• JSTS:Journal of Semiconductor Technology and Science
• /
• 제16권2호
• /
• pp.251-254
• /
• 2016

An energy-efficient object matching accelerator is proposed for mobile object recognition based on matching prediction scheme. Conventionally, vocabulary tree has been used to save the external memory bandwidth in object matching process but involved massive internal memory transactions to examine each object in a database. In this paper, a novel object matching accelerator is proposed based on matching predictions to reduce unnecessary internal memory transactions by mitigating non-target object examinations, thereby improving the energy-efficiency. Experimental results show a 26% reduction in power-delay product compared to the prior art.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2006년도 하계종합학술대회
• /
• pp.3-4
• /
• 2006

Fast IP address lookup in routers is essential to achieve packet forwarding in wire-speed. The longest prefix matching for IP address lookup is more complex than exact matching because of its dual dimensions, length and value. By thoroughly studying the current proposals for IP address lookup, we find out that the binary search could be a low-cost solution while providing high performance. Most of the existing binary search algorithms based on trie have simple data structures which can be easily implemented, but they have some problems because of empty internal nodes. The proposed algorithm is based on trie structure, but empty internal nodes are replaced by priority prefixes. The best-matching-prefix search in the proposed algorithm is more efficiently performed since search can be finished earlier when input is matched with a priority prefix. The performance evaluation results show that the constructed priority-trie has very good performance in the lookup speed and the scalability.

• 정보보호학회논문지
• /
• 제18권3호
• /
• pp.79-88
• /
• 2008

인터넷 환경에서 내부 네트워크를 보호하기 위하여 침입탐지시스템이 광범위하게 사용되고 있다. 침입탐지시스템은 비정상 패킷의 특성을 분석하여 규칙을 생성하고 이 규칙들을 이용하여 패킷들을 필터링함으로써 내부 시스템들을 보호한다. 최근 공격 사례가 많아지고, 공격 형태가 구조화되면서 이를 탐지하는 규칙의 수도 지속적으로 증가하고 있다. 이에 따라 침입탐지시스템이 규칙을 적용하는 과정에서의 성능 하락 정도도 커지고 있다. 본 논문은 규칙을 적용하는 과정에서 상대적으로 오버헤드가 큰 문자열 검색 성능을 개선하고자 복수개의 부분패턴을 이용한 다중 패턴매칭 기법을 제안한다. 그리고 대표적인 고성능의 다중 패턴매칭 알고리즘인 Wu-Manber 알고리즘과 성능을 비교하고 그 결과를 보인다.

• 동의생리병리학회지
• /
• 제24권1호
• /
• pp.165-170
• /
• 2010

This study assessed effects of magnetic acu-point therapy as an add-on to conventional antihypertensive managements such as medication or lifestyle modification for hypertensive or pre-hypertensive subjects. This study was conducted with the purpose to evaluate the decreasing effect of the blood pressure on the hypertensive patients who were given to magnetic acu-point therapy. A single-blind, placebo-controlled trial by cross matching was conducted at Daejeon University Oriental Hospital. The washout periods of cross matching are 10 days. 22 hypertensive (systolic BP>140 mmHg or diastolic BP>90 mmHg) volunteers were recruited and treatments were donemagnetic acu-points on Oegwan(TE 5), Jogimeup(GB 41), Naegwan(PC 6), Gongson(SP 4), Hugye(SI 3), Sinmaek(BL 62). The treated group was statistically significant decrease of the blood pressure, while control group show no statistically significant difference of BP. From the results, magnetic acu-point therapy seems to offer a benefit to the treatment of hypertensive patients.

• 한국결정학회:학술대회논문집
• /
• 한국결정학회 2002년도 정기총회 및 추계학술연구발표회
• /
• pp.18-18
• /
• 2002

Although it was suggested in 1962 that an efficient wavelength conversion could be achieved using ferroelectric crystals of periodic 180° domains, it was not until 1990's that quasi-phase-matching (QPM) became realized, as technology for periodic poling of LiNbO₃ crystals was readily available. Since ferroelectric domain inversion brings about change of the sign of second-order nonlinear susceptibility, periodically poled ferroelectric structures provide an ideal way of achieving QPM for second-harmonic generation and optical parametric oscillation. Periodically poled ferroelectric domains can also be utilized for optical devices, such as Brags electrooptic modulators. fabrication of stable periodic domain structures depends on a number of poling parameters of a ferroelectric crystal, such as coercive field, internal field and electrical conductivity. We present poling kinetics of KNbO₃ crystals, which involve domain nucleation and growth, backswitching, relaxation of internal field. Optimum poling conditions were established by designing a proper wave shape of external field. We demonstrate an efficient second-harmonic generation using QPM in a periodically poled KNbO₃ crystal.

• 한국유체기계학회 논문집
• /
• 제18권6호
• /
• pp.31-36
• /
• 2015

A multi-stage turbocharger system has been constructed for HALE UAV internal combustion engine. To boost rarefied intake air up to sea level condition, the turbocharger system should consist of 3 stages including heat exchanger located after compressor outlet to drop compressed air temperature. One dimensional system analysis has been conducted by matching required power between compressor and turbine and adequate turbochargers have been searched for from commercially available models targeting for automobiles. By applying commercial automobile turbochargers to the multi-stage turbocharger system, it is expected that considerable amount of research resources will be saved.

• ETRI Journal
• /
• 제34권3호
• /
• pp.454-457
• /
• 2012

A miniaturized planar digital video broadcasting terrestrial (DVB-T) antenna, which is composed of a patterned helical line, an open stub, and an impedance matching circuit on an FR4 (${\varepsilon}_r$=4.4) substrate for portable media player applications, is presented in this letter. The antenna has monopole-like, omni-directional radiation characteristics and a wide impedance bandwidth (VSWR<3) in the DVB-T band from 174 MHz to 230 MHz at the VHF band.

• 한국컴퓨터정보학회논문지
• /
• 제28권4호
• /
• pp.31-39
• /
• 2023

본 논문에서는 이종 모델의 특징맵 간 상관관계인 외부적 상관관계와 동종 모델 내부 특징맵 간 상관관계인 내부적 상관관계를 활용하여 교사 모델로부터 학생 모델로 지식을 전이하는 Internal/External Knowledge Distillation (IEKD)를 제안한다. 두 상관관계를 모두 활용하기 위하여 특징맵을 시퀀스 형태로 변환하고, 트랜스포머를 통해 내부적/외부적 상관관계를 고려하여 지식 증류에 적합한 새로운 특징맵을 추출한다. 추출된 특징맵을 증류함으로써 내부적 상관관계와 외부적 상관관계를 함께 학습할 수 있다. 또한 추출된 특징맵을 활용하여 feature matching을 수행함으로써 학생 모델의 정확도 향상을 도모한다. 제안한 지식 증류 방법의 효과를 증명하기 위해, CIFAR-100 데이터 셋에서 "ResNet-32×4/VGG-8" 교사/학생 모델 조합으로 최신 지식 증류 방법보다 향상된 76.23% Top-1 이미지 분류 정확도를 달성하였다.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2010년도 하계학술대회 논문집
• /
• pp.122-122
• /
• 2010

A Static Induction Thyristor (SI-Thyristor) has a great potential as power semiconductor switch for pulsed power or high voltage applications with fast turn-on switching time and high switching stress endurance (di/dt, dV/dt). However, due to direct commutation between gate driver and SI-Thyristor, it is difficult to design optimal gate driver at the aspect of impedance matching for fast gate current driving into internal SI-Thyristor. Thus, to penetrate fast positive gate current into steady off state of the SI-Thyristor, it is proposed and proceeded the internal impedance calculation of the SI-Thyristor at steady off state with the gate driver while switching conditions that are indicated applied gate voltage, $V_{GK}$ and applied high voltage across anode and cathode, $V_{AK}$.

• 전자공학회논문지A
• /
• 제33A권12호
• /
• pp.110-119
• /
• 1996

A novel high speed VLSI architecture and its VLSI realization methodologies for a motion estimation processor based on full search block matching algorithm are presentd. The presented architecture is designed in order to be suitable for highly parallel and pipelined processing with identical PE's and adjustable in performance and hardware amount according to various application areas. Also, the throughput is maximized by enhancing PE utilization up to 100% and the chip pin count is reduced by reusing image data with embedded image memories. Also, the uniform and identical data processing structure of PE's eases VLSI implementation and the clock rate of external I/O data can be made slower compared to internal clock rate to resolve I/O bottleneck problem. The logic and spice simulation results of the proposed architecture are presented. The performances of the proposed architecture are evaluated and compared with other architectures. Finally, the chip layout is shown.