• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.383-386
• /
• 2009

In this paper, we propose some parallel structures of the word-based nonlinear combine functions in word-based stream cipher, high-speed versions of general (bit-based) nonlinear combine functions. Especially, we propose the high-speed structures of popular three kinds in word-based nonlinear combiners using by PS-WFSR (Parallel-Shifting or Parallel-Structured Word-based FSR): m-parallel word-based nonlinear combiner without memory, m-parallel word-based nonlinear combiner with memories, and m-parallel word-based nonlinear filter function. Finally, we analyze its cryptographic security and performance.

• Journal of the Korean Data and Information Science Society
• /
• v.12 no.2
• /
• pp.95-102
• /
• 2001

최적의 SVM 가중치를 선택하는 방법 중에서 메모리와 속도의 문제를 해결하는 방법 중 하나가 커널애더트론 방법(Kernel Adatron, KA)이다. 본 연구에서는 KA방법을 제곱무감각손실함수까지 확장을 한 알고리즘을 개발한다. 그리고 추정해야 될 라그랑제 배수(Lagrange multiplier)의 수를 반으로 줄이는 알고리즘을 제시한다. 그리고 제시된 알고리즘의 효율성을 여러 모의실험을 통해서 입증한다.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.12
• /
• pp.149-158
• /
• 2012

Proposed is a hybrid randomizing function using two asymptotically optimal randomizing functions: Elias function and Peres function. Randomizing function is an mathematical abstraction of producing a uniform random bits from a source of randomness with bias. It is known that the output rate of Elias function and Peres function approaches to the information-theoretic upper bound. Especially, for each fixed input length, Elias function is optimal. However, its computation is relatively complicated and depends on input lengths. On the contrary, Peres function is defined by a simple recursion. So its computation is much simpler, uniform over the input lengths, and runs on a small footprint. In view of this tradeoff between computational complexity and output efficiency, we propose a hybrid randomizing function that has strengths of the two randomizing functions and analyze it.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.19 no.2
• /
• pp.121-129
• /
• 2008

In this paper, we present a fast analysis of multilayer microstrip fractal structure by using the fast multipole method (FMM). In the analysis, accurate spatial green's functions from the real-axis integration method(RAIM) are employed to solve the mixed potential integral equation(MPIE) with FMM algorithm. MoM's iteration and memory requirement is $O(N^2)$ in case of calculation using the green function. the problem is the unknown number N can be extremely large for calculation of large scale objects and high accuracy. To improve these problem is fast algorithm FMM. FMM use the addition theorem of green function. So, it reduce the complexity of a matrix-vector multiplication and reduce the cost of calculation to the order of $O(N^{1.5})$, The efficiency is proved from comparing calculation results of the moment method and Fast algorithm.

• Journal of Internet Computing and Services
• /
• v.7 no.1
• /
• pp.39-47
• /
• 2006

The evaluation of MDx family hash functions such as MD5 is difficult because the design background or a generalized model is unknown. In this paper, an approximated model is proposed to generalize hash functions. The diffusion of a input difference is tested by an approximated model for MD5. The results show that MD5 does not provide perfect diffusion, so MD5 is weak against some attacks. We propose a multiple chaining pattern which provides perfect diffusion in approximated model of hash function without extra calculation or memory. And We show the probability of differential characteristics of our proposal.

• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.8
• /
• pp.445-455
• /
• 2002

In this paper, we present a new task scheduling scheme ior the efficient execution of the tasks of compound nodes of hierarchical task graph(HTG) on shared memory system. The proposed scheme for exploitation functional parallelism is autoscheduling that performs the role of scheduling by processor itself without any dedicated global scheduler. To adapt the proposed scheduling scheme for various platforms, Including a uni-processor systems, Java threads were used for implementation, and the performance is analyzed in comparison with a conventional bit vector method. The experimental results showed that the proposed method was found to be more efficient in its execution time and exhibited good load-balancing when using the experimental parameter values. Furthermore, the memory size could be reduced when using the proposed algorithm compared with a conventional scheme.

• Journal of Broadcast Engineering
• /
• v.9 no.4 s.25
• /
• pp.384-390
• /
• 2004

In this paper we compare the performance of 4 different algorithms for converting the sampling frequency of an audio from 44.1KHz to 48KHz. The algorithms considered here include the basic polyphase method. sine function based method. multi-stage method. and B-spline based method. For a fair comparison, the sampling rate converters using the 4 algorithms are redesigned under a high fidelity condition. Then, their H/W complexities are compared in terms of the computational complexity and the memory size. As a result, it is shown that the basic polyphase method and sine function based method outperform the other two in terms of the computational complexity, while the B-spline based method requires less memory than the others.

• Annual Conference of KIPS
• /
• 2009.04a
• /
• pp.635-638
• /
• 2009

컴퓨터 디스플레이를 위한 한글 표현 방법에는 한글 음절의 초성, 중성, 종성 각각에 코드를 부여하고, 이를 조합하여 1 음절을 2 바이트로 처리하는 표준 조합형 코드와 각 음절마다 2 바이트 코드를 부여하는 표준 완성형 코드, 그리고 모든 국가의 언어를 표현하기 위한 만국 공통의 문자부호 체계인 유니코드 방식이 있다. 일반적으로 임베디드 시스템은 PC에 비해 상대적으로 속도가 느리고 저장공간 또한 제한되어 있으나 그 용도에 따라 PC에 필적하는 성능을 가지는 경우도 있다. 따라서 화면에 한글을 출력할 경우, 임베디드 시스템의 자원 환경에 맞는 적합한 방식을 채택해야 한다. 본 논문은 시랩시스(CLabSys)사의 3.5" TFT LCD 키트인 LP35가 부착된 LN2440SBC 임베디드 보드(S3C2440A CPU, 400MHz)의 TFT LCD 드라이버 제작을 위한 초기화 방법과 픽셀 디스플레이 함수를 소개한다. 또한 픽셀 디스플레이 함수와 비트맵 폰트를 사용하여 표준 조합형, 표준 완성형, 유니코드 방식의 3가지 방식에 대한 한글 출력 처리 속도와 필요 메모리 용량을 비교한다. 시험 결과에 따르면, 표준 조합형은 적은 메모리 공간을 차지하지만 초성, 중성, 종성을 조합하는데 시간이 소요되고, 완성형은 조합형에 비해 출력 처리 속도는 빠르나 모든 음절의 저장을 위해 비트맵 폰트의 용량이 크며, 유니코드는 비트맵 폰트의 용량은 가장 크지만, 국가 간 언어의 호환성을 보장하고 출력 속도 또한 세 가지 방식 중 가장 빠른 것으로 나타났다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1649-1654
• /
• 2021

In this paper, we introduce a regularization of long short-term memory (LSTM) based fall detection system using TensorFlow that can detect falls that can occur in the elderly. Fall detection uses data from a 3-axis acceleration sensor attached to the body of an elderly person and learns about a total of 7 behavior patterns, each of which is a pattern that occurs in daily life, and the remaining 3 are patterns for falls. During training, a normalization process is performed to effectively reduce the loss function, and the normalization performs a maximum-minimum normalization for data and a L2 regularization for the loss function. The optimal regularization conditions of LSTM using several falling parameters obtained from the 3-axis accelerometer is explained. When normalization and regularization rate λ for sum vector magnitude (SVM) are 127 and 0.00015, respectively, the best sensitivity, specificity, and accuracy are 98.4, 94.8, and 96.9%, respectively.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.26 no.2
• /
• pp.315-318
• /
• 2016

In this paper, we present a new preimage attack on MJH, a double-block-length block cipher-based hash function. Currently, the best attack requires $O(2^{3n/2})$ queries for the 2n-bit MJH hash function based on an n-bit block cipher, while our attack requires $O(n2^n)$ queries and the same amount of memory, significantly improving the query complexity compared to the existing attack.