利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

27 条记录 (耗时 0.044 秒)

    Microfluidic sorting device
    24.
    发明授权
    Microfluidic sorting device 有权
    微流控分选装置

    公开(公告)号：US07068874B2

    公开(公告)日：2006-06-27

    申请号：US10848972

    申请日：2004-05-18

    申请人： Mark Wang ,  Erhan Polatkan Ata ,  Sadik C. Esener

    发明人： Mark Wang ,  Erhan Polatkan Ata ,  Sadik C. Esener

    IPC分类号： G02B6/26 ,  H01S3/00 ,  G01N27/26

    CPC分类号： H05H3/04 ,  Y10T428/24744

    摘要： Small particles, for example 5 μm diameter microspheres or cells, within, and moving with, a fluid, normally water, that is flowing within microfluidic channels within a radiation-transparent substrate, typically molded PDMS clear plastic, are selectively manipulated, normally by being pushed with optical pressure forces, with laser light, preferably as arises from VCSELs operating in Laguerre-Gaussian mode, at branching junctions in the microfluidic channels so as to enter into selected downstream branches, thereby realizing particle switching and sorting, including in parallel. Transport of the small particles thus transpires by microfluidics while manipulation in the manner of optical tweezers arises either from pushing due to optical scattering force, or from pulling due to an attractive optical gradient force. Whether pushed or pulled, the particles within the flowing fluid may be optically sensed, and highly-parallel, low-cost, cell- and particle-analysis devices efficiently realized, including as integrated on bio-chips.

    摘要翻译： 选择性地操纵在通常模制的PDMS透明塑料的辐射透明基底内的微流体通道内流动的流体（通常为水）内的微粒，例如5微米直径的微球或细胞，通常是通过 通过激光照射，优选地是以在Laguerre-Gaussian模式下操作的VCSEL产生的激光照射在微流体通道中的分支结处，从而进入选择的下游分支，从而实现包括并行的粒子切换和分选。 因此，通过微流体传输小颗粒，而以光学镊子的方式进行操作会由于光学散射力的推动或由于有吸引力的光学梯度力而引起拉动。 无论是推动还是拉动，流动流体中的颗粒可以被光学感测，并且高度并行，低成本的细胞和颗粒分析装置被有效地实现，包括集成在生物芯片上。

    Optoelectronic associative memory using parallel-readout optical disk storage
    26.
    发明授权
    Optoelectronic associative memory using parallel-readout optical disk storage 失效
    使用并行读出光盘存储的光电关联存储器

    公开(公告)号：US5412592A

    公开(公告)日：1995-05-02

    申请号：US785408

    申请日：1991-10-31

    申请人： Ashok V. Krishnamoorthy ,  Philippe J. Marchand ,  Gokce Yayla ,  Sadik C. Esener

    发明人： Ashok V. Krishnamoorthy ,  Philippe J. Marchand ,  Gokce Yayla ,  Sadik C. Esener

    IPC分类号： G06K9/74 ,  G11B7/0065 ,  G11C13/04 ,  G11C15/00

    CPC分类号： G11C13/042 ,  G06K9/74 ,  G11B7/0065 ,  G11C15/00

    摘要： A high capacity (1 Gbyte), high throughput (1.1 Gbyte/s) motionless-head parallel-readout optical disk, and a detector array integrated on substrate with an Si/PLZT Exclusive-NOR gate array, implement a fast retrieval (25 ms) associative memory/content addressable memory capable of 10.sup.10 bit-operations per second. The disk stores arrayed 1-D holograms, preferably computer-generated from 128.times.128 pixel images by Fourier transform. Reverse transform upon disk readout is by lenses, preferably by a single hybrid refractive/diffractive lens. The reconstituted 2-D image, or optical word, is detected and compared to an electrical data word in an fast integrated optoelectronic circuit. The circuit permits (i) a variably preset "match" detection threshold, and (ii) dynamically variable, sub-image, field size of the search. In a first mode of operation all images (or sub-images) that are sufficiently close to a preset query threshold are retrieved in one disk rotation. In a second mode, Hamming distances for all images (sub-images) are calculated during a first disk rotation, and the identified best match(es) is (are) retrieved on a subsequent rotation. In a third mode a query image can be classified relative to all images upon a disk during one rotation of the disk. In still other modes the associative memory can search for images and sub-images (optical words) of variable size in parallel.

    摘要翻译： 高容量（1Gbyte），高吞吐量（1.1Gbyte / s）静止头并行读出光盘以及集成在具有Si / PLZT异或非门阵列的衬底上的检测器阵列，可实现快速检索（25 ms ）关联存储器/内容可寻址存储器，能够每秒进行1010位操作。 磁盘存储阵列的1-D全息图，优选地通过傅里叶变换由128×128像素图像计算机生成。 磁盘读出时的反向变换是透镜，优选通过单个混合折射/衍射透镜。 检测重构的2-D图像或光学字，并将其与快速集成光电子电路中的电数据字进行比较。 电路允许（i）可变地预设的“匹配”检测阈值，以及（ii）搜索的动态变化的子图像的场大小。 在第一操作模式中，在一个盘旋转中检索足够接近于预设查询阈值的所有图像（或子图像）。 在第二模式中，在第一盘旋转期间计算所有图像（子图像）的汉明距离，并且在随后的旋转中检索所识别的最佳匹配。 在第三种模式中，在磁盘旋转一周期间，查询图像可以相对于磁盘上的所有图像分类。 在其他模式中，关联存储器可以并行地搜索可变大小的图像和子图像（光学词）。

    Artificial neuron with switched-capacitor synapses using analog storage of synaptic weights
    27.
    发明授权
    Artificial neuron with switched-capacitor synapses using analog storage of synaptic weights 失效
    具有开关电容器突触的人造神经元使用突触重量的模拟储存

    公开(公告)号：US5343555A

    公开(公告)日：1994-08-30

    申请号：US909563

    申请日：1992-07-06

    申请人： Gokce Yayla ,  Ashok V. Krishnamoorthy ,  Sadik C. Esener

    发明人： Gokce Yayla ,  Ashok V. Krishnamoorthy ,  Sadik C. Esener

    IPC分类号： G06N3/067 ,  G06F15/18

    CPC分类号： G06N3/0675

    摘要： A pseudo-analog electronic or optoelectronic neuron stores synaptic weights as analog quantities, preferably as charges upon capacitors or upon the gates of floating gate transistors. Multiplication of a stored synaptic weight times a binary pulse-width-modulated synapse input signal periodically produces electrical charge of a first polarity on a first synapse capacitor. Meanwhile a fixed charge of opposite polarity is periodically produced at the same frequency upon another, second, synapse capacitor. The charges on both synapse capacitors at many synapses are periodically accumulated, and integrated, at a single neuron soma in the form of pulse-amplitude-modulated charge-encoded signals. This accumulation, and integration, transpires continuously progressively by a switched-capacitor technique, and during the entire duration of the input signal to each synapse. The net final result, expressed in signed electrical charge, is converted back to a PWM binary signal for transmission to further neurons. A fully capacitive synapse typically occupies a compact area of 45.lambda..times.42.lambda., consumes less than 2 .mu.W dynamic power (at 1 MHz) and offers more than 90% of the full voltage scale for linear weight adaptation. It is therefore well suited to large scale parallel implementations of adaptive neural networks.

    摘要翻译： 伪模拟电子或光电神经元将突触权重存储为模拟量，优选地作为电容器上的电荷或浮置栅极晶体管的栅极。 存储的突触体重乘以二进制脉冲宽度调制突触输入信号的周期性地在第一突触电容器上产生第一极性的电荷。 同时，在另一个第二突触电容器上以相同频率周期性地产生相反极性的固定电荷。 两个突触电容器在许多突触上的电荷被周期性地累积，并以脉冲幅度调制的电荷编码信号的形式在单个神经元细胞集成。 这种积累和整合通过开关电容器技术逐渐发展，并且在输入信号的整个持续时间期间到每个突触。 以带符号的电荷表示的净最终结果被转换回到PWM二进制信号以传输到另外的神经元。 完全电容性突触通常占据45λx42λ的紧凑区域，消耗小于2WW的动态功率（1MHz），并提供用于线性重量适配的全电压标度的90％以上。 因此，它非常适合于适应性神经网络的大规模并行实现。