| 48 | 0 | 83 |
| 下载次数 | 被引频次 | 阅读次数 |
辐射源个体识别技术在非合作无人机（Unmanned Aerial Vehicle，UAV）管控中具有重要应用意义。在低空飞行场景下，受山脉、建筑物等遮挡，接收无人机发射的信号时常常存在非视距传播情况，且无人机工作频段与地面现有无线电用频设备的频率相近，通常会导致接收到的无人机信号存在信噪比低、衰落影响大、有干扰等情况，严重影响对无人机信号的感知与个体识别能力。提出一种“信号检测-数据增强-智能识别”的一体化算法，设计了基于时频域二维扫描与导频数据盲提取的信号预处理方法，实现对无人机信号的检测，并有效滤除背景噪声和干扰，提高突发信号检测率和信噪比；设计了基于信道重构与动态加噪融合的数据增强算法，有效缓解信道衰落对无人机个体识别的影响，提升算法的泛化能力；对ResNet网络结构和学习策略进行了改进，提升对无人机信号的个体识别能力。在公开的DroneRFb-DIR数据集上实验结果表明，本文提出的算法能够实现复杂电磁环境下无人机信号的准确检测与识别，视距接收信号识别准确率达到97.1%，非视距接收信号的识别准确率达到81.1%。
Abstract：Specific emitter identification technology is of great application significance in non-cooperative unmanned aerial vehicle (UAV) management and control. In low-altitude flight scenarios， signal reception from UAVs often suffers from non-line-of-sight propagation due to occlusions such as mountains and buildings. Moreover， the operating frequency band of UAVs is close to that of existing terrestrial radio devices， which commonly results in the received UAV signals featuring low signal-to-noise ratio， severe fading and interference. These issues seriously degrade the capability of UAV signal perception and individual recognition. This paper proposes an integrated algorithm framework of “signal detection – data augmentation – intelligent recognition”. A signal preprocessing method based on two-dimensional time-frequency scanning and blind extraction of pilot data is designed to detect UAV signals， effectively suppress background noise and interference， and improve the detection rate and SNR of burst signals. A data augmentation algorithm combining channel reconstruction and dynamic noise addition is developed to mitigate the impact of channel fading on UAV individual recognition and enhance the generalization performance of the algorithm. The ResNet architecture and learning strategy are further modified to boost the individual recognition performance for UAV signals. Experimental results on the public DroneRFb-DIR dataset demonstrate that the proposed algorithm can achieve accurate detection and recognition of UAV signals under complex electromagnetic environments. The recognition accuracy reaches 97.1% for line-of-sight received signals and 81.1% for non-line-of-sight received signals.
[1] 刘正堂， 邹永杰， 孙健， 等. 俄乌战争中无人机攻防运用分析与启示[J]. 火力与指挥控制， 2025， 50(10): 207-212.
[2] FAN R， LI L， SUN P， et al. Feature Correlation Based RF Fingerprinting Identification for Walkie-Talkies Push-to-Talk Communications[J/OL]. Journal of Communications and Information Networks， 2025， 10(1): 57-70.
[3] HU N， LI Y， PAN W， et al. Geometric Distribution of UAV Detection Performance by Bistatic Radar[J/OL]. IEEE Transactions on Aerospace and Electronic Systems， 2024， 60(2): 2445-2452.
[4] 熊天宇， 刘俊， 谷雨， 等. 面向无人机识别的同平台LiDAR-可见光数据集构建[J]. 无线电工程， 2026， 56(4): 614-624.
[5] 马艳艳， 林强， 牛闯， 等. 复杂战场电磁环境研究综述[J/OL]. 现代防御技术， 2024， 52(4)[2026-04-11].
[6] LI D， YANG X， HU A， et al. LTE Device Radio Frequency Fingerprints Blind Extraction Based on Temporal-Frequency Domain PRACH Signals[J/OL]. IEEE Transactions on Vehicular Technology， 2023， 72(10): 13229-13242.
[7] RAJENDRAN S， SUN Z. RF Impairment Model-Based IoT Physical-Layer Identification for Enhanced Domain Generalization[J/OL]. IEEE Transactions on Information Forensics and Security， 2022， 17: 1285-1299.
[8] HENG C， JIN-FENG W， JIN-WEI C. Research on Intelligent SEI Method for Navigation Radar Based on Metric[C/OL]//2023 IEEE 3rd International Conference on Information Technology， Big Data and Artificial Intelligence (ICIBA): 2023: 1672-1678[2024-03-14].
[9] YANG X， LI D. LED-RFF: LTE DMRS-Based Channel Robust Radio Frequency Fingerprint Identification Scheme[J/OL]. IEEE Transactions on Information Forensics and Security， 2024， 19: 1855-1869.
[10] SOLTANIEH N， NOROUZI Y， YANG Y， et al. A Review of Radio Frequency Fingerprinting Techniques[J/OL]. IEEE Journal of Radio Frequency Identification， 2020， 4(3): 222-233.
[11] 王子健， 李歆昊， 谷业伟. 基于多维特征的FPV无人机个体识别方法[J/OL]. 电讯技术， 2025[2025-11-17].
[12] 王加琪， 王威. 多无人机强弱信号混叠下的检测与识别方法[J/OL]. 数据采集与处理， 2025， 40(6): 1464-1476.
[13] LUO S， NAKAMURA O， YAMADA R， et al. Performance Comparison of Neural Network and Minimum Mean Square Error Methods for Channel Estimation in Severe Frequency Selective Fading Channel[J/OL]. IEICE Communications Express， 2025， 14(8): 305-308.
[14] 张煜晟. 面向辐射源个体识别的射频指纹增强技术[D/OL]. 北京邮电大学， 2025[2025-11-19].
[15] SOLTANI N， SANKHE K， DY J， et al. More Is Better: Data Augmentation for Channel-Resilient RF Fingerprinting[J/OL]. IEEE Communications Magazine， 2020， 58(10): 66-72.
[16] LU Q， YANG Z， ZHANG H， et al. MRFE: A Deep-Learning-Based Multidimensional Radio Frequency Fingerprinting Enhancement Approach for IoT Device Identification[J/OL]. IEEE Internet of Things Journal， 2024， 11(18): 30442-30454.
[17] WANG Y， OHTSUKI T， SUN Z， et al. Avoiding Shortcuts: Enhancing Channel-Robust Specific Emitter Identification via Single-Source Domain Generalization[J/OL]. IEEE Transactions on Wireless Communications， 2025， 24(4): 3163-3176.
[18] WAN H， SHI F， WANG Y， et al. SigMix: Robust Specific Emitter Identification Method Enhanced by Cross-Time and Cross-Receiver Mixing Augmentation[J/OL]. IEEE Internet of Things Journal， 2025， 12(12): 21339-21349.
[19] 任俊宇， 俞宁宁， 周成伟， 等. DroneRFb-DIR:用于非合作无人机个体识别的射频信号数据集[J]. 电子与信息学报， 2025， 47(3): 573-581.
[20] 王豪， 罗俊松， 王惠明， 等. 无人机射频指纹识别方法综述[J]. 无线电工程， 2024， 54(11): 2672-2684.
[21] SUZUKI T， MATSUYAMA T， SAITO S， et al. Effectiveness Evaluation of Coded FTN Signaling in the Presence of Nonlinear Distortion Under Multipath Fading Channels[J/OL]. IEICE Communications Express， 2024， 13(7): 252-255.
[22] 王祥. 面向多通道信道仿真仪的信道模型软件开发与验证[D/OL]. 北京邮电大学， 2025[2025-11-20].
[23] ZHENG Y， WANG J， HUANG J. Open space radar specific emitter identification using MSAK‐CNN‐LSTM network[J/OL]. IET Radar， Sonar & Navigation， 2024， 18(7): 1080-1093.
基本信息:
中图分类号:TN911.23;V279
引用信息:
[1]林鹏,张邦宁,杨宁,等.复杂电磁环境下无人机辐射源个体识别算法[J].无线电工程().
2026-09-22
2026-09-22
2026-09-22