• Journal of KIISE
• /
• v.42 no.5
• /
• pp.558-565
• /
• 2015

Recently, the number of new malware and malware variants has dramatically increased. As a result, the time for analyzing malware and the efforts of malware analyzers have also increased. Therefore, malware classification helps malware analyzers decrease the overhead of malware analysis, and the classification is useful in studying the malware's genealogy. In this paper, we proposed a set of key opcode to classify the malware. In our experiments, we selected the top 10-opcode as key opcode, and the key opcode decreased the training time of a Supervised learning algorithm by 91% with preserving classification accuracy.

• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.245-248
• /
• 2001

본 논문에서는 연집 오류(burst error)에 우수한 정정 능력을 보이는 고속 RS(Reed-Solomon) 복호기를 제안한다. 제안된 RS 복호기는 RS(n, k, t); (37 < n ≤ 255, 21 < k ≤ 239, t = 8)의 사양을 지원하며 수정 유클리드 알고리즘(modified Euclid´s algorithm)을 이용한 시스톨릭 어레이(systolic array) 방식의 병렬처리 구조로 설계되었다. 고속 RS 복호기의 효율적인 VSLI 설계를 위하여 새로운 방식의 수정 유클리드 알고리즘 연간 회로를 제안한다. 제안된 수정 유클리드 알고리즘 회로는 2t + 1의 연산 지연 시간을 갖으며 기존 구조의 연산 지연 시간인 3t + 37에 비하여 t = 8 인 경우 약 72%의 연산 지연이 감소하였다. 제안된 구조를 VHDL을 이용하여 설계하였으며 SAMSUNG 0.5㎛(KG80) 라이브러리를 이용하여 논리 합성과 타이밍 검증을 수행하였다. 합성된 RS 복호기의 총 게이트 수는 약 77,000 개이며 최대 80MHz의 동작 속도를 나타내었다.

• Proceedings of the KAIS Fall Conference
• /
• 2011.12b
• /
• pp.441-444
• /
• 2011

본 논문은 2D/3D 영상 처리에서 깊이지도 생성을 위한 Optical flow에서 레벨을 간소화하여 연산량을 감소시키고 객체의 고유벡터를 이용하여 영상의 잡음을 제거하는 연구이다. Optical flow는 움직임추정 알고리즘의 하나로 두 프레임간의 픽셀의 변화 벡터 값을 나타내며 블록 매칭과 같은 알고리즘에 비해 정확도가 높다. 그러나 기존의 Optical flow는 긴 연산 시간과 카메라의 이동이나 조명의 변화에 민감한 문제가 있다. 이를 해결하기 위해 연산 시간의 단축을 위한 레벨 간소화 과정을 거치고 영상에서 고유벡터를 갖는 영역에 한해 Optical flow를 적용하여 잡음을 제거하는 방법을 제안하였다. 제안한 방법으로 2차원 영상을 3차원 입체 영상으로 변환하였고 SSIM(Structural SIMilarity Index)으로 최종 생성된 영상의 오차율을 분석하였다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.04a
• /
• pp.315-318
• /
• 2004

최적해 탐색 도구로 널리 알려진 유전 알고리즘을 이용하여 신경망의 학습을 위한 가중치를 탐색하는 방법은 신경망의 학습 방법의 하나로 사용되고 있다. 신경망의 가중치는 일정 시간의 유전자 연산을 수행하게 되면 최적화된 가중치의 값과 유사하게 되는 특징을 지닌다. 이는 유전자 연산 방법에 의해 가중치가 수렴되고 있음을 의미하며, 그 때의 가중치는 일정한 패턴을 지니는 특징을 발견할 수 있다. 이에, 본 논문에서는 탐색된 가중치의 패턴을 보존하기 위한 방법으로 유전자의 일정 부분을 고정한 후 유전자 연산을 수행하는 개선된 학습 방법을 제안하고자 한다. 이를 이용할 경우에 유전자 탐색의 문제점으로 제시되고 있는 탐색 시간을 효율적으로 감소시킬 수 있는 장점이 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10c
• /
• pp.871-873
• /
• 2001

본 논문에서는 전기 단층 촬영 기법과 같이 많은 양의 데이터에 대해 산술 계산을 수행하는 응용의 수행속도를 개선하기 위하여 이중 CPU PC 상에서 Matlab의 기본연산, 즉 행렬 곱하기, 역행렬 계산, 의사 역행렬 계산 등을 병렬로 수행하는 라이브러리 프로그램을 구현하고 그 성능을 측정한다. 구현된 라이브러리는 행렬의 곱하기, 역행렬 계산, 의사 역행렬 계산 등 기본적인 행렬 연산에 대해 각 CPU에서 수행될 쓰레드를 생성하고 이 쓰레드에 분할 행렬을 인자로 넘겨줌으로써 병렬 계산을 실행하도록 하고 부분 결과를 합성하여 최종적인 결과를 산출하게 된다. 구현된 코드를 수행시켜 속도를 측정한 결과 행렬의 곱하기는 최대 69%, 역행렬은 34.8 %, 의사 역행렬은 52 % 까지 수행시간을 단축시켰다. 이에 의해 전기 단층 촬영 프로그램은 한번의 전류 주입에 대해 영상 복원에 소요되는 시간을 48 %로 감소시켰다.

• Journal of Broadcast Engineering
• /
• v.23 no.4
• /
• pp.559-573
• /
• 2018

Recently, one of the most actively applied image media in the most fields such as virtual reality (VR) is omni-directional or panorama image. This image is generated by stitching images obtained by various methods. In this process, it takes the most time to extract keypoints necessary for stitching. In this paper, we analyze the parameters involved in the extraction of SIFT keypoints with the aim of reducing the computation time for extracting the most widely used SIFT keypoints. The parameters considered in this paper are the initial standard deviation of the Gaussian kernel used for Gaussian filtering, the number of gaussian difference image sets for extracting local extrema, and the number of octaves. As the SIFT algorithm, the Lowe scheme, the originally proposed one, and the Hess scheme which is a convolution cascade scheme, are considered. First, the effect of each parameter value on the computation time is analyzed, and the effect of each parameter on the stitching performance is analyzed by performing actual stitching experiments. Finally, based on the results of the two analyses, we extract parameter value set that minimize computation time without degrading.

• Journal of IKEEE
• /
• v.21 no.2
• /
• pp.136-141
• /
• 2017

In this paper, we propose an algorithm for memory optimization to apply deep learning to PC. The proposed algorithm minimizes the memory and computation processing time by reducing the amount of computation processing and data required in the conventional deep learning structure in a general PC. The algorithm proposed in this paper consists of three steps: a convolution layer configuration process using a random filter with discriminating power, a data reduction process using PCA, and a CNN structure creation using SVM. The learning process is not necessary in the convolution layer construction process using the discriminating random filter, thereby shortening the learning time of the overall deep learning. PCA reduces the amount of memory and computation throughput. The creation of the CNN structure using SVM maximizes the effect of reducing the amount of memory and computational throughput required. In order to evaluate the performance of the proposed algorithm, we experimented with Yale University's Extended Yale B face database. The results show that the algorithm proposed in this paper has a similar performance recognition rate compared with the existing CNN algorithm. And it was confirmed to be excellent. Based on the algorithm proposed in this paper, it is expected that a deep learning algorithm with many data and computation processes can be implemented in a general PC.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.9
• /
• pp.617-625
• /
• 2009

NAND flash memory is widely used in embedded systems because of many attractive features, such as small size, light weight, low power consumption and fast access speed. However, it requires garbage collection, which includes erase operations. Erase operation is very slow. Besides, the number of the erase operations allowed to be carried out for each block is limited. The proposed garbage collection method focuses on minimizing the total number of erase operations, the deviation value of each block and the garbage collection time. NAND flash memory consists of pages of three types, such as valid pages, invalid pages and free pages. In order to achieve above goals, we use a page rate to decide when to do garbage collection and to select the target victim block. Additionally, We implement allocating method and group management method. Simulation results show that the proposed policy performs better than Greedy or CAT with the maximum rate at 82% of reduction in the deviation value of erase operation and 75% reduction in garbage collection time.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.2
• /
• pp.223-233
• /
• 2023

Gauss Sieve is an algorithm for solving SVP and requires exponential time and space complexity. The terminationcondition of the Sieve is determined by the size of the constructed list and the number of collisions related to space complexity. The term 'collision' refers to the state in which the sampled vector is reduced to the vector that is already inthe list. if collisions occur more than a certain number of times, the algorithm terminates. When executing previous algorithms, we noticed that unnecessary operations continued even after the shortest vector was found. This means that the existing termination condition is set larger than necessary. In this paper, after identifying the point where unnecessary operations are repeated, optimization is performed on the number of operations required. The tests are conducted by adjusting the threshold of the collision that becomes the termination condition and the distribution in whichthe sample vector is generated. According to the experiments, the operation that occupies the largest proportion decreased by62.6%. The space and time complexity also decreased by 4.3 and 1.6%, respectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.11A
• /
• pp.1124-1131
• /
• 2008

In this paper, we propose a timing synchronization scheme and its hardware architecture of the next generation IEEE 802.11n wireless LAN standard which is based on MIMO-OFDM technique. Proposed timing synchronization method takes two steps which consist of two modified auto-correlators. For coarse timing synchronization, a sliding window differentiator is used after a conventional auto-correlation in order to avoid plateau problem. The conjugate symmetry property of L-LTS is utilized for the simplification of fine timing synchronization. Since cross-correlation based methods are not required, the computational complexity and the number of multipliers can be reduced. In order to reduce the hardware complexity, we have used sign multipliers. Based on simulation results, the proposed method outperforms a conventional method. The proposed scheme can be applied to IEEE 802.11n systems and can easily be expanded to frequency synchronization scheme.