Aerial Vehicles[J]. Air & Space Defense, 2025, 8(6): 73-84. |
| [4] |
XIA Yilin, LIU Gang, YAN Congqiang, CAI Yunze. Research on Deep Learning-Based Rotation Detection Algorithms for Ship Wakes in SAR Images[J]. Air & Space Defense, 2025, 8(5): 64-74. |
| [5] |
XU Qiang, MA Yuehua, XU Ke, PAN Jun. A Review on Intelligent Radar Target Recognition Methods[J]. Air & Space Defense, 2025, 8(5): 1-9. |
| [6] |
ZHAO Ziyu, WANG Xuquan, MA Jie, XING Yujie, DUN Xiong, WANG Zhanshan, CHENG Xinbin. Edge Chip Deployment Methods for Lightweight Infrared Computational Imaging Reconstruction Algorithms[J]. Air & Space Defense, 2025, 8(4): 85-93. |
| [7] |
CUI Shan, PAN Junyang, WANG Wei, GUO Ye, XU Jiangtao. Air Defence and Anti-Missile Interception Decision-Making Study Based on Deep Learning[J]. Air & Space Defense, 2024, 7(5): 54-64. |
| [8] |
LIU Jing, GUO Xiaolei, ZHANG Xinhai, MAO Jingjun, LYU Ruiheng. A Lightweight Aerial Rotated Object Detection Algorithm of Air-to-Surface Missile[J]. Air & Space Defense, 2024, 7(4): 106-113. |
| [9] |
LIN Zhaochen, ZHANG Xinran, LIU Ziyang, HE Fenghua, OUYANG Lei. Deep Learning-Based Hypersonic Vehicle Motion Behavior Recognition[J]. Air & Space Defense, 2024, 7(1): 48-55. |
| [10] |
WANG Bing, PI Gang, CHEN Wencheng, XIE Haifeng, SHI Xiangling. Deep Learning Based Surface Defect Detection Method of Flexible Solar Array Hinge[J]. Air & Space Defense, 2023, 6(1): 96-101. |
| [11] |
JIN Lijie, WU Yatao. Radar Signal Modulation Type Recognition Based on Double CNN[J]. Air & Space Defense, 2022, 5(1): 66-70. |
| [12] |
TAO Haihong, YAN Yingfei. A Netted Radar Node Selection Algorithm Based on GA-CNN[J]. Air & Space Defense, 2022, 5(1): 1-5. |
| [13] |
CAI Yunze, ZHANG Yanjun. Infrared Dim and Small Target Detection Based on Dual-Channel Feature-Enhancement Integrated Attention Network[J]. Air & Space Defense, 2021, 4(4): 14-22. |
| [14] |
CAO Jinghao, ZHANG Junju, HUANG Wei, YAO Ruotong, ZHANG Ping. UAV Recognition and Detection Based on Multi-scale Feature Fusion[J]. Air & Space Defense, 2021, 4(1): 60-64. |
| [15] |
CHEN Jichi, WEI Guohua, GUO Conglong, ZHANG Lihe. Preliminary Study on a Deep Learning 3D Reconstruction Simulation Method Based on Infrared Image Sequence[J]. Air & Space Defense, 2020, 3(4): 21-29. |
|
|
|
|
2026, Vol. 9 
