摘要
针对复杂体制雷达辐射源信号分类问题,提出一种基于时频分析、图像处理和支持向量机的辐射源信号分类新方法。该方法将辐射源信号分类问题转换为图像处理及识别问题,先对辐射源信号进行时频分析,获得时频分布图,并将其转化为灰度图像并作归一化处理,再用支持向量机对处理后的图像进行分类。5种典型辐射源信号分类实验表明,该方法在信噪比高于2.5 dB时,平均正确分类率达92%以上。
To correctly classify advanced radar emitter signals, this papers presents a novel method consisting of time-frequency analysis, image processing and support vector machine. The method transforms the classification of emitter signals into image processing and image recognition. Emitter signals are analyzed in time-frequency domain and their time-frequency images are obtained. These images are transformed into grayscale images and normalized ones. Support vector machines are used to design classifiers for recognizing the processed images and the images correspond to different radar emitter signals. Experiments conducted on five typical emitter signals show that the correct classification rate of the method is more than 92% when the signal-to-noise ratio（SNR） is above 2. 5dB.
出处
《数据采集与处理》
CSCD
北大核心
2009年第4期487-492,共6页
Journal of Data Acquisition and Processing
基金
国家自然科学基金(60702026
60572143)资助项目
四川省青年科技基金(09ZQ026-040)资助项目
关键词
雷达辐射源信号
自动分类
时频分析
图像处理
支持向量机
radar emitter signals
automatic classification
time-frequency analysis
image processing
support vector machine
