이상준 교수

With the evolution of semiconducting industries, thermomechanical failure induced in a multilayered structure with a high aspect ratio during manufacturing and operation has become one of the critical reliability issues. In this work, the effect of thermomechanical stress on the failure of multilayered thin films on Si substrates was studied using analytical calculations and various thermomechanical tests. The residual stress induced during material processing was calculated based on plate bendi

Abstract Colorimetric sensors that respond to mechanical stimuli have garnered increased attention for various applications such as healthcare, display, electronic skin, and camouflage. To date, these mechanochromic sensors have been largely fabricated and studied with periodic structures such as surface relief patterns and photonic crystals. Although periodic structures can provide high wavelength selectivity, the strong viewing‐angle‐dependence of the color hinders their practical application

A scan architecture is the most widely used for obtaining high test coverage in manufacturing tests. However, the recent increase in circuit size has caused the power consumption during scan testing to become higher than that in the functional mode. Thus, the reliability of scan testing has become a concern. In this brief, a new logic topology-based scan chain stitching method is proposed to reduce the test power. The proposed method uses the topology of logic circuits to analyze them without re

We address the problem of efficient similarity search based on the minimum distance in large time series databases. To support minimum distance queries, most of previous work has to take the preprocessing step of vertical shifting. However, the vertical shifting has an additional overhead in building index. In this paper, we propose a novel dimensionality reduction technique for indexing time series based on the minimum distance. We call our approach the SSV-indexing (Segmented Sum of Variation

이 연구의 목적은 DEA 및 DEA Window를 이용하여 국내 프로축구단의 정태적 효율성과 동태적 효율성을 측정하여 구단의 경영효율성을 제고하는 데 있다. 2017년부터 2020년까지 K리그 1에 잔류한 8개 프로축구단을 선정하였으며, 이 중에서 기업 구단은 FC서울, 수원삼성, 울산현대, 전북현대, 포항스틸러스이고, 시․도민 구단은 강원FC, 대구FC, 인천Utd다. 상대적 효율성 측정을 위해 투입변수로는 매출원가, 산출변수로는 매출액과 정규시즌 승점으로 하였고, R 프로그램을 활용하여 데이터를 처리하였다. DEA를 이용한 특정시점의 정태적 효율성 분석결과 첫째, CCR 모형을 적용할 때 가장 효율적인 구단은 대구FC로 나타났다. 둘째, BCC 모형을 적용할 때 가장 효율적인 구단은 대구FC, 전북현대로 나타났다. 셋째, 비효율성의 주된 원인이 순수기술효율성(PTE) 보다는 규모효율성(SE)에 있는 것으로 분석되었다. 또한 DEA Window를 이용한 동태적 효율성 분석결과 동태

Purpose: To evaluate the classification, diagnosis, and natural course of ophthalmoplegia associated with mitochondrial disease. Materials and Methods: Among 372 patients with mitochondrial disease who visited our hospital between January 2006 and January2016, 21 patients with ophthalmoplegia were retrospectively identified. Inclusion criteria included onset before 20 years of age, pigmentary retinopathy, and cardiac involvement. The 16 patients who were finally included in the study were divide

Abstract As metals form the backbone of many engineering structures including bridges and buildings, the detection of their mechanical deformation is of technological importance. Several optical techniques such as holographic interferometry, speckle interferometry, and digital image correlation have been developed to map the surface strains and displacements of various materials. However, these techniques have stringent requirements for stability and repeatability, which limit their application

Scan-based testing, though widely used in modern digital designs, causes more power consumption than in functional mode. This excessive power consumption can cause severe hazards such as circuit reliability and yield loss. To solve this problem, a new scan chain reordering method based on care bit density is proposed in this paper. This proposed method helps merging care bits toward the front end of scan chains. Thus, it can reduce scan cell switching activities during scan shift operation. Expe

Cryptographic circuits generally are used for applications of wireless sensor networks to ensure security and must be tested in a manufacturing process to guarantee their quality. Therefore, a scan architecture is widely used for testing the circuits in the manufacturing test to improve testability. However, during scan testing, test-power consumption becomes more serious as the number of transistors and the complexity of chips increase. Hence, the scan chain reordering method is widely applied

Artificial intelligence (AI) accelerators that support AI services consist of multiple identical cores for parallel computation to accelerate artificial neural networks. Recently, systolic array-based architectures for low-power AI accelerators have been studied to support battery-operated edge devices and reduce power consumption. In this paper, a zero-overhead method is proposed for efficient testing of systolic array-based low-power AI accelerators. By using the structural characteristics of

Recently, efficient partial relay selection (e-PRS) was proposed as an enhanced version of PRS. In comparing e-PRS, PRS, and the best relay selection (BRS), there is a tradeoff between complexity and performance; that is, the complexity for PRS, e-PRS, and BRS is low to high, respectively, but vice versa for performance. In this paper, we study the outage probability for e-PRS in decode-and-forward (DF) relaying systems over non-identical Nakagami-m fading channels, where the fading parameter m

Recently, efficient partial relay selection (e-PRS) was proposed as an enhanced version of PRS. In comparing e-PRS, PRS, and the best relay selection (BRS), there is a tradeoff between complexity and performance; that is, the complexity for PRS, e-PRS, and BRS is low to high, respectively, but vice versa for performance. In this paper, we study the outage probability for e-PRS in decode-and-forward (DF) relaying systems over non-identical Nakagami-m fading channels, where the fading parameter m

To enable broader deployment of Large Language Models (LLMs), it is essential to identify the best-performing model under strict memory constraints.We present AMQ, Automated Mixed-Precision Weight-Only Quantization, a framework that assigns layer-wise quantization bit-widths to optimally balance model quality and memory usage.However, the combinatorial search space, with over 10 100 possible configurations, makes conventional black-box optimization infeasible.AMQ overcomes this challenge through