• 대한전자공학회논문지SD
• /
• 제48권4호
• /
• pp.39-50
• /
• 2011

본 논문에서는 패널 내부 인터페이스의 하나인 2세대 AiPi+의 클록-데이터 복원 회로(Clock & Data Recovery)를 제안하였다. 제안하는 클록-데이터 복원 회로의 속도는 기존 AiPi+ 보다 빠른 1.25 Gbps 로 향상되었으며 다중 위상 클록을 생성하기 위하여 Delay-Locked Loop(DLL)를 사용하였다. 본 논문에서는 패널 내부 인터페이스의 저전력, 작은 면적의 이슈를 만족하는 클록-데이터 복원 회로를 설계하였다. 매우 간단한 방법으로 자동적으로 Harmonic-locking 문제를 해결할 수 있는 주파수 검출기 구조를 제안하여 기존 주파수 검출기(Frequency Detector)의 복잡도, 전류 소모, 그리고 외부 인가에 따른 문제를 개선하였으며, 전압 제어 지연 라인(Voltage Controlled Delay Line) 에서 상승/하강 시간 차이에 따른 에지의 사라짐 현상을 막기 위해서 펄스 폭의 최대치를 제한하는 펄스 폭 오류 보정 방법을 사용하였다. 제안하는 클록-데이터 복원 회로는 CMOS 0.18 ${\mu}m$ 공정으로 제작되었으며 면적은 $660\;{\mu}m\;{\times}\;250\;{\mu}m$이고, 공급 전압은 1.8 V이다. Peak-to-Peak 지터는 15 ps, 입력 버퍼, 이퀄라이저, 병렬화기를 제외한 클록-데이터 복원 회로의 소모 전력은 5.94 mW 이다.

• 예술인문사회 융합 멀티미디어 논문지
• /
• 제7권2호
• /
• pp.199-208
• /
• 2017

제4차 산업혁명의 주축으로 불리우고 있는 IoT, Cloud, Bigdata, Mobile, AI, 3D print 등의 기술들이 군에 접목되었을 경우 큰 변화가 생길것으로 예측할 수 있다. 특히, 전투라는 목적을 생각하였을 경우 그 중 IoT, Cloud, Bigdata, Mobile, AI 이 5가지 기술이 많은 역할을 할 것으로 생각된다. 따라서, 본 논문에서는 이 5가지 기술이 접목된 미래 군의 모습을 Smart Military라고 정의하고, 이에 따른 아키텍처를 정립하고, 적합한 정보보안 모델에 대해서 연구하였다. 이를 위해 IoT, Cloud, Bigdata, Mobile, AI와 관련된 기존 문헌들을 연구하여, 공통적인 요소를 도출하였으며, 이에 따른 아키텍처를 제시하였다. 제시된 아키텍처를 중심으로 Smart Military 환경에서의 정보보안을 전략적 정보보안과 전술적 정보보안으로 구분하고, 상기 구분된 정보보안 형태에 따라 취약성이 존재하더라도 감내할만한 수준의 경우를 고려하여 전략적 측면을 중심으로 정보보호체계를 구축한다면 효율적인 예산범위 내에서 최적의 정보보호 구축이 가능할 것으로 기대된다.

• 한국수정란이식학회지
• /
• 제31권4호
• /
• pp.349-353
• /
• 2016

This study was carried out to evaluate the influencing factors that affect the reproductive performance of cows at the Monirampur upazila in Jessore district of Bangladesh. A total of 224 cows were brought to the upazila livestock hospital for artificial insemination (AI). The cows were inseminated between 12 to 18 hours from the onset of estrus and data was obtained from the owner. Out of 224 cows, 133 became pregnant and 91 were non pregnant. In this study, the overall pregnancy rate was 59.29%. Among the age variability, the highest pregnancy rate (70.27%) was at the age of 4 years old. In case of breed variation, the highest pregnancy rate was observed in local breed (69.07%) compared with other crossbred cows. Hence the breed variations significantly influence the conception rate of cows. According to the parity, we found that the pregnancy rate was increasing with their parity but decreasing after 4th parity. The highest conception rate was observed in 3rd parity (67.74%) which was significantly higher than that of heifers (Parity-0). Here we also found that the types of bull semen used for AI had no significant effect for pregnancy rate. The skills of AI technician for AI to cows were significantly affecting the pregnancy rate. However, this study is not enough for rating and comment about the reproduction performance of cows. Therefore, further extensive study is needed for rating and recommendation for the cattle up gradation at that particular area.

• Genomics & Informatics
• /
• 제21권2호
• /
• pp.20.1-20.10
• /
• 2023

Aromatase inhibitors (AI) are drugs that are widely used in treating estrogen receptor (ER)-positive breast cancer patients. Drug resistance is a major obstacle to aromatase inhibition therapy. There are diverse reasons behind acquired AI resistance. This study aims at identifying the plausible cause of acquired AI resistance in patients administered with non-steroidal AIs (anastrozole and letrozole). We used genomic, transcriptomic, epigenetic, and mutation data of breast invasive carcinoma from The Cancer Genomic Atlas database. The data was then separated into sensitive and resistant sets based on patients' responsiveness to the non-steroidal AIs. A sensitive set of 150 patients and a resistant set of 172 patients were included for the study. These data were collectively analyzed to probe into the factors that might be responsible for AI resistance. We identified 17 differentially regulated genes (DEGs) among the two groups. Then, methylation, mutation, miRNA, copy number variation, and pathway analyses were performed for these DEGs. The top mutated genes (FGFR3, CDKN2A, RNF208, MAPK4, MAPK15, HSD3B1, CRYBB2, CDC20B, TP53TG5, and MAPK8IP3) were predicted. We also identified a key miRNA - hsa-mir-1264 regulating the expression of CDC20B. Pathway analysis revealed HSD3B1 to be involved in estrogen biosynthesis. This study reveals the involvement of key genes that might be associated with the development of AI resistance in ER-positive breast cancers and hence may act as a potential prognostic and diagnostic biomarker for these patients.

• 한국정보전자통신기술학회논문지
• /
• 제16권5호
• /
• pp.279-291
• /
• 2023

최근 몇 년 동안 언어 모델 기반의 생성형 인공지능 기술은 눈에 띄게 발전하고 있다. 특히, 요약, 코드 작성과 같은 다양한 분야에서 활용 가능성이 증가하고 있어 큰 관심을 받고 있다. 이러한 관심의 반영으로, 생성형 인공지능 관련 특허 출원이 급격히 증가하는 추세를 보인다. 이러한 동향을 파악하고 이에 따른 전략을 수립하기 위해 미래 예측이 핵심적이다. 예측을 통해 해당 기술 분야의 미래 동향을 정확히 파악하여 더 효과적인 전략을 수립할 수 있다. 본 논문에서는 언어 모델 기반 생성형 인공지능 발전 방향을 확인하기 위해 현재까지 출원된 특허들을 분석하였다. 특히, 각 국가에서의 연구 및 발명 활동을 깊게 살펴보았으며, 연도별 및 세부 기술별 출원 동향을 중점적으로 분석하였다. 이러한 분석을 통해 핵심 특허들이 포함하고 있는 세부 기술을 이해하고, 향후 생성형 인공지능의 기술 개발 트렌드를 예측해 보고자 하였다.

• 드라이브 ㆍ 컨트롤
• /
• 제21권3호
• /
• pp.36-45
• /
• 2024

Collaborative robots, designed for direct interaction with humans have limited adaptability to environmental changes. This study addresses this limitation by implementing a barista robot system using AI technology. To overcome limitations of traditional collaborative robots, a model that applies a real-time object detection algorithm to a 6-degree-of-freedom robot arm to recognize and control the position of random cups is proposed. A coffee ordering application is developed, allowing users to place orders through the app, which the robot arm then automatically prepares. The system is connected to ROS via TCP/IP socket communication, performing various tasks through state transitions and gripper control. Experimental results confirmed that the barista robot could autonomously handle processes of ordering, preparing, and serving coffee.

• 로봇학회논문지
• /
• 제17권2호
• /
• pp.198-208
• /
• 2022

In this paper, an autonomous driving system is implemented for the Segye AI Robot Race Competition that multiple vehicles drive simultaneously. By utilizing the ERP42-racing platform, RTK-GPS, and LiDAR sensors provided in the competition, we propose an autonomous driving system that can drive safely and quickly in a road environment with multiple vehicles. This system consists of a recognition, judgement, and control parts. In the recognition stage, vehicle localization and obstacle detection through waypoint-based LiDAR ROI were performed. In the judgement stage, target velocity setting and obstacle avoidance judgement are determined in consideration of the straight/curved section and the distance between the vehicle and the neighboring vehicle. In the control stage, adaptive cruise longitudinal velocity control based on safe distance and lateral velocity control based on pure-pursuit are performed. To overcome the limited experimental environment, simulation and partial actual experiments were conducted together to develop and verify the proposed algorithms. After that, we participated in the Segye AI Robot Race Competition and performed autonomous driving racing with verified algorithms.

• Korean Journal of Radiology
• /
• 제24권11호
• /
• pp.1061-1080
• /
• 2023

Artificial intelligence (AI) in radiology is a rapidly developing field with several prospective clinical studies demonstrating its benefits in clinical practice. In 2022, the Korean Society of Radiology held a forum to discuss the challenges and drawbacks in AI development and implementation. Various barriers hinder the successful application and widespread adoption of AI in radiology, such as limited annotated data, data privacy and security, data heterogeneity, imbalanced data, model interpretability, overfitting, and integration with clinical workflows. In this review, some of the various possible solutions to these challenges are presented and discussed; these include training with longitudinal and multimodal datasets, dense training with multitask learning and multimodal learning, self-supervised contrastive learning, various image modifications and syntheses using generative models, explainable AI, causal learning, federated learning with large data models, and digital twins.

• International Journal of Contents
• /
• 제15권4호
• /
• pp.107-112
• /
• 2019

With the development of AI (artificial intelligence), animation digital publishing has been integrated with intellectualization. This paper adopts the theory of the global value chain, and analyzes the basic structure of the animation publishing value chain. Then focuses on the analysis of digital technology and artificial intelligence technology to play an active role in the topic selection and content customization of animation digital publishing products, optimization of publishing platforms, and user experience of publishing products. Finally, it proposes the use of artificial intelligence data analysis and deep learning technology. The purpose of this paper is to realize the upgrading of animation digital publishing, product upgrading, industrial chain upgrading, and identify some promotion methods for the value chain, such as copyright protection.

• International Journal of Contents
• /
• 제2권3호
• /
• pp.18-21
• /
• 2006

Nowadays, many people enjoy playing games in computer. In this kind of game, people often meet NPC (Non Player Character). It is the virtual character in simplified form of real player and exits in most of current computer games. Various NPCs add the reality and atmosphere of the game as well as help players. There are several techniques to embody NPC, but developers generally use AI technique. This paper discusses some artificial intelligence techniques used in game contents. Especially this paper focuses on the AI techniques used in computer games in terms of the two main approaches, symbolic approach and sub-symbolic approach.