二维解卷积波束形成水下高分辨三维声成像

针对水下三维成像的空间分辨率难以提高,且具有较高旁瓣级的问题,提出了一种二维解卷积波束形成高分辨三维声成像算法,该算法首先完成任意距离切片的平面阵波束形成,近场情况下采用菲涅尔近似实现近场平面阵波束形成,然后通过二维解卷积技术对任意距离切片的二维波束形成结果进行解卷积处理,去除阵列指向性函数的影响,改善波束响应非理想冲击函数所造成波束形成主瓣宽及旁瓣级高的问题。通过计算机仿真分析,新算法可以有效的提高水下三维成像的空间分辨率,抑制旁瓣级,并能够在较宽频带和不同阵列孔径内保持与常规波束形成相当的稳定性。通过试验研究,新方法比常规波束形成实际目标成像分辨率提高一倍,最高旁瓣级下降20 dB,验证了该算法在实际系统中的有效性.      

渐进式频率-波数域全聚焦超声成像

为了提高全聚焦超声成像的质量和速度,提出了渐进式频率波数域全聚焦超声成像方法（GTFM-FW）。首先通过逐次激励超声相控阵各阵元并全孔径采集回波,得到全矩阵数据;然后通过傅里叶变换将全矩阵数据中的一组或多组对角线数据变换到频率波数域,对各组对角线数据的变换结果进行对齐并叠加;最后利用Stolt变换和二维逆傅里叶变换得到全聚焦成像结果。为了验证算法的有效性,对仿真点目标以及试块进行了详细成像实验。结果表明:本文方法可以通过改变全矩阵数据的使用量来实现成像分辨率与旁瓣干扰间的折中。并且由于采用频率波数域处理,在成像分辨率优于传统全聚焦成像算法的情况下,运算量大大降低。      

合成孔径激光成像雷达的二维傅里叶变换成像算法

提出了合成孔径激光成像雷达(SAIL)的二维傅里叶变换成像算法,即对回波信号进行顺轨向相位二次项共轭补偿后直接实施二维傅里叶变换。归纳了啁啾光源侧视SAIL,平移二次项波面直视SAIL和偏转平面波面直视SAIL的数据收集方程,采用连续变量和函数说明了算法的成像过程,并分析了矩形和圆形天线孔径下的成像分辨率,最后给出了离散傅里叶变换的表达形式。算法中交轨向和顺轨向的时间域数据均直接变换到频率域成像,给出了圆形孔径天线SAIL的随交轨向变化的顺轨向成像分辨率的解析解。     

宽带波束域相干信号子空间高分辨方位估计

提出了宽带波束域相干信号子空间目标高分辨方位估计的频域处理与时域处理方法。分别设计出一组覆盖观察扇面的频域和时域低旁瓣恒定主瓣响应宽带波束形成器,然后对频域或时域波束形成器输出数据运用MUSIC方法估计目标方位。时域处理方法不需要进行频带分解,相对于频域处理需要较少的缓存和运算量。所提出的方法还可以在保证波束主瓣响应恒定的同时,通过在旁瓣区域形成期望宽度和深度的凹槽,用于抑制观察扇面外强干扰源。计算机仿真结果表明,本文的时域处理方法能够获得与本文频域处理方法相近的方位分辨与估计性能,它们都能够分辨相干信号源,且方位分辨与估计性能都高于已有方法。通过在干扰方位设计凹槽,较好地抑制了观察扇面外强干扰源,改善了对相干的弱信号源的方位分辨与估计性能。     

三维大型任意阵列宽带波束指向性图的快速计算

为解决常规频域直接方法计算大型三维阵列宽带波束图(BP)速度较慢,无法满足大型三维宽带阵优化设计需求的问题,提出了一种适用于大型三维任意阵列的宽带BP图的快速算法.在利用二维非均匀快速傅里叶变换(NUFFT)伴随操作实现二维阵列BP频域快速计算方法的基础上,引入第三空间频率变量,将与阵元纵向坐标有关的相位因子转化为第三...     

利用宽带声场频率-掠射角干涉结构的深海直达声区目标深度估计方法

针对典型深海环境中宽带声源的深度分辨问题,通过研究深海声场随频率起伏的干涉结构与垂直线阵频域波束输出图中的干涉结构,给出一种直达声区内可区分多水下目标的宽带声源深度估计方法。该方法以近水面目标的射线声场模型为基础,推导出近海面宽带声源接收声场的波束输出表达式,阐明了频域波束输出图中干涉结构与声源深度的对应关系。然后利用改进的傅里叶变换方法将二维频域波束输出图映射到声源深度-掠射角度域,可实现声源深度信息的有效分离。最后开展了深海实验验证,利用垂直阵接收拖曳声源发射的宽带白噪声信号,拖曳声源深度计算结果与实测声源深度基本一致。数值仿真与实验结果均表明该方法可以在多目标复杂环境下准确估计出水下宽带声源的深度。      

宽带低旁瓣时域波束形成

针对有源和无源声呐中的宽带波束形成,提出了一种低旁瓣时域实现方法。声阵列各个阵元的输出首先经过数字延迟线实现数倍采样间隔的时延补偿,然后由FIR数字滤波器来补偿剩余的时延,并实现低旁瓣波束所需的频率加权,最后再把滤波器输出相加得到时域波束输出。FIR数字滤波器的设计采用改进的自适应方法,其期望响应由工作频带划分成的等间隔窄子带上的低旁瓣波束形成权系数构成。各个子带上的低旁瓣波束则采用基于MVDR波束形成原理的波束优化设计方法得到。针对圆弧阵的计算机仿真设计结果和湖上实验验证了本文方法的正确性和有效性。     

基于子带分解的宽带波束域最小方差无畸变响应高分辨方位估计方法研究

针对最小方差无畸变响应法对宽带不适用、计算量较大和如何与实际声呐系统结合这三个问题,提出了基于子带分解的宽带波束域MVDR高分辨方位估计方法:它以子带分解的方法实现MVDR在宽带情形下的扩展;利用波束转换实现降维运算,大大减小运算量和提高算法的稳健性;并且提出了在声呐系统使用中的新思路:可将该方法与声呐的多波束系统结合,直接利用声呐多波束系统的输出,可省去从阵元域到波束域的转换,进一步减少运算量;可用于替代传统的分裂波束精测定向系统或单波束系统,提高整个声呐定向系统的性能。经过理论分析、仿真实验和湖试实验,证明了该方法是简单的、有效的、可行的。     

运动目标时空频信号变换与多维信息关联

提出了一种针对水中匀速直线运动目标的时空频信号变换方法。首先对接收信号作时频分析和频域波束形成,根据选定的目标航向对频率-方位-时间三维信号空间中的接收信号进行由基阵坐标系到目标坐标系的视在方位变换,使目标线谱信号由三维空间轨迹变换为变换域中对应平面切片上的二维轨迹,该切片由频率与方位信息联合构成的轴和时间轴构成。研究结果表明,该方法能够在变换后的信号空间中将目标线谱信号的三维轨迹转换为相应切片上的二维轨迹,相比三维信号空间更容易获得多维信息联合处理的信号处理增益。     

改进粒子群算法的立体传声器阵列声成像系统阵形优化设计

提出改进的粒子群优化算法,获得波束方向图主瓣宽度和旁瓣级折中的优化立体阵形,避免基本实数粒子群算法仅采用旁瓣级或主瓣宽度一个性能指标优化而导致另一个性能指标恶化的问题,利用阵列视角限制进一步优化立体阵形并设计了声成像测量系统。改进的粒子群算法与基本粒子群算法仿真优化阵形比较表明改进粒子群算法设计的优化阵形在保持较窄的主瓣宽度的条件下具有较低的旁瓣级。阵列声成像测量系统的性能测量分析结果表明该系统的空间分辨率和旁瓣抑制能力与理论结果接近,验证了所提算法的有效性。     

基于分数阶傅里叶变换的水下目标宽带回波精确提取

线性调频信号可能将成为水下目标强度宽带测试的一种主要信号形式。本文在分析水中目标回声构成特点的基础上,探讨了一种滤除界面反射精确提取目标回波的方法。利用线性调频信号在一定分数阶傅立叶域呈现高度聚集的特性,将线性调频信号作为收发合置换能器的发射信号,在适当的分数阶傅立叶域中,目标回波和界面反射波如果表现为两个可分离的谱峰,就可以用窄带滤波的方法去除界面反射。本文给出了算法仿真及水池试验结果,即使对于和线性调频回波相距很近的界面反射,仍然可以有效去除,精确提取出目标回波。     

Fourier domain optical coherence tomography using optical demultiplexers imaging at 60,000,000 lines/s

We describe high-speed Fourier domain optical coherence tomography (OCT) using optical demultiplexers (ODs) for spectral dispersion. The OD enables separation of a narrow spectral band of 14 GHz (0.11 nm) from a broadband incident light at 256 different frequencies in 25.0 GHz intervals centered at 192.2 THz (1559.8 nm). OCT imaging of 60,000,000 axial scans per second was achieved through parallel signal acquisition using 256 balanced photoreceivers to simultaneously detect all the output signals from the ODs in a Fourier domain OCT system. OCT imaging at a 16 kHz frame rate, 1100 A-lines per frame, 3 mm depth range, and 23 microm resolution was demonstrated using a resonant scanner for lateral scanning.     

Imaging through scattering medium by recording 3D “spatial-frequential” interferograms

Speed acquisition for image formation process through scattering medium is a challenge in optical coherence tomography (OCT) approach. Besides time domain (TD), spectral Fourier domain (FD) is now widely studied. By using a swept laser source, we demonstrate that a particular time domain OCT method (optical SISAM correlator) can be simultaneously implemented in a single set-up with the corresponding Fourier domain OCT approach (spectral interferometry). Then, FD-OCT and TD-OCT signals are obtained by processing a 3D “spatial-frequential” interferences pattern. We show that these two numerical approaches can be complementary when imaging in scattering medium is achieved.     

Optical color image hiding scheme by using Gerchberg–Saxton algorithm in fractional Fourier domain

We proposed an optical color image hiding algorithm based on Gerchberg–Saxton retrieval algorithm in fractional Fourier domain. The RGB components of the color image are converted into a scrambled image by using 3D Arnold transform before the hiding operation simultaneously and these changed images are regarded as the amplitude of fractional Fourier spectrum. Subsequently the unknown phase functions in fractional Fourier domain are calculated by the retrieval algorithm, in which the host RBG components are the part of amplitude of the input functions. The 3D Arnold transform is performed with different parameters to enhance the security of the hiding and extracting algorithm. Some numerical simulations are made to test the validity and capability of the proposed color hiding encryption algorithm.     

A fast wave superposition spectral method with complex radius vector combined with two-dimensional fast Fourier transform algorithm for acoustic radiation of axisymmetric bodies

Based on the two-dimensional fast Fourier transform (2D FFT) algorithm, a wave superposition spectral method with complex radius vector has been proposed to efficiently analyze the acoustic radiation from an axisymmetric body. First, the complex Fourier series are used along both circumferential and meridian directions, to expand the integral kernel function and unknown source strength density distributed function. Then, by means of the rectangular integral formula, the radiation sound pressure is described in the form of two-dimensional discrete Fourier transform and generalized through 2D FFT algorithm. Finally, several numerical examples are performed to verify the accuracy and efficiency of the present method. Comparing with the other existing analysis ways, the present method has three different characteristics: (1) there is no singular integral in the numerical computation; (2) the unique solution can be given for all eigen wavenumbers owing to the application of the virtual boundary technology with complex radius vector; and (3) the computational efficiency is improved remarkably because all Fourier terms are calculated simultaneously through 2D FFT algorithm.     

Three-dimensional weight-accumulation algorithm for generating multiple excitation spots in fast optical stimulation

We report here an algorithm for calculating a hologram to be employed in a high-access speed microscope for observing sensory-driven synaptic activity across all inputs to single living neurons in an intact cerebral cortex. The system is based on holographic multi-beam generation using a two-dimensional phase-only spatial light modulator to excite multiple locations in three dimensions with a single hologram. The hologram was calculated with a three-dimensional weighted iterative Fourier transform method using the Ewald sphere restriction to increase the calculation speed. Our algorithm achieved good uniformity of three dimensionally generated excitation spots; the standard deviation of the spot intensities was reduced by a factor of two compared with a conventional algorithm.     

Tomographic implementation of astronomical speckle imaging from bispectra

A bispectral method for astronomical speckle imaging utilizes an average speckle bispectrum of an object to derive its Fourier phase. There has been, however, a problem in conventional bispectral algorithm owing to difficulty in processing bispectral data in a four-dimensional (4D) space. In this paper, we propose an implementation to overcome this problem, where a one-dimensional (1D) object projection is reconstructed from a two-dimensional (2D) average bispectrum of speckle projections, and object projections so obtained at various angles are then tomographically combined into a 2D object image. In this tomographic approach, processes are separable into those for individual projection angles, implying that bispectral data required to be stored at a time are from 4D to 2D and computation time can be substantially reduced by parallelizing angle-by-angle processes. We have performed experiments using simulated and observed data, and have demonstrated the feasibility of the present approach with an achievable accuracy comparable to that of a conventional approach.     

An efficient broadband coupled-mode model using the Hamiltonian method for modal solutions

A numerically efficient broadband, range-dependent propagation model is proposed, which incorporates the Hamiltonian method into the coupled-mode model DGMCM. The Hamiltonian method is highly efficient for finding broadband eigenvalues, and DGMCM is an accurate model for range-dependent propagation in the frequency domain. Consequently, the proposed broadband model combining the Hamiltonian method and DGMCM has significant virtue in terms of both efficiency and accuracy. Numerical simulations are also provided. The numerical results indicate that the proposed model has a better performance over the broadband model using the Fourier synthesis and COUPLE, while retaining the same level of accuracy.     

分数傅里叶域中二维复图像相位恢复的混合输入输出算法

通过推广经典傅里叶域内的混合输入输出算法，研究分数傅里叶域内二维复图像的相位恢复问题。给出算法的具体实施过程，并进行了数值模拟。结果表明：对于级次在0与2之间的分数傅里叶变换，当级次在1/2与3/2之间时，该算法较好；对于其余的级次，该算法的效果不能令人满意。     

Physically Realizable Broadband Acoustic Metamaterials with Anisotropic Density

Transformation acoustics are concentrated for the purpose of designing novel acoustic devices to tailor acoustic waves to achieve desirable characteristics. However, these devices require fluid or fluid-like materials with an anisotropic density that generally does not exist in nature. Therefore, we introduce pentamode metamaterials into an alternating multilayer isotropic medium model to build fluid-like metamaterials with an anisotropic density. A 2D acoustic bending based on transformation acoustics is established and investigated to verify our method. This idea provides a method to design broadband and physically realizable acoustic metamaterials with an anisotropic density and is meaningful for the design of acoustic metamaterials.