机载雷达前视探测方位超分辨算法

发布时间：2018-05-24 11:26

  本文选题：机载雷达 + 前视探测　； 参考：《信号处理》2014年12期

【摘要】：机载雷达前视探测在民用与军用领域都具有广泛应用,例如跑道障碍物探测、低空飞行、自主着陆、侦察等。由于探测区域方位向多普勒带宽几乎为零,不满足DBS与SAR技术处理的条件,因此方位分辨率由天线波束3d B宽度决定。针对雷达前视探测方位低分辨率问题,本文提出一种基于改进雷达信号模型的统计优化超分辨算法。算法同时利用泊松与高斯噪声的分布先验信息进行数学建模,选择贝叶斯最大似然准则对超分辨反问题正则化,通过近似迭代求解实现雷达方位向超分辨。仿真与实测数据结果表明,与改进维纳滤波算法相比较,该算法降低了噪声敏感性,有效抑制虚假目标产生,对于提高波束主瓣内目标分辨能力具有实际应用意义。
[Abstract]:Airborne radar forward detection is widely used in both civil and military fields, such as runway obstacle detection, low altitude flight, autonomous landing, reconnaissance and so on. The azimuth Doppler bandwidth of the detected region is almost zero, which does not satisfy the processing conditions of DBS and SAR, so the azimuth resolution is determined by the antenna beam 3D B width. Aiming at the problem of low azimuth resolution of radar forward-looking detection, a statistical optimization super-resolution algorithm based on improved radar signal model is proposed in this paper. At the same time, the algorithm uses the prior information of Poisson and Gao Si noise distribution for mathematical modeling, selects Bayesian maximum likelihood criterion to regularize the super-resolution inverse problem, and realizes the radar azimuth super-resolution by approximate iterative solution. The simulation results show that compared with the improved Wiener filtering algorithm, the proposed algorithm reduces the noise sensitivity and effectively suppresses the false target generation. It is of practical significance for improving the resolution of the target in the main lobe of the beam.
【作者单位】： 电子科技大学电子工程学院;
【基金】：国防基础科研研究计划(B1420110182) 国家自然科学基金(61301273)
【分类号】：TN959.73
，

本文编号：1928878