• Journal of Positioning, Navigation, and Timing
• /
• 제9권3호
• /
• pp.175-182
• /
• 2020

This paper introduces a study to estimate the user altitude in need of rescue in an emergency. The altitude is estimated by using the barometric pressure sensor embedded in the smartphone. Compared to GPS, which is degraded in urban or indoor environments, it has the advantage of not having spatial restrictions. With the endless development of smartphone hardware, it is possible to estimate the absolute altitude using the measured value if only the bias of the embedded barometric pressure sensor is applied. The altitude information of the person in need of rescue in an emergency is a great help in reducing rescue time. Since time is tight, we propose online calibration that provides the barometric pressure sensor bias used for altitude estimation through database. Furthermore, experiments were conducted to understand the characteristics of the barometric pressure sensor, which is greatly affected by wind. At the end, the altitude estimation performance was confirmed through an actual field tests in various floors in the building.

• 한국정보전자통신기술학회논문지
• /
• 제4권2호
• /
• pp.103-110
• /
• 2011

u-Health 시스템의 센서들로부터 측정된 데이터에 대한 정확하고 에너지 효율적인 관리가 필요하다. 센서네트워크에서 대용량의 입력 스트림 데이터 전체를 데이터베이스에 모두 저장하여 한꺼번에 처리하는 것은 효율적이지 못하다. 본 논문에서는 u-Health 시스템 내 센서 네트워크의 에너지 효율성과 정확성을 고려하여 여러 센서에서 지속적으로 들어오는 다차원 스트림 데이터의 처리 성능을 높이고자 한다. 효율적인 입력 스트림 처리를 위해서 슬라이딩 윈도우 기반으로 질의를 처리하고 Mjoin 방법으로 다중 질의 계획을 수립한 후 역전파 알고리즘을 통해 저장 데이터를 축소하는 효율적인 처리 기법을 제안한다. 14,324개의 데이터 집합을 사용하여 실험한 결과 실제 입력되는 데이터보다 저장 공간의 18.3%를 축소함으로써 효과적임을 보였다.

• 대한임베디드공학회논문지
• /
• 제4권3호
• /
• pp.126-133
• /
• 2009

There are already many researches providing u-Healthcare service, but they have left problems to be improved. First of all, the transmission range between sensor nodes and the gateway are restricted. Hence, patients feel uncomfortable because of they need to possess or locate closed to a gateway all the time when they aggregates their medical data. Also, the existing systems have not considered life environment that is important to analyze patient's diseases. Moreover, a guardian need to located close to patient or possess a mobile device that monitors a patients' status in real time when they are in outdoor. In this research, we present multi-hop packet transfer algorithm and compilation of life environment which help improve the problem of the existing researches. Likewise, we designed and implemented a medical information database and a real-time web monitoring system that manage patients' personal history and monitor a patients' status in real time. In this paper, we design and implement the u-Healthcare system based on mobile environment and we present a result when we tested our u-Healthcare system in scenario environment.

• 한국산학기술학회논문지
• /
• 제13권2호
• /
• pp.856-865
• /
• 2012

실시간으로 악취 유발 물진의 배출원에 대한 데이터를 구축하여 악취 민원에 효율적으로 대처할 수 있는 악취 정보 관리 시스템이 필요하다. 본 논문에서는 이러한 문제를 해결하기 위하여 유비쿼터스 센서 네트워크(Ubiquitous Sensor Network : USN)기반의 악취 측정 시스템을 통하여 실시간적으로 수집된 악취 데이터를 데이터베이스에 축적 및 주기적으로 데이터를 조회하여 지도상에 표시함으로서 즉각적인 악취 배출상황을 확인할 수 있는 시스템을 구축하였다. 또한, 지도상의 센서노드 아이콘을 클릭함으로서 센서노드 주변의 업체 목록을 지도상 좌표 기준으로 조회, 강한 악취가 발생한 부근의 주변 업체 종류와 배출물질 종류를 파악할 수 있도록 하였다. 제안된 시스템은 이러한 정보로 부터 업체관리 방안을 계획하여 악취배출 업체에 대한 관리 효율을 높일 수 있을 것으로 사료된다.

• Journal of Biosystems Engineering
• /
• 제27권3호
• /
• pp.259-266
• /
• 2002

It is crucial to know spatial soil variability for precision farming. However, it is time-consuming, and difficult to measure spatial soil properties. Therefore, there are needs fur sensing technology to estimate spatial soil variability, and for electronic mapping technology to store, manipulate and process the sampled data. This research was conducted to develop a real-time soil organic matter sensor and an electronic mapping system. A soil organic matter sensor was developed with a spectrophotometer in the 900∼1,700 nm range. It was designed in a penetrator type to measure reflectance of soil at 15cm depth. The signal was calibrated with organic matter content (OMC) of the soil which was sampled in the field. The OMC was measured by the Walkeley-Black method. The soil OMCs were ranged from 0.07 to 7.96%. Statistical partial least square and principle component regression analyses were used as calibration methods. Coefficient of determination, standard error prediction and bias were 0.85 0.72 and -0.13, respectively. The electronic mapping system was consisted of the soil OMC sensor, a DGPS, a database and a makeshift vehicle. An algorithm was developed to acquire data on sampling position and its OMC and to store the data in the database. Fifty samples in fields were taken to make an N-fertilizer dosage map. Mean absolute error of these data was 0.59. The Kring method was used to interpolate data between sampling nodes. The interpolated data was used to make a soil OMC map. Also an N-fertilizer dosage map was drawn using the soil OMC map. The N-fertilizer dosage was determined by the fertilizing equation recommended by National Institute of Agricultural Science and Technology in Korea. Use of the N-fertilizer dosage map would increase precision fertilization up to 91% compared with conventional fertilization. Therefore, the developed electronic mapping system was feasible to not only precision determination of N-fertilizer dosage, but also reduction of environmental pollution.

• 전자공학회논문지CI
• /
• 제45권3호
• /
• pp.103-112
• /
• 2008

본 논문에서는 u-healthcare system과 트래픽 모니터링 및 분석 시스템인 MAGI의 시스템에 대해 설명하고 각각의 데이터베이스를 분석한다. u-healthcare system은 생체 신호 센서 네트워크와 모바일 게이트웨이를 이용하여 환자의 생체 신호를 환자의 장소와 시간에 관계없이 모니터링하고 분석하는 시스템이다. MAGI는 기존의 트래픽 모니터링 툴의 문제점을 보완한 실시간 트래픽 모니터링 및 분석 시스템이다. MAGI 실시간 트래픽을 분석을 통하여 다이나믹한 방화벽을 구현 할 수 있다. 본 논문은 구현한 시스템의 분석은 물론 데이터베이스의 스키마 및 저장된 데이터를 자세히 분석 하였다. 또한 본 논문은 이론적인 전개와 실제 테스트를 통한 정성적인 성능 분석을 통하여 시스템의 적절성을 평가하였다.

• 대한임베디드공학회논문지
• /
• 제10권2호
• /
• pp.101-108
• /
• 2015

In this paper, a robust traffic light detection method is proposed by using vision sensor and DGPS(Difference Global Positioning System). The conventional vision-based detection methods are very sensitive to illumination change, for instance, low visibility at night time or highly reflection by bright light. To solve these limitations in visual sensor, DGPS is incorporated to determine the location and shape of traffic lights which are available from traffic light database. Furthermore the geometric relationship between traffic light and vision sensor is used to locate the traffic light in the image by using DGPS information. The empirical results show that the proposed method improves by 51% in detection rate for night time with marginal improvement in daytime environment.

• 한국소프트웨어감정평가학회 논문지
• /
• 제16권2호
• /
• pp.127-133
• /
• 2020

인공지능의 활용성이 다양해지면서 소형 휴대용기기에 알고리즘을 탑재하려는 요구가 증가하고 있다. 또한 임베디드 시스템이 고성능화하면서 운영체제는 물론 고속연산 및 머신러닝의 알고리즘 구현이 가능해 지고 있다. 그러나 반복연산과 방대한 학습데이터를 처리하는 머신러닝알고리즘의 특성으로 네트워크 연결에 의한 클라우드 환경에 의존하고 있다. 임베디드 시스템에서의 독자적인 운영을 위해서는 저 전력화 및 최적화 알고리즘에 의한 빠른 실행이 요구된다. 본 연구에서는 스마트 제어를 목적으로 임베디드 시스템에 에너지 측정용 센서를 연결하고, 실시간 측정 및 모니터링 시스템으로 측정정보를 데이터베이스로 저장하는 장치를 구현하였다. 연속적으로 측정되어 저장된 데이터는 학습 알고리즘에 적용하여, 최적화 전력제어에 활용가능하며, 에너지 측정에 요구되는 다양한 센서의 인터페이스가 가능한 시스템을 구성하였다.

• 한국철도학회논문집
• /
• 제5권3호
• /
• pp.148-157
• /
• 2002

A health monitoring system becomes a useful tool to obtain information on long term behavior of the important railway structures such as very long span and special type bridges. The health monitoring system not only gives the direct measurement data of the railway bridges but also provides the basic data on the maintenance of the structures. Therefore, periodic calibrations of the health monitoring system will be a necessary step toward precise and accurate assessment of the railway bridges. In this study, the calibration and gauge factor readjustment process made for the health monitoring system installed in the railroad bridges is reviewed and some findings are explained in detail: specifically, the calibrators made for this purpose are illustrated and the regression processes of the calibration on long-term displacement using water level sensor, longitudinal displacement using LVDT sensor, instantaneous displacement using LVDT sensors and accelerometer are described in full length. Based on the regression results, it was found that the gauge factors need to be readjusted according to the regression equation but, since the deviation or shift is not serious so far, long-term observation on each sensor is also recommended. Future work will be concentrated on the long-term analysis of each sensor and on the database creation so that the assessment of the structures is possible.

• 제어로봇시스템학회논문지
• /
• 제14권3호
• /
• pp.248-253
• /
• 2008

This paper presents a sensing system for smart home which can detect an location transition events such as entrance/exit of a member and identify the user in a group at the same time. The proposed system is compose of two sub-systems; a wireless sensor network system and a database server system. The wireless sensing system is designed as a star network where each of sensing modules with ultrasonic sensors and a Bluetooth RF module connect to a central receiver called Bluetooth access point. We propose a method to discriminate a user by measuring the height of the user. The differences in the height of users is a key feature for discrimination. At the same time, the each sensing module can recognize whether the user goes into or out a room by using two ultrasonic sensors. The server subsystem is a sort of data logging system which read the detected event from the access point and then write it into a database system. The database system could provide the location transition information to wide range of context-aware applications for smart home easily and conveniently. We evaluate the developed method with experiments for three subjects in a family with the installation of the developed system into a real house.