I am a Ph.D. student in Computer Science at Georgia Institute of Technology working with Dr. Gregory Abowd and Dr. Rosa Arriaga.
I was a Visiting Researcher at Carnegie Mellon University and conducted two research projects: the first aims to measure the cognitive load of drivers who encounter distractions, and the second develops novel emotion assessment methods, such as wearable sensors, for college students who have depression and anxiety.
Prior to CMU, I developed wearable and smart devices at LG Electronics as a Research Engineer. I was also a Visiting Researcher at Mobile and Pervasive Computing Laboratory at the University of Florida, and researched pervasive computing technologies under the guidance of Dr. Sumi Helal.
I received a Bachelor of Science in Electrical Engineering and a Master of Science in Electrical and Computer Engineering from Ajou University, South Korea.
Email _ firstname.lastname@example.org
Web _ www.jungwook.com
Doctor of Philosophy
in Computer Science
Georgia Institute of Technology, Atlanta, GA, USA, [2016 - Current]
Master of Science
in Electrical and Computer Engineering
Ajou University, South Korea, 
Bachelor of Science
in Electrical Engineering
Ajou University, South Korea, 
Yonsei University Health System
Visiting Researcher [2015 - 2016]
Human-Computer Interaction Institute
Carnegie Mellon University
Research Engineer [2012 - 2015]
Creative Innovation Center
CTO Division, LG Electronics
Visiting Researcher [2008 - 2010]
Mobile and Pervasive Computing Lab
University of Florida
UbiMi 2016, Germany
Web Chair 
ACM MobiCom 2013, USA
Online Community Chair 
ACM Ubicomp 2009, USA
 CHI EA
 MDPI Sensors
 IEEE EMBC, IEEE JBHI, JAISE
 ACM Ubicomp, MELT, Percom
BRIEF | Onset of most mental disorders appears between ages 15 and 24. However, many students who need to get appropriate psychological treatments are not clearly aware their mental disorder. Thus, early detection of depression is a critical part of improving the efficacy of mental health management. In order to recognize depression at an early stage, I am looking to explore novel sensing technologies and modeling approaches. For example, I am developing a UV sensor-enabled wristband to estimate students’ vitamin D levels.
BRIEF | Interruptions while driving, such as text messages, can be quite perilous. These distractions increase driver workload and reduce performance on the primary driving task. Being able to identify when a driver is susceptible to interruptions is critical for building systems that can mediate these interruptions. In this project, I develop a driving sensing platform that integrates four different Bluetooth sensors and one Bluetooth-LE haptic feedback actuator. Preliminary results of the project demonstrate the effectiveness of the platform.
BRIEF | As the potential of wristband technologies is still being explored, LG created Smart Band to be a more effective and convenient device to manage user's health status. I developed firmware, device drivers, and Bluetooth-LE functions for a custom-designed ultra low power microprocessor in the Smart Band project. The Final outcome of the project is not commercialized, but utilized as an advanced wearable prototype inside LG Electronics. The form factor of the device is similar to the image on the left.
BRIEF | Imagine a drawer that contains your most important belongings. I have developed IoT(Internet of Things) sensors that can be placed into the drawer, and protect against unauthorized access using vibration monitoring function. My IoT research outcomes can make any mundane appliance “smarter.” In IFA 2015, LG introduced a slightly different version of my research outcome, which called Smart ThinQ Sensor.
BRIEF | Since previous heart rate sensors have been uncomfortable and restricting on everyday use, we developed a novel heart rate (HR) monitoring approach in which we measure the pressure variance of the surface of the ear canal. In order detect pressure variance in a comfortable way, a scissor-shaped apparatus and a customized piezoelectric film sensor were designed for high wearability. In the proposed device, the film sensor converts in-ear pulse waves (EPW) into electrical current, and the circuit module enhances the EPW and suppresses noise. A real-time algorithm performs morphological conversions and knowledge-based rules are used to detect EPW peaks.
INTRODUCTION | Cardiovascular disease (CVD), which includes hypertension, myocardial infarction, coronary artery diseases, and stroke, has become a serious problem and is currently one of the major causes of disability and death globally. In fact, more than 17 million people worldwide die each year because of CVD, and it accounts for about 30% of all deaths per year. While many of the risk factors for CVD are known, such as tobacco use, physical inactivity, obesity, and diabetes, resting heart rate (HR) is one of the simplest cardiovascular parameters and an independent risk factor. Resting HR has prognostic importance in that an elevated resting HR is strongly related with mortality in the general population, the elderly, and patients with myocardial infarction, congestive heart failure, diabetes, or hypertension. Thus, frequent HR monitoring in everyday life is necessary to assess and prevent cardiovascular problems in advance.
Vital signs such as HR, respiration, and blood pressure were in the past possible to measure by medical systems in hospitals. However, recent technological advances in physiological sensors, processing capabilities, and wireless communications have enabled individuals to monitor their health status outside of hospitals via wearable biosensor systems. Compared with previous medical devices, wearable sensor systems provide nonintrusive solutions for long-term monitoring without constraints on time and place. A further advantage is that such systems can offer real-time feedback about their health status to the patients themselves or even a professional physician at a hospital. Therefore, wearable HR monitoring systems play an important role in the monitoring of cardiovascular exacerbations in the early stage, during catastrophes that may occur with high risk individuals, or during emergency situations.
Many technologies and methodologies have been introduced for wearable HR monitoring. Among these technologies, those that monitor photoplethysmogram (PPG), which is obtained by optical detection of blood volume changes in the microvascular bed of the tissue using light-emitting diodes and photodetectors, are the most widely utilized. PPG sensors have mainly been applied to fingers, earlobes, or wrists in the form of tweezers- or watch-shaped apparatuses. Also, a customized ear mold equipped with a reflective PPG sensor has been applied to the ear channel. Conventional measurement of electrocardiogram (ECG), which is the gold-standard method for HR monitoring, is not appropriate for wearable sensing because of the adhesive electrodes and wires utilized. However, recent studies have proposed and investigated T-shirts and band type wearable ECG systems that use conductive fabric, flexible printed circuit boards, or textile circuits. In addition, stethoscopes that use a microphone to measure phonocardiography have also been proposed as wearable and portable devices.
However, despite significant progress in wearable systems, various issues that impede extensive application have been reported. These issues include bulky sensors, connections to additional hardware modules, skin irritation, and laundry. In order to solve these problems, Teichmann et al. recently presented a bendable and flexible device that can be placed in a shirt pocket using noncontact sensor principles. However, very few studies have presented a high acceptance rate by users, and there are still many challenges that have to be resolved in order for wearable systems to become accepted by normal people and patients. For examples, existing methods have limitations to utilization in particular situations when the user is running or swimming in which hands and feet are excessively moving. The human head is rather the least moving and least affected part by strenuous exercise or movement according to circumstances. Thus, new methods or alternatives are required for vigorous follow-up studies and applications.
In this project, we present our proposed wearable HR monitoring sensor device developed based on pressure variance in the ear canal surface. The ear is an excellent site for users to accept a wearable device. In addition, the ear canal is partially composed of cartilage and bone and located in the temporal bone, thus providing the proper conditions for anchoring the sensor apparatus. We also present the physiological principle that enables HR monitoring from the ear canal surface, the material comprising the sensor, design and implementation of the wearable sensor device, the verification process and the results obtained, and our conclusions.
BRIEF | More than 3 million children experience an unintentional home accident every year. This project presents a child activity recognition approach that prevents child accidents such as falls. These injuries can be prevented by wearable sensors and activity recognition algorithms. The overall accuracy of our activity recognition was 98.43% using only a single-wearable triaxial accelerometer sensor and a barometric pressure sensor with a support vector machine.
INTRODUCTION | As babies usually start walking between 9 and 16 months, they are at risk of falling from furniture or stairs. As toddlers learn to climb, they are at risk of falling from windows and beds. Falls are a frequent cause of injury in children. Accident and emergency departments and outpatient surveillance systems show that falls are one of the most common mechanisms of injuries that require medical care, and the most common nonfatal injury that at times needs hospitalisation. In children younger than four years of age, most fall-related injuries occur at home. Thus, a new safety management method for children is required to prevent child home accidents. Since the major causes of fall-related injuries change as a child grows and develops, fall prevention needs to be addressed. One of the most challenging issues in this context is to classify daily activities of children into safe and dangerous activities.
We have developed a wearable sensor device and a monitoring application to collect information and to recognize baby activities. We classified baby activities into 11 daily activities which are wiggling, rolling, standing still, standing up, sitting down, walking, toddling, crawling, climbing up, climbing down, and stopping. Multiple sensors embedded in a wearable device are more accurate for collecting different types of sensing information , but would be very inconvenient for users. For this reason, we present only one single unit of sensor nodes, which collects multiple types of information. The nature of information interaction involved in sensor fusion can be classified as competitive, complementary, and cooperative fusion –. In competitive fusion, each sensor provides equivalent information about the process being monitored. In complementary fusion, sensors do not depend on each other directly, as each sensor captures different aspects of the physical process. The measured information is merged to form a more complete picture of the phenomenon. Cooperative fusion of the two sensors enables recognition of the activity that could not be detected by each single sensor. Due to the compounding effect, the accuracy and reliability of cooperative fusion is sensitive to inaccuracies in all simple sensor components used. In this project, we select the cooperative fusion model to combine information from sensors to capture data with improved reliability, precision, fault tolerance, and reasoning power to a degree that is beyond the capacity of each sensor.
The main contributions of this project over the earlier previous work are 1) to extend the method to work with arbitrary every day activities not just walking by improving the feature selection and recognition procedure; 2) to perform evaluation on a large (50 h) dataset recorded from real life activities; 3) to have studied ten divers subjects: 16, 17, 20, 25, 27 months-old baby boys and 21, 23, 24, 26, 29 months-old girls; and 4) to employ a barometric pressure sensor for improving upon the previous algorithms. The proposed method classified daily physical activity of children by a diaper worn device consisting of a single-triaxial accelerometer and a barometric air pressure sensor. We demonstrate our improvements in comparison to the accuracy results of only a single-wearable device and multiple feature sets to find an optimized classification method.
BRIEF | Recognition of sleep patterns is very important for numerous reasons such as the treatment of sleep disorders and the assessment of sleep quality. DreamSleep, a wearable sleep sensor, is developed to monitor sleep patterns in a more comfortable way. It is composed of two different parts: i) a wearable sleep sensor and ii) an Android application. The wearable sleep sensor is to be attached to the user’s abdominal area. Then it senses user movements and snoring sounds to recognize sleep status. Main features of the application include visualization of sleep patterns, database for sensing data, and Bluetooth communication.
Sleep Disorders Detection | The finite impulse response – low pass filter (LPF) is commonly applied to reduce high frequency noise, then the knowledge based rule is utilized to detect sleep disorders. In the case of snoring detection, morphological conversion/transformation of the raw wave signal into a more robust signal that is easy to process is adopted.
The microphone sensor output amplitude is varied according to the user’s snoring. As Fig 1 describes the sample snoring waveforms, the Absolute snoring signal (AbsSnr) can be derived by taking the absolute value from the raw snoring signal (RawSnr). The decimation process is applied to AbsSnr to reduce the amount of data. Finally, the decimation snoring signal (DcmSnr) passes LPF (cut off frequency: 50Hz), then the knowledge (interval, adaptive threshold) based peak detection algorithm is applied to detect snoring.
As the user breathes during sleep, the abdominal region repeatedly, expands and contracts. Accordingly, the output signal of the z-axis accelerometer, attached on the abdominal area is also changed, as Fig 2 presents the sample breathing signal. LPF (cut off frequency: 10Mz) is applied to the raw breathing signal (RawBrt), then knowledge based criteria (threshold, time, amplitude) are adopted to determine the unchanged section in the filtered breathing signal (FltBrt). An unchanged section of more than 10 seconds is regarded as an apnea.
When the user tosses and turns, his entire body rotates to the left or right side. The output signal of the y-axis accelerometer is therefore changed, as the body rotates. If the body rotates to the right or left side, the output signal has a plus or minus value, respectively. Thus, the filtered ‘toss and turn’ signal (FltTst) value changes.Form Factor
BRIEF | Management services for pregnant women have been in high demand because woman have displayed strong desire to control unfamiliar parts of their lives such as pregnancy and motherhood. DreamMom is developed to provide total management services for these pregnant women. Main features of the application include pregnancy diary, hospital review, weight management, step-by-step recipes, and pregnancy social media service. The social media service allows pregnant women to share their questions or experiences with each other in a way that is comforting during a new era of their lives.
BRIEF | Drug information should be easily accessible at any time. For these purpose, I developed a mobile application, DrugInfo, that includes a drug information database jointly developed by BIT Computer and Seoul National University in South Korea. It contains the latest information on pharmaceutical products including drug usage guide, patient education on disease, drug identification by its image, drug administration for pregnant women, and scientific information of drugs.
BRIEF | If a patient has metabolic syndrome, he needs to provide a large amount of personal information related to his disease periodically. Unfortunately, these recurrent questionnaires are not only making him feel uncomfortable, but also becoming prohibitive. To solve the challengeable problems, we attempted to research numerous approaches with respect to the diagnosis and treatment of metabolic syndrome with the Ajou University Hospital. As a result of this research, we developed a specialized ‘Well-being index model’ for treatments of metabolic syndrome, and also designed a context-aware system to deploy the index model in the real world.
BRIEF | Developmental disabilities such as Mucopoly-saccharidoses (MPS) require minute-to-minute, burdening care on the part of the parents. Often, one of the parents has to be dedicated 100%, 24/7 to support and sustain their child’s life. To enhance children’s cognitive ability in expressing some of their desires, we designed a simple DVD controller in a specific form factor suitable to MPS children. Our design was participatory, allowing one family with MPS children to provide guidelines. TouchView is intended to provide intuitive way for MPS children to express their wishes and desires of watching a particular movie.
BRIEF | A ubiquitous environment such as the Gator Tech Smart House (GTSH) provides services that incorporate many sensors and smart appliances. These smart objects lend themselves to other services and allow interaction with users. However, many mundane objects do not have such capabilities. In this project, we demonstrate the SmartPlug, a tool deployed at the GTSH that integrates everyday appliances such as lamps and fans into the smart house through power outlets and switches. We show a design of the SmartPlug and the automatic integration process through the Atlas middleware and the Device Description Language (DDL).
BRIEF | To promote personal health for the elderly and the disabled, and to support independent living at reasonable cost, it is agreeable that a home-based sensor network that collects various data and vital signs of the residents is a promising approach. In this project, networked MEMS accelerometer sensors are considered as a superior technology for localizing footstep source and computing correlation between the level of energy expenditure and the level of floor vibration.
BRIEF | In a ubiquitous environment, system makes their own decision without or with minimized user interaction to provide required services to users. To fulfill this requirement the system needs to collect the proper information of surroundings and define how to react to the changes based on found information autonomously. To achieve this process, the system includes three basic components that are sensing infrastructure, context aware or intelligent decision, and smart services. Main purpose of this research is how to enable this process efficiently and suggest the way of developing smart space.
BRIEF | Since medication taking is such a routine and mundane behavior, self-monitoring of medication taking is an onerous and difficult task. In order to manage the task more effectively and efficiently, I developed a sensor- augmented cup, which called SmartCup. This cup consists of a tilt sensor that can detect if the cup is shaking and an implemented OLED display to notify medication alarms. A central server manages a user’s medication schedule, and sends an alarm signal to the SmartCup. Then the SmartCup keeps sensing its shaking status to recognize user’s intake motion.
BRIEF | In class of Sensor and Actuator Engineering in 2006, I designed and implemented an augmented glove that can recognize user’s hand gestures by using two flex sensors and two gyro sensors. A PC application gathered real-time sensing data through Bluetooth communication and fused the two different sensing data to control a mouse cursor.