• Structural Monitoring and Maintenance
• /
• v.5 no.1
• /
• pp.51-65
• /
• 2018

A number of sensing techniques have been implemented for detecting defects in civil infrastructures instead of onsite human inspections in structural health monitoring. However, the issue of faults in sensors has not received much attention. This issue may lead to incorrect interpretation of data and false alarms. To overcome these challenges, this article presents a deep learning-based method with a new architecture of Stateful Long Short Term Memory Neural Networks (S-LSTM NN) for detecting sensor fault without going into details of the fault features. As LSTMs are capable of learning data features automatically, and the proposed method works without an accurate mathematical model. The detection of four types of sensor faults are studied in this paper. Non-stationary acceleration responses of a three-span continuous bridge when under operational conditions are studied. A deep network model is applied to the measured bridge data with estimation to detect the sensor fault. Another set of sensor output data is used to supervise the network parameters and backpropagation algorithm to fine tune the parameters to establish a deep self-coding network model. The response residuals between the true value and the predicted value of the deep S-LSTM network was statistically analyzed to determine the fault threshold of sensor. Experimental study with a cable-stayed bridge further indicated that the proposed method is robust in the detection of the sensor fault.

• Journal of IKEEE
• /
• v.13 no.3
• /
• pp.39-46
• /
• 2009

Wireless sensor networks consist of small, autonomous devices with wireless networking capabilities. In order to further increase the applicability in real world applications, minimizing energy consumption is one of the most critical issues. Therefore, accurate energy model is required for the evaluation of wireless sensor networks. In this paper we analyze the power consumption for wireless sensor networks. To develop the power consumption model, we have measured the power characteristics of commercial Kmote node based on TelosB platforms running TinyOS. Based on our model, the estimated lifetime of a battery powered sensor node can use about 6.9 months for application of human detection using PIR sensors. This result indicates that sensor nodes can be used in a monitoring system for elderly living alone.

• Journal of Korean Society of Water and Wastewater
• /
• v.20 no.4
• /
• pp.585-594
• /
• 2006

This study focused on the causes of fish kills and its prevention methods in Yudeung Stream, Daejeon, Korea. Intense field data, continuous water quality monitoring system and water quality modeling were applied to analyze the causes. Pollutant can be delivered to urban streams by surface runoff and combined sewer overflows in rainfall events. However, water quality analysis and water quality modeling results indicate that the abrupt fish kills in the Yudeung stream seems to be caused by combined effect of DO depletion, increase in turbidity and other toxic material. Excessive fish population in the study area may harm the aesthetic value of the stream and also has greater potential for massive fish kills. It is suggested to implement methods to reduce delivery of pollutants to the stream not only to prevent fish kills but also to keep balance of ecosystem including human uses. Frequent clean up of the urban surface and CSO, installation of detention basin will be helpful. In the long run, it seems combined sewer system has be replaced with separate sewer system for more effective pollutant removal in the urban area.

• Korean Journal of Clinical Laboratory Science
• /
• v.46 no.1
• /
• pp.38-45
• /
• 2014

Intraoperative Neurophysiological Monitoring (INM) inspection has a very important role. While preserving the patient's neurological function be sure to safe surgery, neurological examination should thank. Cerebello pontine angle tumor surgery, especially in the nervous system is more important to the meaning of INM. In cochlear nerve, facial nerve, trigeminal nerve, which are intricate brain surgery, doctors are only human eye and brain to the brain that it is virtually impossible to distinguish the nervous system. They receives a lot of help from INM. In this paper, we examined six kinds broadly. First, the methods of spontaneous EMG and Free-running EMG, which can instantly detect a damage inflicted on a nerve during surgery. Second, methods of triggered EMG and direct nerve electrical stimulation, which directly stimulate a nerve using electricity to distinguish between nerves and brain tumors. Third, the method of knowing a more accurate neurologic status by informing neurological surgeons about Free-running EMG wave forms that are segmetalized into four. Fourth, three ways of knowing when a patient will be awaken from intraoperative anesthesia, which happens due to a weak anesthetic. Fifth, a method of understanding the structures of a brain tumor and a facial nerve as five dividend segments. Sixth, comparisons between cases normal facial nerve recovery and occurrence of a facial nerve paralysis during the postoperative course.

• International Journal of Advanced Culture Technology
• /
• v.4 no.3
• /
• pp.13-19
• /
• 2016

The growth of mobile devices in Internet of Things (IoT) leads to a number of remote and controlling system related IoT applications. For instance, home automation controlling system uses client system such web apps on smartphone or web service to access the home server by sending control commands. The home server receives the command, then controls for instance the light system. The web service gateway responsible for handling clients' requests attests an internet latency when an increasing number of end users requests submit toward it. Therefore, this web service gateway fails to detect several commands, slows down predefined actions which should be performed without human intervention. In this paper, we investigate the performance of a web server-side platgorm based event-driven, non-blocking approach called Node.js against traditional thread-based server side approach to handle a large number of client requests simultaneously for remote and controlling system in IoT remote monitoring applications. The Node.JS is 40% faster than the traditional web server side features thread-based approach. The use of Node.js server-side handles a large number of clients' requests, then therefore, reduces delay in performing predefined actions automatically in IoT environment.

• Proceedings of the KSR Conference
• /
• 2008.06a
• /
• pp.224-231
• /
• 2008

A subway or an underground railway is one of the representative public transportations which lots of people take everyday. Then, subway station, which is also one of the very important public civil infrastructures, generally services for a long period of time. During the service time of stations, they are easily damaged from environmental corrosion, material aging, fatigue, and the coupling effects with long-term loads and extreme loads. Recently, civil construction work on the places near station often creates lots of damages to the station. As these damages accumulate, the performance of station degenerates due to the above factors. They would inevitably reduce the resisting capacity of station against the disaster; even they bring into the collapse of stations with the structural failure under long-term loads and extreme loads. And, if disaster such as earthquake, fire, etc. happens, it causes huge property damage and threatens the human lives. Because of these above reasons, the structural health monitoring system need to be developed for ensuring the safety of station. In this paper, the development directions of the structural health monitoring system with fiber optic sensor and USN for subway station are briefly described.

• International Journal of Advanced Culture Technology
• /
• v.5 no.1
• /
• pp.64-69
• /
• 2017

The rapid growth of algae occurs can induce the algae bloom when nutrients are supplied from anthropogenic sources such as fertilizer, animal waste or sewage in runoff the water currents or upwelling naturally. The algae blooms creates the human health problem in the environment as well as in the water resource managers including hypoxic dead zones and harmful toxins and pose challenges to water treatment systems. The algal blooms in the source water in water treatment systems affects the drinking water taste & odor while clogging or damaging filtration systems and putting a strain on the systems designed to remove algal toxins from the source water. This paper propose the emerging In-Situ self-diagnosable smart algae sensing device with wireless connectivity for smart remote monitoring and control. In this research, we developed the In-Site Algae diagnosable sensing device with wireless sensor network (WSN) connectivity with Optical Biological Sensor and environmental sensor to monitor the water treatment systems. The proposed system emulated in real-time on the water treatment plant and functional evaluation parameters are presented as part of the conceptual proof to the proposed research.

• Journal of Auto-vehicle Safety Association
• /
• v.12 no.1
• /
• pp.52-61
• /
• 2020

In recent years, the increasing interest in health-care requires the industrial products to be well-designed ergonomically. In the commercial vehicle industry, several researchers have demonstrated the driver's posture has great effect on the orthopedic desease such as fatigue, back pain, scoliosis, and so on. However, the existing sensor systems developed for measuring the driver posture in real time have suffered from inaccuracy and low reliability issues. Here, we suggest our smart vehicle seat system capable of real-time driver posture monitoring by using the air bag sensor package with high sensitivity and reliability. The ergonomic numerical model which can evaluate a driver's posture has been developed on the basis of the human body segmentation method followed by simulation-based validation. Our experimental analysis of obtained pressure distribution of a vehicle seat under the different driver's postures revealed our smart vehicle system successfully achieved the driver's real-time posture data in great agreement with our numerical model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.7
• /
• pp.2414-2433
• /
• 2021

This article focuses on providing remedial solutions for COVID disease through the data collection process. Recently, In India, sudden human losses are happening due to the spread of infectious viruses. All people are not able to differentiate the number of affected people and their locations. Therefore, the proposed method integrates robotic technology for monitoring the health condition of different people. If any individual is affected by infectious disease, then data will be collected and within a short span of time, it will be reported to the control center. Once, the information is collected, then all individuals can access the same using an application platform. The application platform will be developed based on certain parametric values, where the location of each individual will be retained. For precise application development, the parametric values related to the identification process such as sub-interval points and intensity of detection should be established. Therefore, to check the effectiveness of the proposed robotic technology, an online monitoring system is employed where the output is realized using MATLAB. From simulated values, it is observed that the proposed method outperforms the existing method in terms of data quality with an observed percentage of 82.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.4
• /
• pp.169-176
• /
• 2021

Maritime monitoring requirements have been beyond human operators capabilities due to the broadness of the coverage area and the variety of monitoring activities, e.g. illegal migration, or security threats by foreign warships. Abnormal vessel movement can be defined as an unreasonable movement deviation from the usual trajectory, speed, or other traffic parameters. Detection of the abnormal vessel movement requires the operators not only to pay short-term attention but also to have long-term trajectory trace ability. Recent advances in deep learning have shown the potential of deep learning techniques to discover hidden and more complex relations that often lie in low dimensional latent spaces. In this paper, we propose a deep autoencoder-based clustering model for automatic detection of vessel movement anomaly to assist monitoring operators to take actions on the vessel for more investigation. We first generate gridded trajectory images by mapping the raw vessel trajectories into two dimensional matrix. Based on the gridded image input, we test the proposed model along with the other deep autoencoder-based models for the abnormal trajectory data generated through rotation and speed variation from normal trajectories. We show that the proposed model improves detection accuracy for the generated abnormal trajectories compared to the other models.