基于k分辨FB级数展开的多分量LFM信号分离  

Multi-component LFM Signal Separation Based on k-FB Series Expansion

作　　者：何其芳[1] 张群[1] 罗迎[1] 梁必帅 HE Qifang ZHANG Qun LUO Ying LIANG Bishuai(Information and Navigation College, Air Force Engineering University, Xi＇an 710077, China Border Academy, Xi＇an 710108, China)

机构地区：[1]空军工程大学信息与导航学院,西安710077 [2]边防学院,西安710108

出　　处：《空军工程大学学报（自然科学版）》2017年第4期60-65,共6页Journal of Air Force Engineering University(Natural Science Edition)

基　　金：国家自然科学基金(61571457;61471386);陕西省青年科技新星计划(2016KJXX-49)

摘　　要：针对傅里叶-贝塞尔变换(FBT)难以估计和有效分离多分量LFM信号的问题,提出了一种k分辨-FB(k-FB)级数展开结合dechirp的信号分离与估计算法。在FB级数的基础上引入k分辨参数,通过理论推导,得出了信号频率与级数的关系,证明了参数估计精度与k取值正相关。通过解线频调和k-FB级数计算,实现了信号分离重构和参数估计。在不同信噪比、信号功率比和k分辨条件下对信号的分离精度进行了仿真研究,并与基于分数阶傅里叶变换(FrFT)的方法进行了对比。仿真结果验证了算法的有效性。Aimed at the problems that parameters are very difficult to be estimated and multi-component LFM signals are difficult to be separated by using the Fourier-Bessel transform（FBT）algorithm,a new method based on k-resolution FB（k-FB）series in conjunction with the Dechirp technique is proposed.The k-resolution is introduced into the FB series,and the relationship between the frequency content and the order of k-FB series is derived.Meanwhile,the positive correlation between the k-resolution and the estimation precision is proved.With the Dechirp technique and the k-FB series expansion,multi-component signal separation and parameter estimation are both realized.The performance of the signal separation precision under various SNR conditions with different k-resolution and signal power ratios is simulated.Moreover,aperformance comparison between the proposed method and the traditional fractional Fourier transform（FrFT）based method is also provided.The simulation results show that the method is valid.

关 键 词：k分辨-FB级数 多分量LFM信号 分离重构 参数估计 

分 类 号：TN971.1[电子电信—信号与信息处理]