• Journal of KIISE:Software and Applications
• /
• v.27 no.11
• /
• pp.1073-1081
• /
• 2000

본 연구에서는 효과적인 시계열 예측을 위하여 구간연산 기능을 갖는 신경망모델을 제안한다. 이는 기존의 FIR 네트워크의 동작특성을 구간연산으로 일반화함으로써 시계열 신호의 동적 특성을 효과적으로 반영할 뿐만 아니라 학습데이타 표현의 유연성을 증대시킨다. 이는 또한 실제 응용에서 신경망 입력 데이타에 내재할 수 있는 측정치 오류의 영향을 보완할 수 있게 하며 데이타 그룹화를 효과적으로 이룰 수 있게 함으로써 입력데이타의 양을 감축시키고 이로부터 학습의 효율을 개선한다. 본 논문에서는 구간연산을 이용하여 네트워크 동작특성, 학습알고리즘을 제시하고 실제 응용시스템에 이를 적용함으로써 제안된 이론의 유용성을 평가한다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.10 no.1
• /
• pp.77-87
• /
• 2000

타원곡선 이산대수 문제에 기초한 공개키 암호시스템에서 타원곡선 멱승은 반드시 필요한 연산이며 연산들 중에서 가장 복잡도가 크다. 따라서 효율적인 암호시스템 구현을 위해서는 타원곡선 멱승연산을 효율적으로 구현하는 것이 중요하다. 본 논문에서는 복합 좌표계(mixed coordinate system)를 이용한 멱승 방법을 GF(pm)상에서 정의되는 타원 곡선을 적용하여 최적의 효율성을 갖는 타원곡선 멱승 구현법을 제안한다. 또한 ‘곱셈을 이용한 역원 연산 알고리즘(IM; Inversion with Multiplication)’을 이용하여 더욱 효율적인 구현이 가능함을 보인다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.10 no.1
• /
• pp.89-97
• /
• 2000

공개키 암호 시스템 중에서 가장 널리 사용되는 RSA 암호 시스템은 키의 분배와 권리가 용이하고, 디지털 서명이 가능한 장점이 있으나, 암호화와 복호화 과정에서 512 비트 이상의 큰 수에 대한 멱승과 모듈라 감소 연산이 요구되기 때문에 처리 속도의 지연이 큰 문제가 되므로 모듈라 멱승 연산의 고속 처리가 필수적이다. 따라서 본 논문에서는 몫을 추정하여 중간 곱의 크기를 제한하는 interleaved 모듈라 곱셈 기법을 이용하여 모듈라 멱승 연산을 수행하는 고속 RSA 암호 칩을 VHDL을 이용하여 모델링하고 Faraday FG7000A 라이브러리를 이용하여 합성하고 타이밍 검증하여 단일 칩 IC로 구현하였다. 구현된 암호 칩은 75,000 게이트 수준으로 합성되었으며, 동작 주파수는 50MHz이고 1회의 RSA 연산을 수행하는데 소요되는 전체 클럭 사이클은 0.25M이며 512비트 당 처리 속도는 102.4Kbit/s였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.49-51
• /
• 2022

대규모 데이터를 처리하기 위해 머신러닝, 딥러닝과 같은 AI 활용 연구가 일반화되면서 시스템 환경 또한 병렬처리 연산에 강화된 가속기 기반의 이기종 아키텍처로 확산되고 있다. CPU 기반의 계산 환경과 달리 상대적으로 성능이 낮은 수천 개의 산술연산장치(ALU)를 활용해 쓰레드 방식으로 연산을 수행하며, I/O의 특성 또한 대규모의 데이터들이 수많은 연산장치에 전달되기 위한 Small I/O, High-throughput 처리 성능이 애플리케이션에 큰 영향을 끼친다. 본 논문에서는 병렬 컴퓨팅 환경에 AI 애플리케이션이 접목되면서 요구되는 스토리지, 파일시스템의 환경을 분석하고 나아가 성능 검증을 통해 I/O 특성을 파악하고자 한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.796-799
• /
• 2006

This paper presents a new method for automatic detection of skew in a document image using mathematical morphology. To speed up processing, we use reduced image but it still requires long time to estimate the skew angle so the proposed method works with region of interest, not with whole image. Character strings are connected by using morphological closing operation and a component labeling is used to select region of interest. The method considers the lowermost pixels of characters in candidate regions in the binary image of original document image. Experimental results shows that the proposed method is extremely fast and robust as well as independent of script forms.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.3
• /
• pp.714-720
• /
• 2004

In detecting human faces in color images, additional geometric computation is often necessary for validating the face-candidate regions having various forms. In this paper, we propose a method that detects the facial features using chrominance components of color which do not affected by face occlusion and orientation. The proposed algorithm uses the property that the Cb and Cr components have consistent differences around the facial features, especially eye-area. We designed the Mean-Gray Morphology operator to emphasize the feature areas in the eye-map image which generated by basic chrominance differences. Experimental results show that this method can detect the facial features under various face candidate regions effectively.

• KIPS Transactions on Computer and Communication Systems
• /
• v.5 no.8
• /
• pp.181-188
• /
• 2016

Recently, as the amount of spatial information increases, an interest in the study of spatial information processing has been increased. Spatial database systems extended from the traditional relational database systems are difficult to handle large data sets because of the scalability. SpatialHadoop extended from Hadoop system has a low performance, because spatial computations in SpationHadoop require a lot of write operations of intermediate results to the disk, resulting in the performance degradation. In this paper, Spatial Computation Spark(SC-Spark) is proposed, which is an in-memory based distributed processing framework. SC-Spark is extended from Spark in order to efficiently perform the spatial operation for large-scale data. In addition, SC-Spark based on the GPGPU is developed to improve the performance of the SC-Spark. SC-Spark uses the advantage of the Spark holding intermediate results in the memory. And GPGPU-based SC-Spark can perform spatial operations in parallel using a plurality of processing elements of an GPU. To verify the proposed work, experiments on a single AMD system were performed using SC-Spark and GPGPU-based SC-Spark for Point-in-Polygon and spatial join operation. The experimental results showed that the performance of SC-Spark and GPGPU-based SC-Spark were up-to 8 times faster than SpatialHadoop.

• Journal of Korea Spatial Information System Society
• /
• v.1 no.1 s.1
• /
• pp.49-55
• /
• 1999

Extracting information of maps is necessary to establish the GIS. In this paper, a house recognition method on the scanned topographic map is described. A contour detection method is used to extract houses from the scanned maps and RLE (run-length encoding) method is used for manipulating houses touching grid lines. To handle houses touched to roads and borderlines, morphological operation is used. To remove misrecognition occurred by morphological operation, the legions which contain characters on the map are also automatically eliminated.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.1
• /
• pp.91-95
• /
• 2008

Flash memory is widely used in embedded systems because of its benefits such as non-volatile, shock resistant, and low power consumption. But NAND flash memory suffers from out-place-update, limited erase cycles, and page based read/write operations. To solve these problems, log-structured filesystem was proposed such as YAFFS. However, YAFFS sequentially retrieves an array of all block information to allocate free block for a write operation. Also before the write operation, YAFPS read the array of block information to find invalid block for erase. These could reduce the performance of the filesystem. This paper suggests fast operation method for NAND flash filesystem that solves the above-mentioned problems. We implemented the proposed methods in YAFFS. And we measured the performance compared with the original technique.

• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.3
• /
• pp.134-142
• /
• 2002

Quadtree, which is a hierarchical data structure, is a very important data structure to represent binary images. The linear quadtree representation as a way to store a quadtree is efficient to save space compared with other representations. Therefore, it has been widely studied to develop efficient algorithms to execute operations related with quadtrees. The window operation is one of important geometry operations in image processing, which extracts a sub-image indicated by a window in the image. In this paper, we present an algorithm to perform the window operation of binary images represented by quadtrees, using three-dimensional $n{\times}n{\times}n$ processors on RMESH(Reconfigurable MESH). This algorithm has constant-time complexity by using efficient basic operations to route the locational codes of quardtree on the hierarchical structure of $n{\times}n{\times}n$ RMESH.