【作者】 李洪亮；

【导师】 张乃通；

【作者基本信息】 哈尔滨工业大学 ， 通信与信息系统， 2008， 硕士

【摘要】 随着人们对于信息量的需求迅猛的增长,信息获取的方式也出现较大的变化,由原来的接触式向非接触式发展。从原来单纯的各种物理量等数据的采集,逐步发展到利用电磁信号本身来实现对周围环境的感知。由于IR-UWB(Impulse Radio Ultra Wideband)在获取和承载信息方面具有很多优势,所以针对各种应用背景的基于IR-UWB技术的研究也拓展开来,在成像技术方面显得尤为突出。电磁散射是一种物理现象,在非均匀场的前提下,入射波被散射后,空间中各点的总场可以表示为原来的入射场和散射后形成的散射场共同作用的结果。当散射体的几何和物理特性已知时,电磁散射正问题就是确定散射域的问题。然而,逆散射问题是从远场模式的信息推断出非均匀场的信息。也就是说:散射波的采集在距散射体很远的位置完成。古典的电磁逆散射理论提出了大量的具有挑战性的数学问题。首先,这是由于在试图建立求解过程中,对散射体的发生的物理条件的不确定性所决定的。其次,用Hadamard的话来说,几乎所有的逆散射问题尤其是有重要应用意义的,都存在所谓的不适应问题。因此,要想得到可靠的近似解,必须面对(在某一阶段)解的唯一性和数值稳定性。利用UWB信号成像可以获得高质量的解。但是,采用宽带或者超宽带激励信号的成像算法(收发分置的或者收发同置的)通常涉及到动态特性。特别值得关注的是,任何的解决逆散射问题的方法都必须(在某一阶段)引入一定的正则化过程,以便消除由不适应问题产生的虚假波动,这一点是毋庸置疑的。本文介绍了两种基于IR-UWB信号的成像算法:DAS(时延求和算法)和LSM(线性采样算法)。并在此基础上提出了修正的线性采样算法(MSM),以及成功地使用Morozov’s离散原理选择最优正则化参数α。更多还原

【Abstract】 Along with the rushing grow up of the people’s need for information, the way and method of we get the information is not as usual when we use touching; now we use non-touching, i.e. we got the information form the data such as physical properties before, but now we get it form the electromagnetic signal itself which apperceives the ambience. Researches on many applications which base on IR-UWB (Impulse Radio Ultra Wideband) has been developed due to IR-UWB has great superiority on the information capture and carrier, from which the imaging technology stands out.Electromagnetic scattering is a kind of physical phenomenon, in the presence of an in-homogeneity, an electromagnetic incident wave is scattered and the total field at any point in space can be written as the sum of the original incident field and the scattered field.The electromagnetic direct scattering problems are the problem of determining the scattered field when the geometrical and physical properties of the scatter are known. The inverse scattering problem is the problem of inferring information on the in-homogeneity from knowledge of the far-field pattern, i.e. the scattered wave at large distances from the scatter. Classical electromagnetic inverse scattering theory provides a rich supply of challenging mathematical problems. First of all, this is due to the fact that the unknown physical conditions where the scattering takes place create difficulties in trying to construct the solution procedure. Second, all the inverse scattering problems significant in applications belong to the class of so-called ill-posed problems in the sense of Hadamard and therefore any reliable approach to their solution must face, at some stage, questions of uniqueness and numerical stability. Using the UWB signal imagery, may obtain the higher solution. Imaging methods (bistatic or monostatic) base on broadband or UWB stimulation signals are usually referred to as migrations.In particular, it is well established that any numerical implementation of a method for the solution of an inverse scattering problem must, at some step, incorporate a regularization procedure in order to eliminate the artificial oscillations due to the ill-posed of the problem.In this paper, introduce two kinds of method base on IR-UWB: DAS (Delay-And-Sum) and LSM (Linear Sampling Method).Then introduce a modifying method of LSM (MSM). Moreover, we use the Morozov’s discrete principle to choice the parameter of regularization.更多还原