KSKyeong Su Shin

Kyeong Su Shin

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Kyeong Su Shin has a total of 8 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A Wi-Fi Channel State Information-Based Conveyor Belt Thickness Monitoring System

    by Kyeong Su Shin, Younho Nam, Young-Joo Suh

    Conveyor belt failures can inflict significant financial losses on operators by stalling operations. However, belt monitoring systems for preventing such failures are not widely adopted due to their cost and maintenance burden. To address this issue, we propose a Wi-Fi sensing–based conveyor belt wear monitoring system that continuously monitors changes in belt thickness to enable early detection or prediction of failures. By transmitting Wi-Fi frames through the belt and monitoring changes in Wi-Fi channel state information (CSI) caused by the dielectric properties of the belt material, the belt thickness can be estimated. As the belt wears and thins, Wi-Fi signals are less affected by its dielectric properties. Therefore, using a machine learning–based regressor, belt thickness can be estimated from CSI measurements. We validate this proposed approach through a prototype implementation on a conveyor belt testbed. A belt was intentionally ground down to various thicknesses and then installed on the testbed. Wi-Fi sensor units were mounted on the testbed, and data samples were collected. Using the obtained samples, we trained a multilayer perceptron–based belt thickness estimation model and achieved a root mean square error of 0.2292 mm. The proposed system offers a cost-effective and low-maintenance alternative to conventional belt monitoring solutions, as it does not require active electronic components to be mounted directly on the belt conveyor.

  • Open Access

    MIRA: A Transformer-Based Framework for Idler Roller Anomaly Detection and Localization

    by Younho Nam, Su Yeon Shim, Kyeong Su Shin, Young-Joo Suh

    Monitoring the condition of belt conveyor idlers is critical for ensuring safe and efficient operation of industrial conveying systems. However, existing methods often suffer from limited scalability and delayed fault detection, particularly under variable environmental conditions. In this work, we propose MIRA, a transformer-based framework for monitoring idler roller anomalies, which detects and localizes faults using acoustic and vibration signals collected from low-cost sensors. MIRA employs a masked transformer-based autoencoder trained in an unsupervised manner to reconstruct normal patterns and detect deviations via reconstruction error. MIRA can also infer the fault location, enabling spatially aware anomaly detection without the need for labeled data. We validated the system on a custom-built conveyor belt testbed equipped with sensor units, each measuring sound and two-axis vibration signals. We evaluated MIRA on four types of idler faults across 14 roller locations and 6 belt speeds. The results show that MIRA achieves an anomaly detection accuracy of 98.70% and a fault localization accuracy of 96.09%, demonstrating its robustness and practical applicability in complex operational settings.


  • Open Access

    Designing a Multivariate Belt Conveyor Idler Stall Detection and Identification System with Scalability Analysis

    by Kyeong Su Shin, Younho Nam, Young-Joo Suh

    Belt conveyor idlers are freely rotating idlers supporting the belt of a conveyor, and can induce severe frictional damage to the belt as they fail. Therefore, fast and accurate detection of idler faults is crucial for the effective maintenance of belt conveyor systems. In this article, we implement and evaluate the performance of an idler stall detection system based on a multivariate deep learning model using accelerometers and microphone sensor data. Emphasis is place on the scalability of the system, as large belt conveyor installations can span multiple kilometers, potentially requiring hundreds or even thousands of sensor units to monitor. The accuracy of the proposed system are analyzed and reported, along with its network bandwidth and energy requirements. The results suggest that while implementing accurate large-scale idler stall detection is feasible, careful consideration must be paid to observing the available network bandwidth and energy budget in order to avoid prolonged downtimes.


  • UbiGest: Smartphone-Based Ubiquitous Gesture Recognition With Wi-Fi

    by Seung-Hyun Jeong, Kyeong Su Shin, Jihun Park, Sanghyeok Jo, Young-Joo Suh

    Wi-Fi channel state information (CSI)-based gesture recognition systems have recently gained significant attention. One major challenge faced by these systems is that CSI measurements are affected by various conditions, including environmental factors and user orientation, leading to degraded sensing performance in unseen environments. Due to this challenge, most existing systems require multiple transceivers placed only a few meters apart from each other, and users are required to perform gestures within the proximity of these transceivers. Despite these systems robustly sensing users, their sensing range and mobility are severely restricted. In this article, we propose UbiGest, a smartphone-based gesture recognition system. UbiGest leverages beacon frames from multiple nearby APs over adjacent channels, enabling ubiquitous gesture recognition while mitigating the effects of the environment, user position, and user orientation. The system, even with limited computational resources, extracts features from noisy CSI data collected from a smartphone and robustly classifies a user’s gesture. By incorporating few-shot learning, the system maintains its robustness in challenging conditions, such as in entirely new environments with only one or two available APs. The extensive experimental results demonstrate that UbiGest achieves an accuracy of 95.3% in in-domain evaluation and 92.6% in cross-domain evaluation, on average.


  • Women's Empowerment Matters: Evidence from the Health and Education Sectors in Developing Worlds

    by Kyeong Min Shin, Yujin Lee, Jinhwan Oh

    This study examines the effect of female empowerment on child development in developing countries. Using comprehensive panel data from seventy-eight developing countries between 2000 and 2018, this research assesses the impact of political and economic empowerment of females on the health and education of children. The major findings include (1) that GDP per capita, health expenditure, and the official development assistance are not significantly reducing immunization, child mortality, and primary school enrollment; (2) that women's empowerment in political and economic fields is not significantly affecting the above-mentioned dependent variables; and (3) that when those two kinds of explanatory variables interact, women's empowerment is positively associated with immunization rate and primary school enrollment rate in countries with an income level lower than 1,500–2,000 USD, mainly located in East Africa. Such findings suggest that in certain economic levels, female empowerment and child development policies are complementary and should go hand in hand.