• Journal of the Korea Society of Computer and Information
• /
• v.15 no.10
• /
• pp.105-112
• /
• 2010

In this paper, the bio-signals, such as EEG, ECG were measured with a sensor and their characters were drawn out and analyzed. With results from the analysis, four emotion of rest, concentration, tension and depression were inferred. In order to assess one's emotion, the characteristic vectors were drawn out by applying various ways, including the frequency analysis of the bio-signals like the measured EEG and HRV. RBFN, a neural network of the complex structure of unsupervised and supervised learning, was applied to classify and infer the deducted information. Through experiments, the system suggested in this thesis showed better capability to classify and infer than other systems using a different neural network. As follow-up research tasks, the recognizance rate of the measured bio-signals should be improved. Also, the technology which can be applied to the wired or wireless sensor measuring the bio-signals more easily and to wearable computing should be developed.

• Journal of the Institute of Convergence Signal Processing
• /
• v.20 no.1
• /
• pp.45-50
• /
• 2019

When PPG signal is used in mobile healthcare devices, the accuracy of the measured heartbeat decreases from the influence by the movement of the user. The reason is that the frequency band of the noise overlaps the frequency band of the PPG signal. In order to remove these same noises, the methods using frequency analysis method or application of acceleration sensor have been investigated and showed excellent performance. However, in applying these methods to low-cost healthcare devices, it is difficult to apply these methods because of much processing time and sensor's cost. In order to solve these problems, this study proposed the filter design method using an adaptive notch filter and the fuzzy inference system to extract more accurate heart rate in real time and evaluated its performance. As results, it showed better results than the other methods. Based on the results, when applying the proposed method to design the mobile healthcare device, it is possible to measure the heartbeat more accurately in real time.

• 한국공간정보시스템학회:학술대회논문집
• /
• 2007.06a
• /
• pp.194-199
• /
• 2007

최근 차량증가에 따라 교통사고가 날로 증가하고 있고, 안전운전을 위한 보조장치 개발의 필요성이 증대되고 있다. 특히 교통사고 원인중 운전자의 졸음으로 인한 사고가 30%에 달하고 있어 졸음운전 예방장치의 개발이 시급한 실정이다. 이에 본 연구에서는 졸음운전예방을 위한 효율적인 졸음경보 장치시스템을 제안한다. 제안하는 방법은 HRV와 안구활동 신호를 BP(역전파알고리즘)을 이용 보다 더 정확한 측정 및 판단을 행하여 운전자에게 음성경고를 보내주는 방식이다. 제안된 방식은 기존의 연구된 피부전기활동, 영상처리를 이용한 졸음운전 감지 시스템, 심박변동신호 분석 방법보다 훨씬 효율적인 졸음경보를 할 수 있을 것으로 기대된다.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.08a
• /
• pp.289-292
• /
• 2000

선형 자귀회귀(AR) 모델을 근거로한 HRV 파워 스펙트럼해석은 비침습적으로 자율신경의 반응을 정량화하는데 폭넓게 사용된다. 본 연구는 단구간(2분 미만)의 심전도와 맥파 신호로부터 시계열 HRV의 파워스펙트럼을 추정한다. 시계열은 정상인을 대상으로 검출한 심전도와 맥파신호의 특징점 시간간격(RRI, PPI)으로부터 구하였다. 발생된 시계열은 다항식 보간법에 의해 AR모델에 적합하게 재구성하였으며, AR모델 계수는 Burg법에 의해 계산하였다. AR모델을 적용한 단구간의 심전도와 맥파의 심박변동에 대한 파워스펙트럼밀도는 저주파수(LF)와 고주파수(HF)에서 매끄러운 스펙트럼 파워를 나타내고 있다. 또한 동일한 피험자의 심전도와 맥파의 파워스펙트럼밀도를 비교한 결과 동일한 모양을 나타내었다.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2015.07a
• /
• pp.265-266
• /
• 2015

인간의 사고 및 행동능력은 환경부하, 작업의 내용 등 외부환경에 큰 영향을 받는다. 선박운항 시, 인간-기계시스템 환경에서 항해사의 일정한 수준 이상의 판단 및 조선능력은 안전운항을 위해 매우 중요한 요소이다. 특히 협수로 항해시, 선박통항량이 많은 해역을 항해할 경우 선교에서의 작업부하가 커지고 그로 인한 피로 및 긴장에 의한 심신 기능이 저하되어, 항해사의 조선(操船)능력에 영향을 미치게 될 것이다. 그러나 선박운항 또는 입출항 시 항해사의 생리적, 심리적 스트레스를 측정하여 계량화한 연구는 미비하다. 따라서 본 연구에서는 웨어러블 심박수 측정장치를 이용하여 항해사의 심박수 변화를 측정 분석하고, 항해사의 선박 운항작업 시 발생하는 스트레스를 계량화하여 모델을 개발하기 위한 기초연구를 수행하였다.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.12
• /
• pp.17-23
• /
• 2020

As the number of households with companion animals increases, the demand to monitor the health of companion animals in remote locations far away is increasing. We are going to put a wearable device on a companion animal so that it can monitor the heart beat signal from a remote location. However, the monitoring method using Bluetooth has some disadvantages. it can be accessed only in a short distances. In case of WiFi, large power consumption is the problem. To overcome these issues, we propose a system to reduce power consumption by indirectly receiving a user's request using Bluetooth at a time when the user does not need it, and sending sensor data through WiFi when the user makes a monitoring request.

• Science of Emotion and Sensibility
• /
• v.23 no.2
• /
• pp.51-60
• /
• 2020

This study aimed to develop a modular smart clothing system for heart rate monitoring that reduces the inconvenience caused by battery charging and the large size of measurement devices. The heart rate monitoring system was modularized into a temporary device and a continuous device to enable heart rate monitoring depending on the requirement. The temporary device with near-field communication (NFC) and heart rate sensors was developed as a clothing attachment type that enables heart rate monitoring via smart phone tagging when required. The continuous device is based on Bluetooth Low Energy (BLE) communication and batteries and was developed to enable continuous heart rate measurement via a direct connection to the temporary device. Furthermore, the temporary device was configured to connect with a textile electrode made of a silver-based knitted fabric designed to be located below the pectoralis major muscle for heart rate measurement. Considering the user-experience factors, key functions, and the ease of use, we developed an application to automatically log through smart phone tagging to improve usability. To evaluate the accuracy of the heart rate measurement, we recorded the heart rate of 10 healthy male subjects with a modular smart clothing system and compared the results with the heart rate values measured by the Polar RS800. Consequently, the average heart rate value measured by the temporary system was 85.37, while that measured by the reference device was 87.03, corresponding to an accuracy of 96.73%. No significant difference was found in comparison with the reference device (T value = -1.892, p = .091). Similarly, the average heart rate measured by the continuous system was 86.00, while that measured by the reference device was 86.97, corresponding to an accuracy of 97.16%. No significant difference was found in terms of the heart rate value between the two signals (T value = 1.089, p = .304). The significance of this study is to develop and validate a modular clothing system that can measure heart rates according to the purpose of the user. The developed modular smart clothing system for heart rate monitoring enables dual product planning by reducing the price increase due to unnecessary functions.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.9
• /
• pp.328-334
• /
• 2012

Continuous fetal heart beat monitoring has assisted clinicians in assuring fetal well-being during antepartum and intrapartum. Fetal heart rate (FHR) is an important parameter of fetal health during pregnancy. The Doppler ultrasound is one of very useful methods that can non-invasively measure FHR. Although it has been commonly used in clinic, inaccurate heart rate reading has not been completely resolved.. The objective of this study is to improve detection algorithm of FHR from Doppler ultrasound signal with simple method. We modified autocorrelation function to enhance signal periodicity and adopted adaptive window size and shifted for data segment to be analysed. The proposed method was applied to real measured data, and it was verified that beat-to-beat FHR estimation result was comparable with the reference fetal ECG data. This simple and effective method is expected to be implemented in the embedded system.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.47 no.5
• /
• pp.33-42
• /
• 2010

In this study, we have proposed a simple cardiorespiratory monitoring method based on displacements of human body which occurs due to periodic heartbeat and breathing. The proposed system consists of an aircushion, pressure sensing hardware and heartbeat and respiration signals extraction algorithm. The aircushion was used for unconstrained measurement of the respiration and heartbeats without a sensor attached on the subject's skin surface. The displacements of subject sitting on the aircushion cause small pressure variations. These variations are amplified and filtered with the pressure sensing hardware. Finally, heart rate and respiration rate are extracted by signal processing algorithm based on frequency domain filter. To evaluate the performance, extracted respiration and heart rate from proposed system were compared with conventional methods. The average sensitivity of respiration and heart rate are 98.67% and 99.24%, respectively. These results show the proposed method has advantages of installing and processing simplicity so as to be used easily in unconstrained respiration and heart rate monitoring in daily life.

• Proceedings of the KIEE Conference
• /
• 2008.10b
• /
• pp.474-475
• /
• 2008

심실 빈맥은 심실에서 비정상적인 전기자극 발생으로 인하여 심박수가 분당 $100{\sim}120$회를 초과하는 부정맥 증상을 일컫는다. 심실 빈맥이 발생하는 경우 심박출량이 감소하고, 폐에는 미처 나가지 못한 피가 고이는 경우도 발생하여 심부전증이 나타나거나 심실 세동으로 발전하여 사망에 이를 수 있는, 매우 위험한 부정맥 중의 하나이므로 심실 빈맥 검출은 매우 중요한 사안이다. 따라서 본 연구에서는 R-R 간격 정보를 이용하여 심실 빈맥 부정맥 신호를 실시간으로 검출할 수 있는 신호처리 알고리즘을 구현하고자 하였다.