中文摘要：

针对多分量线性调频信号存在重叠、漏检现象,导致波达方向的估计存在较大误差问题,提出多分量线性调频信号分数阶傅里叶域量纲归一化方法。首先给出了多分量LFM信号的阵列模型,讨论了阵列模型信号时频坐标系进行量纲归一化。通过对相干环境下分数阶傅里叶域进行量纲归一化操作,提高多分量信号在分数阶傅里叶域的分辨能力,保证信源数目估计的准确性。仿真结果表明该方法正确的识别信源组数,提高多分量信号波达方向估计精确度,信噪比低于-4 d B的信号抗干扰能力明显增强。

英文摘要：

In order to solve the problem that the multi-component linear frequency modulation（ LFM） signals can overlap,even some signals can＇t be detected,so that the errors of the direction of arrival（ DOA） estimation is too large,the method that DOA estimation for multi-component LFM signals dimensional normalization in fractional Fourier domain is proposed. Firstly,the array model of LFM signal is constructed,then,the dimensional normalization of LFM array model signal in time-frequency domain is discussed,and how to select a suitable dimensional normalization parameter is analyzed. Dimensional normalization in fraction Fourier domain can improve the resolution ability of the FRFT for the multi-component LFM signals,avoid the peaks of multi-component LFM signal overlap,and the accuracy of the number of source can ensure. The simulation result shows the effectiveness of estimating the direction of arrival for multi-component LFM signals,and the method can improve the estimation accuracy of signal parameter and performance analysis. This method can also improve the ant jamming capability of signal that SNR is under- 4 d B.