| 摘要: |
| 在无人机的检测与识别过程中特征提取尤为重要。针对在低信噪比下由于提取特征困难且鲁棒性较差导致的识别效果不理想的问题，提出了一种基于改进的CVD（Cadence-Velocity Diagram）和Radon变换的特征提取方法应用于识别旋翼无人机。该方法通过提取目标的频率信息、峰值信息和边缘信息作为特征，运用K近邻分类器进行分类识别。仿真结果显示，在信噪比为-15 dB的条件下能够达到96.67%的识别精度。所提方法在低信噪比下识别效果明显优于SVM和朴素贝叶斯等算法。 |
| 关键词: 旋翼无人机 检测与识别 微多普勒 特征提取 低信噪比 |
| DOI： |
|
| 基金项目:国家自然科学基金资助项目（61561010）；广西自然科学基金项目（2017GXNSFAA198089）；广西重点研发计划项目（桂科AB18126003、AB16380316）；桂林电子科技大学研究生教育创新计划项目（2018YJCX21） |
|
| Rotorcraft UAV identification based on CVD and Radon transform |
| JIANG Liubing,JING Fengwei,CHE Li |
| (a.Key Laboratory of Wireless Broadband Communication and Signal Processing in Guangxi;b.School of Computer and Information Security，Guilin University of Electronic Technology，Guilin 541004，China;School of Information and Communication，Guilin University of Electronic Technology，Guilin 541004，China) |
| Abstract: |
| Feature extraction is especially important in the detection and identification process of rotorcraft unmanned aerial vehicle(UAV).For the problem that the recognition effect is not ideal due to the difficulty in extracting features and poor robustness at low signal-to-noise ratio(SNR),a feature extraction method based on improved cadence-velocity diagram(CVD) and Radon transform is proposed.Through extracting frequency information,peak information,and edge information of the target as features,this method uses K nearest neighbor(KNN) classifier to classify and recognize the information.The simulation results show that the recognition accuracy of 96.67% can be achieved when SNR is-15 dB.The recognition effect at low SNR is significantly better than that of support vector machine(SVM) and Naive Bayes. |
| Key words: rotorcraft unmanned aerial vehicle(UAV) detection and identification micro-Doppler feature extraction low SNR |
《电讯技术》编辑部 