摘要
给出一种基于谱分析的多分量正弦调频信号估计算法。提取观测信号的时变相位,进行解缠绕处理;构造自适应带通滤波器,按照能量大小依次分离出各分量信号;利用频谱分析和卡森准则估计各分量信号的调制频率和调频指数。所给算法无需进行二维平面的峰值搜索,且基于谱分析的参数估计可借助快速傅里叶变换实现,故运算量较小。选取一段时长为1s的观测信号进行仿真实验,结果显示,信噪比大于8dB时,所给算法能够实现各分量信号的分离,且参数均方误差小于-10dB。
A new approach fo r parameter estimation of multi-component sinusoidal frequency modulation signal based on spectrum analysis is presented. Firstly , extract time-verying phase from the observed signal for unwrapping processing. Then , construct an adaptive bandpass filter to separate out all component signals in proper order by their energy from large to small. Finally , estimate the modulation frequency and frequency modulation index of each component signal by using frequency spectrum analysis and Carson criter ion. The proposed algorithm requires no search for peak value on the two dimensional plane, and based on the spectral analysis, the parameter estimation can be realized by the fast Fourier trans form , soitis of low computational cost. Simulation results based on a1slong signal show that, the proposed algorithm can well realize the separation of each component signal, and the mean square error of the parameter is less than -10 dB when the signal-to-noise ratio is greater than 8 dB.
出处
《西安邮电大学学报》
2017年第4期86-90,共5页
Journal of Xi’an University of Posts and Telecommunications
基金
国家自然科学基金资助项目(61202490)
陕西省自然科学基金资助项目(2014JM2-6117)
陕西省教育厅科学研究计划资助项目(15JK1654)
西安邮电大学校青年教师科研基金资助项目(101-0486)
关键词
参数估计
正弦调频信号
谱分析
卡森准则
调频指数
调制频率
parameter estimation, sinusoidal frequency modulation signal, spectrum analysis, Carson criterion , frequency modulation ind ex, modulation frequency
