雷达阵列综合及关键参数估计算法研究

发布时间：2018-06-28 21:10

本文选题：阵列综合 + 稀疏　；参考：《电子科技大学》2014年硕士论文

【摘要】：雷达诞生至今已接近80年,在此期间各种不同体制的雷达不断涌现,其功能、体积、重量、可靠性以及生存能力等亦发生了相当大的变化。毫无疑问,雷达在国民经济和军事应用领域正扮演着越来越重要的角色。本文主要就雷达信号处理中的阵列综合以及参数估计这两项技术展开了研究。全文分为三个部分:第一部分就稀疏阵列综合展开讨论。首先,介绍了一种基于连续凸优化的稀疏阵列综合方法,其对方向图的功率进行约束,然后将阵列优化问题转换为二阶锥规划(Second Order Cone Program,SOCP)问题并使用SeDuMi进行求解。之后在上述方法模型的基础上,提出了一种基于加权l1范数的稀疏阵列综合方法,在整个观测角度上对波形进行约束,并采用复数求导结合启发式近似方法来求解。将仿真结果与已有的结论相比较,该方法可以得到孔径更短,稀疏程度更高的阵列。第二部分详细阐述了基于稀疏表示的波达方向(Direction Of Arrival,DOA)估计方法。首先简单介绍了三种常见的谱估计方法并指出它们的局限性。然后,通过引入过完备集,将阵列接收模型转化为稀疏表示的DOA估计问题,在此基础上详细介绍了l1-SVD算法。最后,引入一个酉变换矩阵,通过该变换矩阵将上述基于稀疏表示的复接收信号模型转换为实数模型,大幅度降低了l1-SVD算法的计算复杂度。第三部分在多输入多输出(Multiple-Input Multiple-Output,MIMO)雷达体制下讨论了稳健的参数估计方法。首先对MIMO雷达系统进行建模,同时将Capon方法和APES方法在MIMO雷达体制下进行了推导并分析了其性能的优劣;之后考虑雷达阵列部分校准的情况,提出了一种可以对目标的幅度、方位以及阵列的扰动进行准确估计的方法,同时分析了信噪比(Signal to Noise Ratio,SNR)对该方法性能的影响。最后考虑阵列完全失配的情况,首先介绍了一种RCB方法并对其存在的局限性进行了分析。然后在RCB方法的基础上提出了一种IRCB方法,该方法无需事先获知流形矢量的不确定程度,其性能与流形矢量的不确定水平确定已知时的RCB方法相近。
[Abstract]:Radar has been born for nearly 80 years. During this period, various kinds of radars of different systems have been emerging, and their functions, volume, weight, reliability and survivability have also changed a lot. There is no doubt that radar is playing an increasingly important role in national economy and military applications. In this paper, array synthesis and parameter estimation in radar signal processing are studied. The thesis is divided into three parts: the first part is about sparse array synthesis. Firstly, a sparse array synthesis method based on continuous convex optimization is introduced, in which the power of the opposite direction graph is constrained, and then the array optimization problem is transformed into the second order cone programming (SOCP) problem and solved by SeDuMi. Then a sparse array synthesis method based on weighted L 1 norm is proposed based on the above method model. The waveform is constrained from the whole observation angle and solved by complex derivation combined with heuristic approximation. Comparing the simulation results with the existing results, the proposed method can obtain arrays with shorter aperture and higher sparsity. In the second part, the direction of arrival (DOA) estimation method based on sparse representation is described in detail. Firstly, three common spectral estimation methods are introduced and their limitations are pointed out. Then, by introducing over-complete sets, the array reception model is transformed into a sparse representation DOA estimation problem, and then the l1-SVD algorithm is introduced in detail. Finally, a unitary transformation matrix is introduced, by which the complex received signal model based on sparse representation is transformed into a real number model, which greatly reduces the computational complexity of the l1-SVD algorithm. In the third part, the robust parameter estimation method is discussed in the Multiple-Input Multiple-Output MIMO (MIMO) radar system. Firstly, the MIMO radar system is modeled. At the same time, the Capon method and the apes method are deduced and analyzed under MIMO radar system. The influence of signal to noise ratio (SNR) on the performance of the method is analyzed. Finally, a RCB method is introduced and its limitations are analyzed. Then an IRCB method is proposed based on the RCB method. This method does not need to know the degree of uncertainty of the manifold vector in advance, and its performance is similar to that of the RCB method when the uncertainty level of the manifold vector is known.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN957.51

【参考文献】