基于改进FRFT的高动态LFM信号估计算法  

High Dynamic LFM Signal Estimation Algorithm Based on Improved FRFT

作　　者：杜锋[1] 李广侠[1] 向中武 唐璟宇 程磊 云飞龙 

机构地区：[1]解放军理工大学通信工程学院 [2]解放军61541部队

出　　处：《电子质量》2016年第5期5-8,共4页Electronics Quality

摘　　要：高动态环境下线性调频信号存在较大的频偏和多普勒变化率,传统的估计方法是采用似然函数的二维搜索,计算量过于复杂。虽然借助于分数阶傅立叶变换可将二维降为一维,但是该算法的每一次搜索仍是一次分数阶傅立叶变换,难以在计算复杂度和精度之间取得较好的折中。针对这一问题,该文提出一种改进分数阶傅立叶变换辅助的高动态信号估计算法,首先通过分数阶傅立叶变换得到多普勒变化率粗估值,然后采用多层分段FFT进行精估,待动态被削弱之后,最后采用FFT估计频偏。该算法在较低信噪比条件下采用大步进值的分数阶傅立叶变换得到的误差曲线接近于克拉美-罗界,并且计算复杂度仅为传统方法的5%。The linear frequency modulation signal has a large frequency offset and Doppler rate under High dynamic environment,the traditional estimation method is to use the two-dimensional search with the likelihood function,the calculation is too complicated.Although by means of fractional Fourier transforms can be reduced the two-dimensional to one-dimensional,each time the algorithm search is still a fractional Fourier transform.It is difficult to obtain a good compromise between computational complexity and accuracy.To solve this problem,this paper proposes an improved high dynamic signal estimation based on fractional Fourier transform secondary,first the crude valuation of the Doppler rate is obtained by Fractional Fourier Transform and then the fine estimation by multi-layer and segmented FFT,after the weakening of dynamic,the final FFT frequency offset estimation is used.The estimation Error curve with a large step value of Fractional Fourier transform under lower SNR is close to Cramer-Rao bound,and the computational complexity of the conventional method is of only 5% relatively.

关 键 词：分数阶傅立叶变换 高动态 线性调频信号 多层分段FFT 多普勒变化率 

分 类 号：TN911.23[电子电信—通信与信息系统]