• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 추계학술대회
• /
• pp.597-599
• /
• 2022

IoT 디바이스에서 on-device AI를 수행할 때, 타겟 서비스나 디바이스의 환경에 따라 필요한 AI 모델이 달라질 수 있다. 또한, 기존 AI 모델도 federated learning과 같이 추가적인 데이터를 이용해 트레이닝을 하거나 보다 향상된 새로운 기법을 사용하는 등 업데이트가 일어날 수 있다. 이에 따라 IoT 디바이스에서 양질의 AI 서비스를 수행하기 위해서는 상황에 따라 필요한 AI 모델을 선택적으로 사용하거나 최적화된 최신 버전의 AI 모델로 업데이트 할 수 있어야 한다. 본 논문에서는 이를 지원하기 위한 AI 모델 레포지토리를 제안한다. 레포지토리는 AI 모델의 등록, 검색, 관리 및 배포를 지원하며 실사용을 위한 웹 포털을 포함한다. 제안하는 시스템의 실효성 확인을 위해 Node.js와 Vue.js로 구현하여 동작을 확인하였다.

• Journal of Information Processing Systems
• /
• 제20권2호
• /
• pp.215-225
• /
• 2024

With the rapid development of Internet of Things (IoT) and big data technology, a large amount of data will be generated during the operation of related industries. How to classify the generated data accurately has become the core of research on data mining and processing in IoT industry chain. This study constructs a classification model of IoT industry chain based on improved random forest algorithm and text analysis, aiming to achieve efficient and accurate classification of IoT industry chain big data by improving traditional algorithms. The accuracy, precision, recall, and AUC value size of the traditional Random Forest algorithm and the algorithm used in the paper are compared on different datasets. The experimental results show that the algorithm model used in this paper has better performance on different datasets, and the accuracy and recall performance on four datasets are better than the traditional algorithm, and the accuracy performance on two datasets, P-I Diabetes and Loan Default, is better than the random forest model, and its final data classification results are better. Through the construction of this model, we can accurately classify the massive data generated in the IoT industry chain, thus providing more research value for the data mining and processing technology of the IoT industry chain.

• 예술인문사회 융합 멀티미디어 논문지
• /
• 제7권4호
• /
• pp.623-630
• /
• 2017

본 논문은 라즈베리파이3를 활용한 IoT 교육과정의 커리큘럼 설계요소를 도출하기 위한 한국의 IoT 사용서비스의 현황을 고찰한다. 첫째, 한국의 Korea Telecom(KT), LGU+ Telecome (LGT), SK Telecom(SKT)와 같은 3대 메이저 통신업체의 IoT 연구 및 상용서비스를 고찰한다. 둘째, 현재로서 IoT 교육도구로서 가장 적합하다고 각광받는 라즈베리파이3의 구성요소 및 장점을 고찰한다. 마지막으로 이 둘의 공통분모를 토대로 IoT의 사회적 수요인력을 양성하기 위한 라즈베리파이3를 활용한 IoT 교육과정의 중요 설계요소를 도출하게 된다. 해당 설계요소로는 플랫폼, 하드웨어, 소프트웨어, 빅데이터 네트워크가 있다. 이와 같이 도출된 라즈베리파이3를 활용한 IoT 교육과정의 중요 설계요소를 바탕으로 본 필자의 대학의 임베디드 시스템 과목을 1학기동안 교육하였고, 성공적으로 수강생 25명중 90% 이상의 우수 강의 만족도와 강의 분량을 검증하였다.

• 전기학회논문지
• /
• 제67권2호
• /
• pp.301-307
• /
• 2018

Unmanned Aerial Vehicles (i.e., drone) are gaining a lot of interest from a wide range of application domains such as infrastructure monitoring and parcel delivery service. In those service scenarios, multiple UAVs are involved and should be reliably operated by so-called UAV management system. For that, we propose oneM2M standard based UAV management and control system which is specifically targeted at traffic management of low-altitude UAVs. In this paper, we include oneM2M platform architecture and its implementation for UAV management system in conjunction with UAV interworking procedure.

• International journal of advanced smart convergence
• /
• 제7권4호
• /
• pp.75-83
• /
• 2018

Predictive IoT Sensor Algorithm is a technique of data science that helps computers learn from existing data to predict future behaviors, outcomes, and trends. This algorithm is a cloud predictive analytics service that makes it possible to quickly create and deploy predictive models as analytics solutions. Sensors and computers collect and analyze data. Using the time series prediction algorithm helps to predict future temperature. The application of this IoT in industrial environments like power plants and factories will allow organizations to process much larger data sets much faster and precisely. This rich source of sensor data can be networked, gathered and analyzed by super smart software which will help to detect problems, work more productively. Using predictive IoT technology - sensors and real-time monitoring - can help organizations exactly where and when equipment needs to be adjusted, replaced or how to act in a given situation.

• 유통과학연구
• /
• 제18권5호
• /
• pp.15-23
• /
• 2020

Purpose: This research focused on the fact that the Internet platform is integral to IoT products such as Smart home and studied consumer buying decisions when products are sold bundled with internet service. Contrary to the sales strategies of telecommunication companies, some companies sell IoT products alone, for example Google, Kakao, and Naver. In this market situation, the sales strategies of Korean telecommunication companies were analyzed with bundling theory and technology acceptance model, then it was conducted to figure out which sales and distribution strategies could affect consumers' purchase behavior. Research design, data, methodology: Data was collected by149 questionnaires from groups who are familiar with IoT smart home systems, then exploratory factor analysis and regression were used to analyze the research model. Results: The results revealed that the perceived ease of use and the perceived usefulness affect the purchase intention of IoT-based products; however, this effect was not found in the case of bundled products. In other words, it is found that selling and distributing Internet services and IoT products together does not affect consumers' purchases. Conclusion: It is suggested that Korean telecommunications companies' existing sales and distribution strategies for IoT products need to be changed according to its characteristics.

• 한국멀티미디어학회논문지
• /
• 제17권12호
• /
• pp.1461-1472
• /
• 2014

Recently, smart devices has been spreaded and the definition of RFID/USN(Radio Frequency Identification/Ubiquitous Sensor Network) technology has also extended. So, originally, human is an agent of communications but now, 'the things' are also an agent of communication. That is, Internet of Things(IoT). IoT is very big IT infra technology that is based on wire-wireless network technologies such as Bluetooth and Wi-Fi. However, in reality, IoT service that can provide various meaning and integrate smoothly is not appear yet. To solve these problems of IoT, we implemented Publish/Subscribe system, which operate in Zigbee based wireless sensor network by using MQTT-SN protocol. Moreover, this system can operate with combining MQTT system that is already existed in external Internet network so, extensibility and flexibility aspect is excellent. We show that possibility of IoT services that is combined between different heterogeneous network through the MQTT-SN.

• 한국IT서비스학회지
• /
• 제16권4호
• /
• pp.213-222
• /
• 2017

The advancement of ICT is changing every field of life. It becomes possible with the penetration of personalized devices, that is, smartphone. The boom of IoT will come when there exist diversified and personalized services. In general, we might admit that it is needed that the more privatized services than the overall serviced. Jeju Island is the only one special self-governing province in Republic of Korea and deserves to be proud of the unique culture from its long historical background. One of the very regional culture which performs by women divers, Haenyeo activity or culture, was registered as Intangible Cultural Heritage. When authors were researched Jeju Haenyeo as a worthy reserving service, we recognized that it has never considered to use any point of ICT/IoT yet. Because IoT holds the high potentiality to create any service scenario between interesting groups. We will design a few services for Haenyeo which covers their job territory or daily life, adopts up-to-date technology or method such as sensored network, smart contract and App/Web. In this paper, we intent to show the simplicity and easiness of the application of IoT not to much inconspicuous target. So, we suggest a specialized IoT service for the reservation and promotion of Haenyeo Culture. This service would be composed of sensors, IoT network and App/Web at home and office. This service can be used among interesting groups : Haenyeo, policy maker, manufacturer, service provider and culture consumer.

• 한국항해항만학회지
• /
• 제45권4호
• /
• pp.181-185
• /
• 2021

본 연구에서는 어구의 정상 또는 유실 등의 상태 여부를 판단하기 위한 어구정보를 IoT 기반의 통신망을 이용하여 수신하고 분석하여, 어구의 현재 상태를 관리할 수 있고, 어구에 이상 상태 발생 시 이를 확인하고, 신속한 어구 회수 등의 관리를 수행할 수 있도록 지원하는 어구 부이 관리시스템을 개발하는 것에 관한 내용을 다룬다. IoT 기반의 통신망을 사용한 어구 관리 체계 및 통합 관리 구조 설계를 수행하였으며 이를 이용한 어구 부이 통합 관리 시스템을 개발하여 이에 대한 시스템 시험 및 검증을 수행하여 실효성을 확인하였다.

• 문화기술의 융합
• /
• 제8권2호
• /
• pp.355-360
• /
• 2022

본 연구에서는 마이크로 데이터 센터의 환경 모니터링 시스템(EMS)을 설계하고 개발하는 방법을 나타낸다. 설계하는 환경 모니터링 시스템은 IoT 센서 및 싱글보드 컴퓨터를 이용하여 데이터센터 운영환경을 실시간으로 감지하여 데이터를 해석하고 운영환경을 모니터한다. 먼저, EMS 하드웨어 플랫폼 및 IoT 센싱 패키지 설계 방법에 대하여 나타낸다. 또한 서로 다른 프로토콜을 갖는 IoT 센서에 대한 데이터 수집 SW을 설계하고 원격에서 모니터링이 가능한 EMS 모니터링 SW를 개발한다. 데이터 수집 SW는 센서 수집 모듈과 수집 메니저 모듈로 구성된다. EMS 소프트웨어는 마이크로 서비스 아키텍츠(MSA) 개념을 적용하여 설계하며 각각의 서비스는 독립적인 비즈니스 로직을 포함한다.