论文部分内容阅读
在日益复杂的电磁环境中,如何提取有效特征是解决目标识别难题的一个关键。通过对目标信号的分析,发现它们虽呈现非平稳性,但却具有循环平稳性。因此,循环谱在分析此类信号方面具有优越的潜力,但是采用循环谱通常导致高维问题。针对这个问题,这里提出了降维循环谱的特征提取与目标识别方法,该方法以循环谱的相同频率点在不同循环频率下的相关性作为识别特征,并用主成分分析方法对该特征降维。实验结果表明,基于降维循环谱的方法具有很好的鲁棒性。
In the increasingly complex electromagnetic environment, how to extract effective features is a key to solve the problem of target identification. Through the analysis of the target signals, we found that although they exhibit nonstationarity, they are cyclostationary. Therefore, cyclic spectra have great potential in analyzing such signals, but the use of cyclic spectra often leads to high-dimensional problems. In order to solve this problem, a method of feature extraction and target recognition is proposed. The method takes the correlation of the same frequency points of the cyclic spectrum at different cyclic frequencies as the recognition feature and reduces the dimension of the feature by principal component analysis . Experimental results show that the method based on reduced-dimensional cyclic spectrum has good robustness.