多状态,多阈值神经网络模型的光电混合实现

提出了一种采用高分辨率液晶电视（ＬＣＴＶ）实现Ｈｏｐｆｉｅｌｄ神经网络多值算法的光电系统，文章给出了平面多状态，多阈值的全互连Ｈｏｐｆｉｅｄ神经网络模型，并采用该系统对颜色进行了联想和记忆的实验，初步的实验结果可以证实，此种高分辨率液晶电视神经网络系统是可行的。     

一种改进型HOPFIELD神经网络模型及其单通道光学实现

提出一种改进型HOPFIELD神经网络模型。通过对存贮模式进行互补扩展,消除了存贮模式中0和1个数不等问题。利用扩展模式互补性和由扩展模式所形成的连接权的镜象对称性,在不增加神经元个数和连接权矩阵维数的情况下,提高了网络的存贮能力和容错能力。在此基础上设计了全正光学连接权矩阵,在单通道内实现了双极寻址,降低了光学系统的复杂性。     

光折变多量子阱光学寻址空间光调制器的理论分析

利用瞬态二维输运模型,给出了半绝缘多量子阱光学寻址空间光调制器在纵向场几何的理论分析.建立了空间电荷场分量的偏微分方程和边值方程,并通过数值方法进行了求解.在推导方程中考虑了双极、各向异性输运和带边共振激发等因素.结果表明,在小光栅间距下横向场显著影响体电荷的分布,体电荷的分布效应又强烈影响器件的分辨率和时间响应. 关键词：     

液晶光阀用于光学傅里叶变换

结合我校开设的液晶光阀实验，介绍了几个典型的静态傅里叶变换实验内容，以液晶光阀在计算全息扫描器中的应用为例，对其在动态傅里叶变换中的应用进行了讨论。     

仅用激发神经元的IPA神经网络模型及其光学实现

仅用激发神经元的IPA(Interpattern Association)型神经网络模型与既有激发又有抑制的IPA模型具有相似的性能。仅用激发神经元后,不需要光强的相减,这样可以较简单地实现全光神经网络系统,对相似的存储模式有较强的分辨力。此文提出了一套二维(8×8)光学神经网络实验系统,用透镜阵列实现互连,并给出了理论描述和光学实验结果。     

Hartline侧抑制神经网络模型的光学实现

Hartline模型是从神经生理实验得出的普遍适用于神经系统对信息处理的一种方式的模型.本文提出了一种光电混合结构实现Hartline模型的方法:其中非相干光学卷积器实现所要求的卷积运算,用偏振编码的方法实现权重系数的编码,而减法运算仍由电子学完成.最后用二维图象边缘增强的光学实验验证了该方法的有效性.     

时间离散细胞神经网络及其光学实现

提出一种并行图像处理递归算法－时间离散细胞神经网络，基于常规微分方程比较法则设计“填空”这一图像处理任务模板参数，利用空间面积码和离焦多重成像的光电混合系统实施此算法。     

用二值空间光调制器实现多灰阶全息存储

为了利用二值空间光调制器实现多灰阶全息存储,将待存储数据页分解成一系列等价二值数据页,并将记录时间分割成与二值数据页对应的一系列递减时间段,每个时间段使用二值空间光调制器加载相应的二值数据页与参考光在存储介质内部干涉记录一系列全息图。从理论上证明了再现所得数据页的灰阶分布与待存储的数据页一致,并从实验上验证了用二值空间光调制器实现四灰阶存储的可行性。多灰阶全息存储提高了每幅数据页的信息量,更加充分地利用了存储介质的动态范围,提高了存储密度与读出数据传输速率。     

高准确度LCOS自适应光学成像系统的研究

设计了一种液晶自适应光学成像系统.系统中高准确度硅基板液晶(LCOS)器件被用来校正波前位相.LCOS的优势在于它具有几十万个校正单元,而传统的变形镜最多只有1 000左右的校正单元,这种驱动单元的优势使LCOS拥有更高的校正准确度.系统中哈特曼波前传感器被用来测量波前畸变.对点光源进行闭环校正实验中,CCD相机获取了一个高清晰光点图像.实验中波前校正准确度为PV=0.08 λ,rms=0.015 λ(λ=632.8 nm),系统接近衍射极限.     

正补态光学神经网络模型性能评估

本文对一种多值光学神经网络模型——正补态模型的性能,存贮容量、容错性及收敛性进行了评估,并在N=16的网络上进行了计算机模拟.结果表明,正补态模型的性能比其它光学神经网络模型有所改善.     

Optical implementation of (3, 3, 2) regular rectangular CC-Banyan optical network

CC-Banyan network plays an important role in the optical interconnection network. Based on previous reports of (2, 2, 3) the CC-Banyan network, another rectangular-Banyan network, i.e. (3, 3, 2) rectangular CC-Banyan network, has been discussed. First, according to its construction principle, the topological graph and the routing rule of (3, 3, 2) rectangular CC-Banyan network have been proposed. Then, the optically experimental setup of (3, 3, 2) rectangular CC-Banyan network has been designed and achieved. Each stage of node switch consists of phase spatial light modulator (PSLM) and polarizing beam-splitter (PBS), and fiber has been used to perform connection between adjacent stages. PBS features that s-component (perpendicular to the incident plane) of the incident light beam is reflected, and p-component (parallel to the incident plane) passes through it. According to switching logic, under the control of external electrical signals, PSLM functions to control routing paths of the signal beams, i.e. the polarization of each optical signal is rotated or not rotated 90° by a programmable PSLM. Finally, the discussion and analysis show that the experimental setup designed here can realize many functions such as optical signal switch and permutation. It has advantages of large number of input/output-ports, compact in structure, and low energy loss. Hence, the experimental setup can be used in optical communication and optical information processing.     

Cell culture device using spatial light modulator

Spatial light modulator is introduced for cell culturing and related illumination experiment. Two kinds of designs were used. The first type put the cell along with the bio-medium directly on top of the analyzer of the microdisplay and set a cover glass on it to retain the medium environment, which turned the microdisplay into a bio-container. The second type introduced an optical lens system placed below the spatial light modulator to focus the light spots on specific position. Details of the advantages and drawbacks for the two different approaches are discussed, and the human melanocyte cell (HMC) is introduced to prove the feasibility of the concept. Results indicate that the second type is much more suitable than the first for precision required application.     

局域互联神经网络的关联存储

基于全局互联的Hopfield模型,本文提出了局域互联关联存储的新概念.与全局互联相比,局域互联具有较小的关联矩阵,因而,有利于用现有的空间光调制器加以实现.同时,计算机模拟结果表明,它仍然具有全局关联存储的能力.     

Controllable optical multi-well trap and its optical lattices using compounded cosine patterns

This paper proposes a flexible scheme to form various optical multi-well traps for cold atoms or molecules by using a simple optical system composed of an compounded amplitude cosine-only grating and a single lens illuminated by a plane light wave or a Gaussian beam.Dynamic manipulation and evolution of multi-well trap can be easily implemented by controlling the modulation frequency of the cosine patterns.It also discusses how to expand this multi-well trap to two-dimensional lattices with single-or multi-well trap by using an orthogonally or non-orthogonally modulated grating,as well as using incoherent multi-beam illumination,and these results show that all the symmetric structures of two-dimensional Bravais lattices can be obtained facilely by using proposed scheme.     

基于混合光模式阵列的自由空间编码通信

光学涡旋的产生、传输与应用是当前光学领域的研究热点之一.光学涡旋具有轨道角动量,作为一种全新的自由度,丰富了目前光通信的方式.利用面向目标的共轭对称延拓傅里叶计算全息技术,基于空间光调制器,用单束激光直接产生混合光模式阵列进行编码通信.采用由单光涡和复合光涡构成的4种易于识别的模式组成2×2混合光模式阵列,进行灰度图像的编码传输.在接收端提取混合光模式阵列图的信息并进行解码,实现零误码的灰度图像再现.以传输一幅Lena图像为例,使用2×2混合光模式阵列进行编码通信,相对于传统单光涡编码通信,其信息容量可增加4倍.该方法光路简单易行,可扩展性强,进一步拓展使用4×4混合光模式阵列进行编码通信,信息容量提升16倍.提出的混合光模式阵列编码通信方法对于提高信息传输容量具有重要价值.     

Matter-wave interference in an axial triple-well optical dipole trap

This paper proposes a scheme of axial triple-well optical dipole trap by employing a simple optical system composed of a circular cosine grating and a lens. Three optical wells separated averagely by ～37 μm were created when illuminating by a YAG laser with power 1 mW. These wells with average trapping depth ～0.5 μK and volume ～74 μm³ are suitable to trap and manipulate an atomic Bose--Einstein condensation (BEC). Due to a controllable grating implemented by a spatial light modulator, an evolution between a triple-well trap and a single-well one is achievable by adjusting the height of potential barrier between adjacent wells. Based on this novel triple-well potentials, the loading and splitting of BEC, as well as the interference between three freely expanding BECs, are also numerically stimulated within the framework of mean-field treatment. By fitting three cosine functions with three Gaussian envelopes to interference fringe, the information of relative phases among three condensates is extracted.     

Pixel size and pitch measurements of liquid crystal spatial light modulator by optical diffraction

We present a simple technique for the determination of pixel size and pitch of liquid crystal (LC) based spatial light modulator (SLM). The proposed method is based on optical diffraction from pixelated LC panel that has been modeled as a two-dimensional array of rectangular apertures. A novel yet simple, two-plane measurement technique is implemented to circumvent the difficulty in absolute distance measurement. Experimental results are presented for electrically addressed twisted nematic LC-SLM removed from the display projector.     

一种适于光学实现的非线性神经网络模型及其蒙特卡洛学习算法

本文用计算机仿真研究了一种适于光学实现的非线性神经网络模型的存储客量α_c和寻址能力,提出了一个改进其触突互联矩阵的蒙特卡洛学习算法.数值研究表明,经过学习修正后的神经网络模型的寻址能力及存储容量都有较大的改进.     

用空间光调制器产生三维光阱阵列

本文提出了用液晶空间光调制器制作复合相位光栅、产生三维光阱阵列的新方案．在本方案中,首先将一维矩形光栅转变为能够产生纵向光阱阵列的环形光栅,再把环形光栅和二维矩形光栅组合成复合光栅．根据现有空间光调制器的技术参数,模拟仿真设计了产生5×5×5光阱阵列的光栅,以普通功率的高斯光波为输入光,正透镜聚焦衍射光,计算输出光强分布,结果表明：在透镜焦点附近获得具有很高峰值光强和光强梯度的三维光阱阵列,囚禁冷原子的光学偶极势达到mK量级,对原子的作用力远大于原子的重力．用大功率激光作为输入光波时,产生的光阱阵列也能用于囚禁Stark减速后的冷分子．