公开(公告)号：KR1020120050313A

公开(公告)日：2012-05-18

申请号：KR1020100111743

申请日：2010-11-10

Applicant: 삼성전자주식회사

Inventor： 류수정 ,  박성배 ,  서웅 ,  조영철 ,  김정욱 ,  정무경 ,  김호영

IPC: G06F9/30 ,  G06T1/60 ,  G06F12/02

CPC classification number: G06T1/60 ,  G06F9/30134 ,  G06F9/30101 ,  G06F12/02

Abstract: PURPOSE: A computing device through an X-Y memory and a method thereof are provided to rapidly process image data without overhead by performing pop or push operation of data. CONSTITUTION: A memory unit(101) defines an address space based on a multidimensional space having two shafts. A memory access unit(102) includes a first pointer register in which a first pointer is stored and a second pointer register in which a second pointer is stored. The memory unit is a stack memory for storing data according to a LIFO(Last-In First-Out) mode. The memory access unit fixes the first pointer and reduces the second pointer.

Abstract translation: 目的：提供通过X-Y存储器的计算设备及其方法，通过执行数据的弹出或推送操作来快速处理图像数据而无需开销。 构成：存储单元（101）基于具有两个轴的多维空间来定义地址空间。 存储器访问单元（102）包括其中存储有第一指针的第一指针寄存器和存储第二指针的第二指针寄存器。 存储器单元是用于根据LIFO（先进先出）模式存储数据的堆栈存储器。 存储器访问单元固定第一指针并减少第二指针。

公开(公告)号：KR1020120001423A

公开(公告)日：2012-01-04

申请号：KR1020100062187

申请日：2010-06-29

Applicant: 삼성전자주식회사

Inventor： 김정욱 ,  최상진 ,  이승황 ,  최일섭 ,  김동억 ,  최광혁 ,  박선우

IPC: B41J2/045 ,  B41J2/05 ,  B41J2/175

CPC classification number: B41J2/04501 ,  B41J2/05 ,  B41J2/175

Abstract: PURPOSE: A sound pressure generation device of an ink-jet printer is provided to prevent the backward flowing of ink from a printer head through a flowing backward prevention filter. CONSTITUTION: A sound pressure generation device of an ink-jet printer comprises a printer head(10), a valve(20), a pump, a reservoir, a sound pressure sensor(50) and a sound pressure control device. The printer head has a plurality of nozzles emitting ink. The valve controls the flow of the sound pressure formed in the printer head. The pump constitutes the sound pressure in the printer head. The reservoir reduces the pulsating of the sound pressure by the pump. The sound pressure sensor measures the sound pressure formed in the printer head. The sound pressure control device controls the sound pressure measured by the sound pressure sensor.

Abstract translation: 目的：提供喷墨打印机的声压产生装置，以防止来自打印机头的墨水通过防逆流过滤器的反向流动。 构成：喷墨打印机的声压产生装置包括打印头（10），阀（20），泵，储存器，声压传感器（50）和声压控制装置。 打印头具有多个喷墨喷嘴。 阀控制打印头中形成的声压的流动。 泵构成打印机头中的声压。 油箱可以减少泵的声压脉动。 声压传感器测量打印头中形成的声压。 声压控制装置控制由声压传感器测量的声压。

公开(公告)号：KR1020090068845A

公开(公告)日：2009-06-29

申请号：KR1020070136629

申请日：2007-12-24

Applicant: 삼성전자주식회사

Inventor： 김정욱 ,  김혁 ,  최상진 ,  정성욱 ,  임은섭 ,  안병일 ,  이중하

IPC: H01L21/66 ,  H01L21/027

CPC classification number: B41J2/125 ,  B41J2/07 ,  H01L21/6715 ,  H01L21/67253 ,  H01L22/12

Abstract: A pattern formation system and a control method using the droplet mass deflection measure system are provided, which shorten the process time for forming pattern and form the pattern which needs the high precision. A droplet mass deflection measure system comprises a plurality of droplets(DR-1), the droplet moving force offering part(230), the discharge solution location perception member(210), and the droplet mass deflection measurement control part. Droplet is discharged in a plurality of droplet discharge units. The droplet moving force offering part individually provides the discharge direction of a plurality of droplets and moving force of the other direction to a plurality of droplets. The discharge solution location perception member acquires droplet location images in which the location of a plurality of droplets is individually reflected. The droplet mass deflection measurement control part measures the mass deflections of a plurality of droplets.

Abstract translation: 提供了一种使用液滴质量偏转测量系统的图案形成系统和控制方法，缩短了形成图案的工艺时间并形成了需要高精度的图案。 液滴质量偏转测量系统包括多个液滴（DR-1），液滴移动力提供部分（230），排出溶液位置感知部件（210）和液滴质量偏转测量控制部分。 液滴在多个液滴喷射单元中排出。 液滴移动力提供部分分别提供多个液滴的排出方向和另一个方向的移动力到多个液滴。 排放溶液位置感知构件获取液滴位置图像，其中多个液滴的位置被单独反射。 液滴质量偏转测量控制部分测量多个液滴的质量偏转。

公开(公告)号：KR100875836B1

公开(公告)日：2008-12-24

申请号：KR1020070028813

申请日：2007-03-23

Applicant: 삼성전자주식회사

Inventor： 백창우 ,  김홍석 ,  김희석 ,  김정욱

IPC: G06F9/06 ,  G06F9/38

CPC classification number: G06F9/3822 ,  G06F9/30156 ,  G06F9/3885

Abstract: An apparatus and a method are provided for a parallel processing very long instruction word (VLIW) computer. The apparatus includes: an index code generation unit sequentially generating an index code, which is associated with a number of no operation (NOP) instruction word between effective instruction words, with respect to each of instruction word groups to be executed in a VLIW computer; an instruction compression unit sequentially deleting the NOP instruction word which corresponds to the index code with respect to each of instruction word groups; and an instruction word conversion unit converting the effective instruction words to include the index code, the effective instruction words corresponding to the NOP instruction words.

Abstract translation: 提供了一种用于并行处理超长指令字（VLIW）计算机的设备和方法。 该设备包括：索引码产生单元，其相对于将在VLIW计算机中执行的每个指令字组，顺序地产生与有效指令字之间的无操作数（NOP）指令字相关联的索引码; 指令压缩单元，相对于每个指令字组，顺序删除与索引码对应的NOP指令字; 以及指令字转换单元，用于转换所述有效指令字以包括所述索引代码，所述有效指令字对应于所述NOP指令字。

公开(公告)号：KR100868451B1

公开(公告)日：2008-11-11

申请号：KR1020070017775

申请日：2007-02-22

Applicant: 삼성전자주식회사

Inventor： 김종면 ,  류수정 ,  유동훈 ,  김홍석 ,  김희석 ,  김정욱 ,  민경준

IPC: G06F12/02 ,  G06F12/08 ,  G06F12/00

CPC classification number: G06F8/443 ,  G06F8/447 ,  G06F12/0207

Abstract: 본 발명은 프로그램 중 3중 루프에 의하여 메모리를 접근하는 코드로부터 a, b, c를 획득하는 단계 - 상기 a는 상기 3중 루프의 가장 안쪽 루프 변수가 가질 수 있는 값의 개수이고, 상기 b는 상기 3중 루프의 가운데 루프 변수가 가질 수 있는 값의 개수이고, 상기 c는 상기 3중 루프의 가장 바깥쪽 루프 변수가 가질 수 있는 값의 개수임 -; 상기 3중 루프에 의하여 접근되는 상기 메모리의 시작 주소를 획득하는 단계; 및 상기 시작 주소 및 소정의 식을 이용하여 상기 3중 루프에 의하여 접근되는 상기 메모리의 a * b * c개의 주소를 획득하는 단계를 포함하는 메모리 접근 방법을 제공한다.
3DAM (3-D Addressable Memory), 3-D 메모리, 2-D 메모리

公开(公告)号：KR1020080095672A

公开(公告)日：2008-10-29

申请号：KR1020070040474

申请日：2007-04-25

Applicant: 삼성전자주식회사

Inventor： 김정욱

IPC: G06F3/12

CPC classification number: H04N1/3877 ,  H04N1/2108

Abstract: An image processing device having an image rotation function and an image rotation method thereof are provided to reduce the transmission time between a memory and a buffer and decrease the size of a hard disk at the same time. A user interface(230) receives a user command to control the operation of an image processing device and transfers the command to a CPU(Central Processing Unit)(220). A memory(210) stores each kinds of control programs necessary for performing an image processing operation, and image data corresponding to an image to be rotated are stored at the memory. The CPU controls the operation of the image processing device by executing each kinds of control programs stored at the memory according to the user command inputted through the user interface. A receiving DMA unit(240) accesses the memory unit directly to read out image data by block unit, then stores the read data at a block buffer(260). A rotation processing unit(270) reads out the image data, which are stored by block unit of first size, by block unit which is smaller than the block unit of first size. A transmitting DMA unit(250) stores the image data stored at the buffer, that is to say the image data into which the image data of first block unit are rotationally-processed, at the memory.

Abstract translation: 提供具有图像旋转功能和图像旋转方法的图像处理装置，以减少存储器和缓冲器之间的传输时间，同时减小硬盘的尺寸。 用户界面（230）接收用户命令来控制图像处理设备的操作，并将命令传送到CPU（中央处理单元）（220）。 存储器（210）存储执行图像处理操作所需的各种控制程序，并且将与要旋转的图像相对应的图像数据存储在存储器中。 CPU根据通过用户界面输入的用户命令，执行存储在存储器中的各种控制程序来控制图像处理装置的操作。 接收DMA单元（240）直接访问存储器单元以通过块单元读出图像数据，然后将读取的数据存储在块缓冲器（260）。 旋转处理单元（270）以小于第一尺寸的块单位的块单位读出以第一尺寸的块为单位存储的图像数据。 发送DMA单元（250）将存储在缓冲器中的图像数据（即将第一块单元的图像数据旋转处理的图像数据）存储在存储器中。

公开(公告)号：KR1020080086739A

公开(公告)日：2008-09-26

申请号：KR1020070028813

申请日：2007-03-23

Applicant: 삼성전자주식회사

Inventor： 백창우 ,  김홍석 ,  김희석 ,  김정욱

IPC: G06F9/06 ,  G06F9/38

CPC classification number: G06F9/3822 ,  G06F9/30156 ,  G06F9/3885

Abstract: An apparatus and a method for compressing instruction words for a parallel processing VLIW(Very Long Instruction Word) computer are provided to reduce an instruction storage space in a parallel processing computer without requiring an additional storage space or additional circuit logic. An apparatus for compressing instruction words includes an index code generator(110), an instruction word compressor(120) and an instruction word converter(130). The index code generator sequentially generates index codes with respect to the number of NOP(No Operation) instruction words among effective instruction words for respective instruction word groups executed in a VLIW computer. The instruction word compressor sequentially deletes the NOP instruction words corresponding to the index codes for the instruction word groups. The instruction word converter embeds the index codes in effective instruction words corresponding to the deleted NOP instruction words.

Abstract translation: 提供用于压缩并行处理VLIW（超长指令字）计算机的指令字的装置和方法，以减少并行处理计算机中的指令存储空间，而不需要额外的存储空间或附加电路逻辑。 用于压缩指令字的装置包括索引码发生器（110），指令字压缩器（120）和指令字转换器（130）。 索引码发生器依次生成关于在VLIW计算机中执行的各指令字组的有效指令字中的NOP（No Operation）指令字数的索引码。 指令字压缩器依次删除与指令字组的索引码对应的NOP指令字。 指令字转换器将索引代码嵌入与删除的NOP指令字对应的有效指令字中。

公开(公告)号：KR100854720B1

公开(公告)日：2008-08-27

申请号：KR1020070028856

申请日：2007-03-23

Applicant: 삼성전자주식회사

Inventor： 김희석 ,  김홍석 ,  백창우 ,  김정욱

IPC: G06F9/06 ,  G06F9/00

CPC classification number: G06F8/4441

Abstract: A method and a device for coalescing loops are provided to coalesce a plurality of loops efficiently, remove branch of a conditional sentence without changing a flow of the program, calculate a guard with a simple operation by calculating the periodically set guard with a new instruction, and operate the loops efficiently by reusing resources of a processor. A loop remover(710) removes the innermost loop by executing an external operation of the innermost loop among a plurality of loops when a condition of a conditional sentence is satisfied. A guard code generator(720) generates a guard code by performing if-conversion of the conditional sentence. A guard converter(730) determines a logical value of a guard based on information related to a cycle corresponding to an iteration frequency of the innermost loop by using an instruction for determining the logical value of the guard code. An operation executor performs the external operation of the innermost loop based on the determined logical value. The information related to the cycle is stored in a register and includes a plurality of bits. The instruction determines one of the bits as the logical value of the guard.

Abstract translation: 提供了一种用于聚结循环的方法和装置，用于有效地合并多个循环，在不改变程序流程的情况下移除条件句的分支，通过用新指令计算周期性设置的守卫，用简单的操作计算一个保卫， 并通过重新使用处理器的资源来有效地操作循环。 当满足条件语句的条件时，循环去除器（710）通过执行多个循环中的最内循环的外部操作来去除最内循环。 保护码发生器（720）通过执行条件语句的转换来产生保护码。 保护转换器（730）通过使用用于确定保护码的逻辑值的指令，基于与对应于最内循环的迭代频率的周期相关的信息来确定防护装置的逻辑值。 操作执行器基于所确定的逻辑值执行最内循环的外部操作。 与周期相关的信息被存储在寄存器中并且包括多个位。 指令将其中一个位确定为保护的逻辑值。

公开(公告)号：KR100790392B1

公开(公告)日：2008-01-02

申请号：KR1020040092544

申请日：2004-11-12

Applicant: 삼성전자주식회사

Inventor： 권태용 ,  김정욱

IPC: H01L21/3065

CPC classification number: H01J37/32623 ,  H01J37/32935 ,  H01L21/67069

Abstract: 본 발명은 구조를 단순화하면서도 플라즈마 영역을 용이하게 가변시킬 수 있도록 하는 플라즈마 한정장치를 갖춘 반도체 제조장치를 개시한 것이다.
개시한 반도체 제조장치는 반응실 내부의 플라즈마 영역을 한정하는 플라즈마 한정장치를 구비하고, 플라즈마 한정장치는 반응실 내부를 제1플라즈마 영역으로 한정하는 제1한정장치와, 반응실 내부를 제1플라즈마 영역보다 큰 제2플라즈마 영역으로 한정하는 제2한정장치와, 플라즈마 영역의 가변을 위해 제1한정장치와 제2한정장치를 함께 이동시키는 구동장치를 포함하는 것이다.

公开(公告)号：KR100707781B1

公开(公告)日：2007-04-17

申请号：KR1020060005619

申请日：2006-01-19

Applicant: 삼성전자주식회사

Inventor： 유동훈 ,  김희석 ,  김정욱 ,  류수정

IPC: G06F7/00 ,  G06F12/00

Abstract: 본 발명은 루프에 포함되는 배열 변수의 오프셋 및 메모리 뱅크의 계산을 효율적으로 수행하기 위한 방법 및 장치에 관한 것으로, 본 발명은, 루프 연산 중에 포함된 다차원 배열에 대해 소정의(predetermined) 언롤링 인자(unrolling factor)에 따라 루프 언롤링(loop unrolling)을 수행하는 단계, 및 상기 루프 언롤링에 의하여 상기 루프 연산 내에서 언롤된(unrolled) 다차원 배열(multi-dimensional array)을 루프 카운터 변수(loop counter variable)에 대한 어파인 함수(affine function) 형태의 배열 서브스크립트 표현을 가지는 1차원 배열로 변경하는 단계를 포함하는 컴퓨터 시스템에서의 배열 연산 처리 방법을 제공한다.
루프 언롤링(loop unrolling), 메모리 인터리빙(memory interleaving), 배열 서브스크립트 표현(array subscript expression), 재구성 아키텍처(reconfigurable architecture)

Abstract translation: 本发明涉及用于高效地执行包括在循环中的阵列变量和存储体的偏移的计算的方法和装置，并且本发明涉及用于高效地执行存储体的计算的方法和装置， 在循环操作中根据循环计数器和展开的多维数组的展开因子执行循环展开，借助于循环展开作为循环计数器变量 所述二维阵列具有用于计算机系统中的变量的仿射函数形式的阵列下标表达式。