• Journal of the Korea Society of Computer and Information
• /
• v.29 no.3
• /
• pp.75-81
• /
• 2024

In this paper, we proposes a method for real-time processing of inter-floor noise problems by embedding TinyML, which includes a deep learning model, into ultra-low-power systems. The reason this method is feasible is because of lightweight deep learning model technology, which allows even systems with small computing resources to perform inference autonomously. The conventional method proposed to solve inter-floor noise problems was to send data collected from sensors to a server for analysis and processing. However, this centralized processing method has issues with high costs, complexity, and difficulty in real-time processing. In this paper, we address these limitations by employing On-Sensor AI using TinyML. The method presented in this paper is simple to install, cost-effective, and capable of processing problems in real-time.

• Journal of the Korean Institute of Gas
• /
• v.14 no.4
• /
• pp.56-62
• /
• 2010

The purpose of this paper analyzes and investigates the failure case studies of electronic control sensors for a LP gas engine. The malfunction of crank angle sensor, which controls a fuel injection volume of LP gas, displays an irregular and non-uniform pulse wave form. The pulse form, which is related to the noise of the crank angle sensor, displays at the rectangular peak with a saw-toothed shape and is intermittently generated with a level of 2.46V noise signal. The malfunction of No. 1 TDC sensor in which is caused from the internal disorder affects to the reduction of engine power and engine stop suddenly. If the malfunction of oxygen sensor is occurred due to a wiring problem of a sensor connector, the LP gas vehicle may produce a shaking and disharmony of an engine because of no signal supply from the oxygen sensor. The air cleaner replica produces the clogging of continuous supply of fresh air. This may cause the retardation of vehicle acceleration and engine disharmony intermittently.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.2
• /
• pp.113-118
• /
• 2011

In this paper, we develop a speaker gender classification technique using collaborative sensor fusion for use in a wireless sensor network. The distributed sensor nodes remove the unwanted input data using the BER(Band Energy Ration) based voice activity detection, process only the relevant data, and transmit the hard labeled decisions to the fusion center where a global decision fusion is carried out. This takes advantages of power consumption and network resource management. The Bayesian sensor fusion and the global weighting decision fusion methods are proposed to achieve the gender classification. As the number of the sensor nodes varies, the Bayesian sensor fusion yields the best classification accuracy using the optimal operating points of the ROC(Receiver Operating Characteristic) curves_ For the weights used in the global decision fusion, the BER and MCL(Mutual Confidence Level) are employed to effectively combined at the fusion center. The simulation results show that as the number of the sensor nodes increases, the classification accuracy was even more improved in the low SNR(Signal to Noise Ration) condition.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.13 no.3
• /
• pp.71-78
• /
• 2013

In this paper, we propose a method to infer the user's behavior and situation through collected data from multi-sensor equipped with a smart clothing and it was implemented as a smartphone App. User context reasoning and behavior determine is very difficult using single sensor depending on the measured value of the sensor varies from environmental noise. So, the reasoning and the digital filter algorithms to determine user behavior reducing noise and are required. In this paper, we used EWMA, Kalman Filter and SVM processing behavior for 3-axis value as a representative value of one.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.65 no.10
• /
• pp.1746-1754
• /
• 2016

In this paper, we provide a development and evaluation method for an archery posture analyzing system, using an inertial sensor. The system was developed using LabVIEW2014 by National Instruments and evaluated using the DTW algorithm. To convert the voltage value of the inertial sensor into a physical value, a coordinate transformation matrix bias was applied. To evaluate the similarity of movement in archery shooting, the DTW distance was calculated and similarity was confirmed based on simple mechanical movement, the same person's shooting movement, shooting movement with another person, and the noise signal. The average similarity comparison results were as follows: simple mechanical movement was 17.05%, the same person's shooting movement was 26.48%, shooting movement with another person was 62.8%, and the noise signal was 328.5%; a smaller value indicates a higher level of similarity. We confirmed the possibility of analyzing the archery posture using 3-axis acceleration of the inertial sensor. We inferred that the proposed method might be important means for assessing shooting skills, evaluation of archer's progress, and finding talented archers in advance.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2007.05a
• /
• pp.81-84
• /
• 2007

In this paper, fiber optic sound and vibration monitoring sensor which is latticed shape structure based on Sagnac interferometer is fabricated and tested in laboratory conditions. To detect external vibrations surface mounted fibers on the latticed steel wire fence with a dimension of 170cm by 180cm is used. To detect external sound frequency the tightened fiber optic itself wire netting fence with a dimension of 50cm by 50cm is used. Experiments for the detection of the excited vibration and sound signals were performed. A small vibrator induced external vibration signal and it is applied to the latticed structure in the range of 100Hz to several kHz. External sound signal applied to the fiber optic sensor net using non-directional sound speaker. The detected optical signals were compared and analyzed to the detected both accelerometer and microphone signals in the time and frequency domain. Based on the experimental results, distributed fiber optic sensor using Sagnac interferometer detected effectively external vibration and sound signal and had a good performance. This system can be expanded to the monitoring of a significant system and to the structural health monitoring system.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.27 no.2
• /
• pp.148-154
• /
• 2017

This paper discussed design for resonance avoidance of sensor plates of loose-parts monitoring systems (LPMS) in nuclear power plants (NPP). An LPMS monitors impact of loose parts in primary loop of NPP by using accelerometers, which is mounted on sensor plates. Resonance of the plates may cause false alarms at frequencies over 10 kHz, which can be misunderstood as impact signals of loose parts with small mass and cause unnecessary response of NPP operators. Modal analysis was carried out for the existing sensor plate and design parameters affecting natural frequencies were chosen. Frequency response functions of plates were analyzed by changing the parameters and the optimized plate design for avoiding resonance was determined. Experiments was carried out for the plate specimen with improved design and verified the proposed approach and design.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.18 no.8
• /
• pp.872-878
• /
• 2008

Many industrial operations require continuous or nearly-continuous operation of machines, interruption of which can result in significant cost loss. The condition monitoring of these machines has received considerable attentions in recent years. Rapid developments in semiconductor, computing, and communication with a remote site have led to a new generation of sensor called "smart" sensors which are capable of wireless communication with a remote site. The purpose of this research is to develop a new type of smart sensor for on-line condition monitoring. This system is addressed to detect conditions that may lead to equipment failure when it is running. Moreover it will reduce condition monitoring expense using low cost MEMS accelerometer. This system is capable for signal preprocessing task and analog to digital converter which is controlled by CPU. This sensor communicates with a remote site PC using TCP/IP protocols. The developed sensor executes performance tests for data acquisition accuracy estimations.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.17 no.11
• /
• pp.1028-1036
• /
• 2007

A method for measurement and monitoring of mechanical loads in large slender structures such as wind turbine blade and tower is presented based on continuous strain data obtained from distributed fiber optic sensor. An experimental study was carried out on an aluminum cantilever beam. A static load test was performed and the calculated moment from the distributed fiber optic sensor agree well with the actual applied moment. A series of damages was inflicted on the beam, and vibration tests were carried out for each damage case. The estimated natural frequencies from the distributed fiber optic sensor for each damage case are found to compare well with those from a conventional accelerometer and a numerical analysis based on an energy method.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.25 no.6
• /
• pp.440-446
• /
• 2015

Acoustic signal is emitted from a vessel and received by a cylindrical array sensor at some distance from the vessel. Acoustic signal is the source for a cylindrical array sensor which is designed to detect the acoustic signal. Cylindrical array sensors seldom have an ideal hydrodynamic shape and are not sufficiently robust to survive without some protection and they are normally housed in a sonar dome. Reflected signals by some structure inside a sonar dome make unwanted signals. Therefore, an acoustic baffle is used to minimize unwanted signals. The performance of the acoustic baffles can be determined from the acoustic numerical analysis at the design stage. In this study, finite element method was used to analyze the acoustic field around the cylindrical array sensor and baffle effects. The baffle performance can be defined the echo reduction. To show the baffle performance, the specimens were made for pulse tube test and echo reductions were measured during the test. In this paper, the effect of echo reduction of the acoustic baffle was discussed.