【作者】 吴露露；

【导师】 朱耀庭；

【作者基本信息】 华中科技大学 ， 信息与通信工程， 2009， 博士

【摘要】 有效地探测隐身目标已成为当前国防急需解决的问题。利用微波辐射计探测隐身目标是一种有效的无源探测方法，该方法具有隐蔽性好，不受气候条件和战场烟尘影响等优点。采用天线阵列可将多个小口径天线合成为一个等效的大天线，以提高空间分辨率，但传统的阵列信号处理方法(如：综合孔径算法)的空间分辨率受到天线阵列最大基线长度的限制。为了进一步提高分辨率，论文将阵列信号处理中的空间谱估计方法应用于微波热辐射稀疏阵列接收系统，提出一种新的基于目标微波热辐射信号和空间谱估计的全被动探测方法，该方法利用天线阵列实现高分辨率；利用空间谱估计进一步提高分辨率(实现超分辨率，超越硬件系统空间分辨率的极限)；利用稀疏排列减少所需的阵元数，以保证在不降低系统分辨力的情况下降低系统的复杂度和成本。论文对隐身目标的微波热辐射及传输特性进行了分析，建立了集空中隐身目标、大气传输信道及微波热辐射探测系统于一体的隐身目标微波热辐射信号传输模型，可给出隐身目标和背景的辐射亮温分布，并辅以计算机仿真和实测实验验证了该模型的正确性和有效性。论文从本质上对空间谱估计算法的内在联系和区别作了归纳和总结，选取部分主流算法对其性能进了分析及仿真。并研究了将空间谱估计算法直接用于稀疏阵列所带来的阵列流型模糊进而导致的信号方向估计模糊问题，提出了一种基于任意整数倍间距排列的线阵阵列流型维数的计算方法，可以快速得到以任意整数倍间距排列线阵的维数及与入射信号数相对应的所有模糊角度，可用于解决部分线阵由于稀疏排列所引起信号方向估计模糊问题。在此基础上，论文建立了基于微波热辐射信号及空间谱估计的隐身目标探测模型，利用该模型分析了理想探测系统的性能，包括灵敏度和分辨率，并与采用其他信号处理方法的探测系统的性能通过理论分析和实验进行了对比，结果表明，采用空间谱估计的探测系统的分辨率高于采用其他信号处理方法的探测系统的分辨率，可实现目标的超分辨探测。实际系统不可避免地存在误差，误差会降低空间谱估计的性能，需要研究误差模型和误差校正方法。论文首先对阵列误差、阵列误差模型和校正算法进行了概述和分类。然后针对微波热辐射信号弱的特点，建立了低信噪比阵列误差模型，并基于该模型提出了两种新的阵列误差校正算法。最后通过仿真和实验验证了在低信噪比和存在误差的情况下，论文所提出的阵列误差模型及校正算法依然使得具有高分辨率的空间谱估计算法能够很好地应用于微波热辐射阵列接收系统进行隐身目标的探测。更多还原

【Abstract】 How to effectively detect stealthy targets has become a critical problem as far as national defense issues are concerned. The microwave radiometer detection approach, which is highly invisible and barely influenced by weather or battlefield environment, has proved itself a potent way to detect targets passively. The antenna array constructed of several small antennas has a very high space resolution which is equivalent to that of a big antenna. While the space resolution of traditional array signal processing algorithms, such as inversion algorithm of synthetic aperture microwave radiometer, are determined by the maximum baseline of antenna array.In this paper, spatial spectrum estimation algorithm is employed to calculate the directions of microwave radiation signals received by the microwave radiation array systems. A novel spatial spectrum estimation method for passive detection is proposed. By using this method a super resolution-beyond resolution of the hardware system- can be achieved. Furthermore, with the same resolution performance, system complexity and equipment expenditure can be reduced by sparse array arrangement.Based on the analysis of microwave radiation signals characteristic and the atmosphere channel, a transmission model of stealthy targets, atmosphere channel and microwave radiation detecting system is established, which can show the distribution of stealthy targets’ and the background’s brightness temperature. The validity of this model is proved by the apropos simulations and the experiments.The intrinsic relationships of those common spatial spectrum estimation algorithms are discussed. And the performances of some algorithms are analyzed and simulated. By applying spatial spectrum estimation algorithms to the sparse array system, the problem of DOA (direction of arrival) ambiguity which results from array manifold ambiguity is discussed. Based on that, a novel algorithm is proposed to calculate the manifold dimension of a linear array arranged by arbitrarily integral multiple of half-wavelength, which can easily find out the manifold dimension and all the possible ambiguous angles due to the impinging of multi-signals. This method can be used to resolve the problem of the DOA ambiguity by removing some sensors of certain positions in the array. Furthermore, a model based on spatial spectrum estimation and the radiation signals is proposed to detect stealthy targets. The performances of this model, including sensitivity and space resolution, are analyzed and comparisons are made between our model and others employing other signal processing methods. The results demonstrate that our model is superior to others in the aspect of resolution and even a super-resolution could be achieved.In the practical scenarios, there are many inevitable errors which would deteriorate the performance of spatial spectrum estimation algorithms. Therefore, error calibration algorithms must be investigated to model the practical environments. A brief review of various errors models and errors calibration algorithms is first presented. Considering the weakness of radiation signals’ energy, a new array errors model for low SNR scenario is proposed, then two calibration methods based on this model are further presented. In the situation of low SNR and in the presence of practical errors, the proposed array errors model and the calibration methods can facilitate the application of spatial spectrum estimation algorithms in the microwave radiation array systems effectively, which is proved by the simulations and the experiments.更多还原