• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.50.3-50
• /
• 2002

Our living spaces are becoming increasingly populated with infinite numbers of intelligent embedded agents we are interacting with. in ubiquitous computing environments the aim is to support the occupants during their everyday lives and enhance their living conditions. In this paper we introduce such an environment, the iDorm, which has arisen from our work in careAgents, a project supported by the UK-Korean S&T collaboration fund and the EU's Disappearing Computer Initiative eGadgets project We discuss the research challenges involved, particularly those relating to intelligently associating and configuring large numbers of embedded agents. The paper presents an intelligent association sys...

• 대한임베디드공학회논문지
• /
• 제9권2호
• /
• pp.67-73
• /
• 2014

When a mobile robot performs in unknown environments, a location identification is an essential task. In this paper, we propose a location identification system that uses a sensor space without additional devices on the robot. Also the sensor space consists of a matrix of CDS sensor; when a robot was positioned on the CDS sensor, we can estimate the coordinate of the location by sensing a light. Based on KS illumination standard, experiments are performed in various environments. By evaluating the experimental results, we can show that the proposed system can be applicable to the location identification system of a moving robot.

• 대한임베디드공학회논문지
• /
• 제3권1호
• /
• pp.1-7
• /
• 2008

In this paper, a MSS-NOVO method that combines the HMM composition method with a noise reduction method is proposed for speech recognition in noisy environments. This combined method starts with noise reduction with modified spectral subtraction (MSS) to enhance the input noisy speech, then the noise and voice composition (NOVO) method is applied for making noise adapted models by using the noise in the non-utterance regions of the enhanced noisy speech. In order to evaluate the effectiveness of our proposed method, we compare MSS-NOVO method with other methods, i.e., SS-NOVO, MWF-NOVO. To set up the noisy speech for test, we add White noise to KLE 452 database with different SNRs range from 0dB to 15dB, at 5dB intervals. From the tests, MSS-NOVO method shows average improvement of 66.5% and 13.6% compared with the existing SS-NOVO method and MWF-NOVO method, respectively. Especially our proposed MSS-NOVO method shows a big improvement at low SNRs.

• 대한임베디드공학회논문지
• /
• 제6권1호
• /
• pp.41-47
• /
• 2011

The Pfair scheduling algorithm, which is an optimal scheduling algorithm in the hard real-time multiprocessor environments, propose the necessary and sufficient condition for the schedulability and is based on the fixed quantum size. Recently, several methods that determine the optimal quantum size dynamically were proposed in the mode change environments. But these methods considered only the case in which the period of a task is increased or decreased. In this paper, we also consider the case in which the execution time of a task is increased or decreased, and propose new methods that determine the optimal quantum size dynamically.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2016년도 추계학술대회
• /
• pp.445-447
• /
• 2016

SQLite, Realm 등은 리소스 제약이 큰 소형 기기에 적합한 데이터베이스 엔진들이며 리눅스 기반 모바일 기기에 많이 사용되고 있다. 이 엔진들은 대부분 오픈소스 프로그램들이며 범용 데이터베이스에 비해 가볍고 속도가 빠른 장점을 지닌다. 이 논문에서는 리눅스 기반 환경에서 SQLite 데이터베이스의 검색 성능을 파악하기 위한 테스트 프로그램을 구현하고 성능 실험을 실시한다. 실험은 같은 환경에서 실행되는 Oracle 데이터베이스와 비교하여 진행한다.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2023년도 제68차 하계학술대회논문집 31권2호
• /
• pp.39-40
• /
• 2023

본 논문은 딥러닝 기반의 소리 인식 모델을 기반으로 실내에서 발생하는 다양한 소리를 시각적인 정보로 제공하는 시스템을 위해 경량화된 CNN ResNet 구조의 인공지능 모델을 제시한다. 적용하는 경량화 기법은 모델의 크기와 연산량을 최적화하여 자원이 제한된 장치에서도 효율적으로 동작할 수 있도록 한다. 이를 위해 마이크로 컴퓨터나 휴대용 기기와 같은 임베디드 장치에서도 원활한 인공지능 추론을 가능하게 하는 모델을 양자화 기법을 적용한 경량화 방법들을 실험적으로 비교한다.

• 대한임베디드공학회논문지
• /
• 제18권6호
• /
• pp.293-301
• /
• 2023

With the advancement of artificial intelligence and its various use cases, accessing it through edge computing environments is gaining traction. However, due to the nature of edge computing environments, efficient management and optimization of clusters distributed in different geographical locations is considered a major challenge. To address these issues, this paper proposes a centralization and automation tool called KubEVC-Agent based on Kubernetes. KubEVC-Agent centralizes the deployment, operation, and management of edge clusters and presents a use case of the data transformation for optimizing intra-cluster communication. This paper describes the components of KubEVC-Agent, its working principle, and experimental results to verify its effectiveness.

• 융합보안논문지
• /
• 제7권4호
• /
• pp.15-21
• /
• 2007

임베디드 기기에서 플래시 메모리의 사용량이 증가하고, 임베디드 기기가 여러 computing 환경에서 점점 중요한 위치를 점함에 따라 임베디드 파일 시스템의 보안이 중요한 문제가 된다. 또한 임베디드 기기의 경우 휴대성이 좋은 반면 분실의 위험도 크고, 분실한 경우 기기 내부의 플래시 메모리에 중요 정보가 있다면 사용자에게 큰 손실을 야기할 수 있다. 이처럼 다양한 상황에서 임베디드 기기 내부의 플래시 파일 시스템의 보안성을 향상시키기 위해 암호화 플래시 파일 시스템을 설계 및 구현 하였다. 이를 위해 현재 많이 사용되는 YAFFS 파일 시스템을 수정하였다. 수정된 YAFFS 암호화 파일 시스템을 통해 임베디드 기기의 보안성을 한 층 강화할 수 있다.

• 대한임베디드공학회논문지
• /
• 제4권3호
• /
• pp.139-147
• /
• 2009

This paper proposed the context awareness reasoning system to provide the personalized services dynamically in a ubiquitous mobile environments. The proposed system is designed to provide the personalized services to mobile users and consists of the context aggregator and the knowledge manager. The context aggregator can collect information from networks through Open API Gateway as well as sensors in a various ubiquitous environment. And it can also extract the place types through the geocoding and the social address domain ontology. The knowledge manager is the core component to provide the personalized services, and consists of activity reasoner, user pattern learner and service recommender to provide the services predict by extracting the optimized service from user situations. Activity reasoner uses the ontology reasoning and user pattern learner learns with previous service usage history and contexts. And to design service recommender easy to flexibly apply in dynamic environments, service recommender recommends service in the only use of current accessible contexts. Finally, we evaluate the learner and recommender of proposed system by simulation.

• 대한임베디드공학회논문지
• /
• 제17권2호
• /
• pp.77-83
• /
• 2022

In this paper, the two-stage object detection approach is proposed to implement a deep learning-based label inspection system on edge computing environments. Since the label printed on the products during the production process contains important information related to the product, it is significantly to check the label information is correct. The proposed system uses the lightweight deep learning model that able to employ in the low-performance edge computing devices, and the two-stage object detection approach is applied to compensate for the low accuracy relatively. The proposed Two-Stage object detection approach consists of two object detection networks, Label Area Detection Network and Character Detection Network. Label Area Detection Network finds the label area in the product image, and Character Detection Network detects the words in the label area. Using this approach, we can detect characters precise even with a lightweight deep learning models. The SF-YOLO model applied in the proposed system is the YOLO-based lightweight object detection network designed for edge computing devices. This model showed up to 2 times faster processing time and a considerable improvement in accuracy, compared to other YOLO-based lightweight models such as YOLOv3-tiny and YOLOv4-tiny. Also since the amount of computation is low, it can be easily applied in edge computing environments.