摘要
引入高阶累积量的概念对不同阶次FRFT对LFM信号的聚焦能力进行定量描述,构造了LFM信号的FR-FT-HOC特征谱,以此为基数分别结合步进式搜索和粒子群优化算法提出两种峰值点精搜方法,并根据峰值点坐标与LFM信号之间的定量关系实现了LFM参数估计。由于算法只涉及到一维参数搜索且定义域区间宽度可控,因此运算量大大降低。论文比较了四种FRFT-HOC特征的峰值区分能力,分析了FRFT阶次估计精度对LFM信号参数估计的影响,最后通过仿真实验对改进算法的LFM参数估计性能进行了验证。
In order to quantitatively describe the influence of the energy focusing property of FRFr to LFM signals, higher order cumulants are introduced to form the FRFT-HOC spectrum. And combined with multistage stepping search or PSO algorithm, two spectrum peak searching method are given, on the basis of which, an improved LFM parameter estimation algorithms based on FRFT is proposed. For the reason that the algorithm only involves one dimensional parameter searching and the confidence interval can be adjusted according to the searching step and the peak position, the commutation burden is reduced notably. In the paper, simulation results are given to describe the influence of FRFT order and the performance of the four proposed FRFT-HOC characteristics and the improved LFM parameter estimation algorithm is analyzed.
出处
《信号处理》
CSCD
北大核心
2013年第2期165-170,共6页
Journal of Signal Processing
基金
国家自然科学基金(61172148)
航空科学基金(20112096016)
关键词
高阶累积量
分数阶傅利叶变换
线性调频信号
Higher Order Cumulants
Fractional Fourier Transform
Linear Frequency Modulation
