• The Journal of the Acoustical Society of Korea
• /
• v.13 no.2E
• /
• pp.58-67
• /
• 1994

Hidden Markov Model (HMM)-based algorithms have been used successfully in many speech recognition systems, especially large vocabulary systems. Although general purpose processors can be employed for the system, they inevitably suffer from the computational complexity and enormous data. Therefore, it is essential for real-time speech recognition to develop specialized hardware to accelerate the recognition steps. This paper concerns with a real-time implementation of an isolated word recognition system based on HMM. The speech recognition system consists of a host computer (PC), a DSP board, and a prototype Viterbi scoring board. The DSP board extracts feature vectors of speech signal. The Viterbi scoring board has been implemented using three field-programmable gate array chips. It employs a hardware-efficient Viterbi scoring architecture and performs the Viterbi algorithm for HMM-based speech recognition. At the clock rate of 10 MHz, the system can update about 100,000 states within a single frame of 10ms.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.1
• /
• pp.189-194
• /
• 2021

With the recent increase in the proportion of single-person households, starting in 2016, the domestic shared personnel mobility market such as electric kickboards and electric wheels has grown rapidly. Personal transportation means such as electric kickboards are power devices using electricity and are eco-friendly, lightweight, and do not occupy a separate parking space. Above all, it has the advantage of being convenient to travel short and medium distances, so it has been able to obtain a lot of demand from younger users who pursue reasonable consumption, and accordingly, the related market has grown rapidly. However, as absence of the charging station for electric kickboards, electric kickboards are left everywhere on the road, and are emerging as a threat to safety as well as aesthetics. Therefore, this paper aims to research and propose a new IoT management system for efficient recovery of shared electric kickboards. Through this system, it is expected that the high recovery rate of the electric kickboard can be maintained, and in conclusion, the safety of the user and the surrounding environment can be improved.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.1
• /
• pp.175-179
• /
• 2021

With the recent increase in the proportion of single-person households, the demand for reasonable personal mobility has increased, and the "Personal Mobility" industry that can be used conveniently and concisely has grown rapidly. In fact, according to data from the Korea Transport Institute, the scale of the electric kickboards rental industry, one of the personal mobility industry sectors, is expected to expand to 200,000 units in 2022. Due to the characteristics of electric kickboards that are powered by electricity, stable and efficient battery supply is the most basic and important issue. According to recent reviews from users who have used the electric kickboard, there were cases where the use of the electric kickboard is attempted, but the battery is in a discharged state or the battery charge level is low and thus cannot be used. Therefore, this paper proposes a solar charging system for stable battery use of electric kickboards. When this system is applied, it is expected that it will not only be an eco-friendly charging method for electric kickboards, but also stably supply and demand batteries while driving.

• The Science & Technology
• /
• v.28 no.12 s.319
• /
• pp.88-89
• /
• 1995

중소기업으로 발붙이기 어려운 정보통신ㆍ전자산업에 뛰어들어 핵심제품을 차례로 국산화하겠다는 목표아래 몇몇 젊은이들이 똘똘 뭉쳐 세운 (주)한아시스템. 금성사 정보기기연구소에서 기술을 익힌 신동주사장은 모토로라의 MC68040을 탑재한 싱글보드컴퓨터 KVME040을 탑재한 싱글보드컴퓨터 KVME040을 국내 최초로 개발, 새바람을 일으켰다.

• Journal of Advanced Marine Engineering and Technology
• /
• v.34 no.5
• /
• pp.711-717
• /
• 2010

In this paper, a new automatic punching system that generates pinholes expressing texts or images on a plastic plate is developed. The pin-holed plate is used as a new glamorous display board reflecting colourful lights from the light emitting diode (LED) installed on the edge side of the plate. The punching system has four actuators which work together to make multiple holes with accurate position and depth on the plastic plate. For even reflection of the lights from texts or images on the board and fast production of the pin-holed boards, we developed an accurate actuating structure of the system cooperating with a PID control algorithm. We also built a GUI-based integrated control system to help users easily design luminous texts or images on the plastic plate. Also, we conducted a performance test of the system to verify the punching speed and depth control of the pin holes on the plastic plate.

• Proceedings of the KAIS Fall Conference
• /
• 2011.12b
• /
• pp.558-561
• /
• 2011

본 논문은 임베디드 타겟 보드의 시스템을 활용하여 웹서버를 운용하고, 웹페이지를 통해 임베디드타겟 보드 내의 신호를 제어할 수 있도록 제작한 신호 감지 시스템을 통해 제어 신호를 받아 실제 작동여부를 확인할 수 있는 조형물을 구성하였다. 또한, Linux 커널을 컴파일하여 포팅한 임베디스 시스템과 Linux OS기반의 인증 서버를 구축하여 안전한 홈 시스템의 보안접근 제어가 가능하도록 하였다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.2
• /
• pp.186-192
• /
• 2014

This study presents the remote control of a board robot using an IMU sensor based on Bluetooth communication. The board robot is a kind of riding robot controlled throng wireless communication by a user. The user wears the proposed IMU sensor controller, and changes a direction of the robot by the angles of IMU sensor. Bluetooth is used for wireless communication between the board robot and its user. The IMU sensor in the remote controller is used for recognition of a number of actions, which are measured as analog signals. The user actions have five commands ('1'right '2'neutrality '3'left '4'operation '5'stop), which are transmitted from the user to the board robot through Bluetooth communication. Experimental results show that proposed IMU interface can effectively control the board robot.

• Proceedings of the KIEE Conference
• /
• 2001.07b
• /
• pp.1185-1187
• /
• 2001

본 논문에서는 Motorola사의 MPC555마이크로프로세서를 탑재하여 개발한 제어보드가 소개된다. 이 보드를 사용하여 하이브리드 전기자동차의 발전기를 제어하는 실험을 수행하였다. MFC555는 전력시스템 제어에 필요한 다양한 입출력 장치를 내장하고 있어서 통합제어를 위한 one-chip solution을 가능하게 해 준다. MPC555는 내부 플래시 메모리가 비교적 대용량(448kbytes)이고 부동 소수점 연산이 가능하다. 또한 A/D 채널이 32개이고 SPI(Serial Peripheral Interface) 모듈 1개, SCI(Serial Communication Interface) 모듈 2개, CAN(Contol Area Network) 모듈 2개 등의 다양한 통신채널을 내장하고 있다. MPC555는 TPU(Time Processing Unit) 채널 32개로 다양한 timing function을 구현할 수 있게 해 준다. 개발된 제어 보드를 이용하여 하이브리드 전기자동차의 유도발전기 시스템에 대한 축소 시뮬레이션을 수행하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.598-600
• /
• 2005

본 논문에서는 터치스크린을 사용한 디지털보드게임에서 사용자에게 게임의 전락 정보 제공을 통해 게임의 습득 속도를 증가시켜 줄 수 있는 PGA(Portable game assistant)를 제안한다. 기존의 디지털보드게임은 모든 게이머를 위한 공동영역과 개인만을 위한 개인영역으로 나눈다. 공동영역에서는 모든 사용자가 알아야 하는 게임환경과 도움말을 제공하고, 개인영역에서는 게임 진행에 필요한 게임화면과 말의 정보를 제공한다. 우리는 공동영역인 게임테이블에서 터치스크린을 사용하여 게임을 진행하고, 개인영역에 게임의 재미를 더 하기 위하여 PGA를 사용하여 게임의 전략을 그래픽으로 제공한다. 우리는 배우기 쉽고 전략적으로 지원이 필요한 장기게임에 제안된 시스템을 적용하였으며, 앞으로 본 시스템에 게임 애니메이션, 사운드 등을 추가하여 다양한 테이블보드게임에 전략적인 측면과 동적인 측면을 함께 제공할 것이다.

• Annual Conference of KIPS
• /
• 2019.10a
• /
• pp.827-829
• /
• 2019

본 논문에서는 최근 소개된 구글 TPU 보드를 사용하여 AI 알고리듬을 적용하고 성능 분석을 통하여 TPU 를 통한 AI 에 기반한 영상처리 시스템의 구현 가능성을 검증 하고자 하였다. 구글 TPU 보드는 기계 학습에 특화된 Coral Dev 보드를 사용하였고. 수행하는 인공지능 알고리즘은 객체 인식 알고리즘인 SSD 알고리즘을 사용하였다. 이 후 동일한 알고리즘을 GPU 가 장착되어 있는 고성능 데스크탑과 처리속도를 비교하여, TPU 에 기반한 임베디드 AI 시스템의 활용 가능성을 검증 하였다.