Jeonggil Ko

Yonsei University · School of Integrated Technology (SIT)

Federated Learning (FL) enables collaborative model training across distributed devices while preserving local data privacy, making it ideal for mobile and embedded systems. However, the decentralized nature of FL also opens vulnerabilities to model poisoning attacks, particularly backdoor attacks, where adversaries implant trigger patterns to mani...

While federated learning leverages distributed client resources, it faces challenges due to heterogeneous client capabilities. This necessitates allocating models suited to clients' resources and careful parameter aggregation to accommodate this heterogeneity. We propose HypeMeFed, a novel federated learning framework for supporting client heteroge...

This work presents AttFL, a federated learning framework designed to continuously improve a personalized deep neural network for efficiently analyzing time-series data generated from mobile and embedded sensing applications. To better characterize time-series data features and efficiently abstract model parameters, AttFL appends a set of attention...

This paper presents ArrhyMon, a self-attention-based LSTM-FCN model for arrhythmia classification from ECG signal inputs. ArrhyMon targets to detect and classify six different types of arrhythmia apart from normal ECG patterns. To the best of our knowledge, ArrhyMon is the first end-to-end classification model that successfully targets the classifi...

This work presents SafeFac, an intelligent camera-based system for managing the safety of factory environments. In SafeFac a set of cameras installed on the assembly line are used to capture images of workers that approach the machinery under hazardous situations to alert system managers and halt the line if needed. Given a challenging set of pract...

This work presents DIAMOND, a deep neural network computation offloading scheme consisting of a lightweight client-to-server latency profiling component combined with a server inference time estimation module to accurately assess the expected latency of a deep learning model inference. Latency predictions for both the network and server are compreh...

We introduce Monte-Carlo Attention (MCA), a randomized approximation method for reducing the computational cost of self-attention mechanisms in Transformer architectures. MCA exploits the fact that the importance of each token in an input sequence varies with respect to their attention scores; thus, some degree of error can be tolerable when encodi...

Diagnostic tests for hearing impairment not only determines the presence (or absence) of hearing loss, but also evaluates its degree and type, and provides physicians with essential data for future treatment and rehabilitation. Therefore, accurately measuring hearing loss conditions is very important for proper patient understanding and treatment....

Gaze tracking is a key building block used in many mobile applications including entertainment, personal productivity, accessibility, medical diagnosis, and visual attention monitoring. In this paper, we present iMon, an appearance-based gaze tracking system that is both designed for use on mobile phones and has significantly greater accuracy compa...

PurposeHepatic surface nodularity quantified on CT images has shown promising results in staging hepatic fibrosis in chronic hepatitis C. The aim of this study was to evaluate hepatic surface nodularity, serum fibrosis indices, and a linear combination of them for staging fibrosis in chronic liver disease, mainly chronic hepatitis B.Methods We deve...

In this work we present SUGO, a depth video-based system for translating sign language to text using a smartphone's front camera. While exploiting depth-only videos offer benefits such as being less privacy-invasive compared to using RGB videos, it introduces new challenges which include dealing with low video resolutions and the sensors' sensitive...

This paper presents a year-long study of our project, aiming at (1) understanding the work practices of clinical staff in trauma intensive care units (TICUs) at a trauma center, with respect to their usage of clinical data interface systems, and (2) developing and evaluating an intuitive and user-centered clinical data interface system for their TI...

There is growing research interest from many scientific, healthcare, and industrial applications toward the development of high-precision optical pH sensors that cover a broad pH range. Despite enthusiastic endeavors, however, it remains challenging to develop cost-effective, high-precision, and broadband working paper-strip-type optical pH measure...

Improvements in small sized sensors allow the easy detection of the presence of Volatile Organic Compounds (VOCs) in the air using easy-to-deploy Internet of Things (IoT) devices. However, classifying what VOC exists in the environment still remains as a complex task. Knowing what VOCs are in the air can help us remove the main cause that vents VOC...

Emerging cloud-based video streaming applications suggest limitations on current video streaming systems and emphasize the need to preserve and exchange high-quality contents while minimizing processing latency, given that both are critical in perpetuating a high user-perceived quality. This article presents 4KODEC, a codec for supporting real-time...

Multiplexed analysis allows simultaneous measurements of multiple targets, improving the detection sensitivity and accuracy. However, highly multiplexed analysis has been challenging for point-of-care (POC) sensing, which requires a simple, portable, robust, and affordable detection system. In this work, we developed paper-based POC sensing arrays...

This article presents LiteZKP a framework for supporting multiple anonymous payments using a smart contract-based zero-knowledge proof (ZKP) protocol on resource-limited devices. Specifically, to address challenges related to minimizing the computational overhead and offer a fully anonymous system, LiteZKP includes novel schemes such a new merkle t...

This work presents HeartQuake, a low cost, accurate, non-intrusive, geophone-based sensing system for extracting accurate electrocardiogram (ECG) patterns using heartbeat vibrations that penetrate through a bed mattress. In HeartQuake, cardiac activity-originated vibration patterns are captured on a geophone and sent to a server, where the data is...

To determine whether childhood intermittent exotropia (IXT) affects distance divergence and performance in block-building tasks within a virtual reality (VR) environment. Thirty-nine children with IXT, aged 6–12 years, who underwent muscle surgery and 37 normal controls were enrolled. Children were instructed to watch the target moving away and per...

As emojis are increasingly used in everyday online communication such as messaging, email, and social networks, various techniques have attempted to improve the user experience in communicating emotions and information through emojis. Emoji recommendation is one such example in which machine learning is applied to predict which emojis the user is a...

The ubiquitous deployment of smart wearable devices brings promises for an effective implementation of various healthcare applications in our everyday living environments. However, given that these applications ask for accurate and reliable sensing results of vital signs, there is a need to understand the accuracy of commercial-off-the-shelf wearab...

We present VitaMon, a mobile sensing system that can measure the inter-heartbeat interval (IBI) from the facial video captured by a commodity smartphone's front camera. The continuous IBI measurement is used to compute heart rate variability (HRV), one of the most important markers of the autonomic nervous system (ANS) regulation. The underlying id...

Limitations in battery capacity has held back the active development of novel applications for the Internet of Things (IoT) or have caused embedded systems researchers to design a number of “go-around” schemes, which sacrifice various system performance metrics for energy efficiency. However, with the concept of simultaneous wireless information an...

Internet‐of‐Things (IoT) devices are typically resource constrained in terms of computing capabilities and battery power. Despite the efforts from the Internet Engineering Task Force (IETF) to established standards for IoT such as IPv6 over low‐power wireless personal area networks (6LoWPAN), routing protocol for low‐power lossy networks (RPL), and...

With the advancements in ubiquitous computing, ubicomp technology has deeply spread into our daily lives, including office work, home and house-keeping, health management, transportation, or even urban living environments. Furthermore, beyond the initial metric of computing, such as "efficiency" and "productivity", the benefits that people (users)...

can enable futuristic applications including many Virtual Reality, Augmented Reality, and cloud gaming applications on resource constraint mobile devices. While RGR requires a high-level of networking bandwidth for seamless servicing, emerging high-speed communication technologies such as IEEE 802.11ax and millimeter wavebased communications are ex...

Electrocardiogram (ECG) signals offer rich information for analyzing and understanding the cardiac activity of a person. The continuous monitoring of ECG can help diagnose cardiac disorders, such as arrhythmia, effectively. While many wearable healthcare platforms offer continuous ECG monitoring, these devices are cumbersome in the fact that they n...

We present Reeboc that combines machine learning and k-means clustering to analyze the conversation of a chat, extract different emotions or topics of the conversation, and recommend emojis that represent various contexts to the user. Instead of simply analyzing a single input sentence, we consider recent sentences exchanged in a conversation. we p...

This work presents FDTLS, a security framework that combines storage and network/communication-level security for resource limited Internet of Things (IoT) devices using Datagram Transport Layer Security (DTLS). While coalescing storage and networking security scheme can reduce redundent and unnecessary operations, we identify security- and system-...

We present LpGL, an OpenGL API compatible Low-power Graphics Library for energy efficient AR headset applications. We first characterize the power consumption patterns of a state of the art AR headset, Magic Leap One, and empirically show that its internal GPU is the most impactful and controllable energy consumer. Based on the preliminary studies,...

We present LpGL, a Low-power Graphics Library designed to extend the usage time of mobile AR headsets. LpGL offers a transpatent layer to the application to intercept graphicsrelated calls to reduce unneeded graphics processing overhead without any quality loss for saving mobile device power. Our system reduces power consumption up to ∼22%, with on...

EXtended Reality(XR), which includes the concepts of virtual reality, augmented reality and mixed reality, is a promising technology for the research community and also the commercial domain in the sense that it can open a variety of new applications in a novel computing environment. Most XR applications are "interactive" and this interactivity is...

Intra-body Communication (IBC) is a communication method using the human body as a communication medium, in which body-attached devices exchange electro-magnetic (EM) wave signals with each other. The fact that our human body consists of water and electrolytes allows such communication methods to be possible. Such a communication technology can be...

For retailers, the ability to effectively manage dynamic pricing strategies is critical for various objectives such as managing customer demand, responding to competitors' pricing tactics, achieving internal price communication, maintaining multi-channel price integrity, and ultimately, maximizing revenue and profitability. In recent years, several...

Volatile organic compound (VOC) recognition systems can be helpful tools in monitoring today's living environments surrounded by harmful chemicals including dangerous VOCs. By designing a mobile system where users can easily detect VOC materials in their surroundings, people can avoid VOC-contained environments or take actions to improve their livi...

With the advancements in ubiquitous computing, ubicomp technology has deeply spread into our daily lives, including office work, home and house-keeping, health management, transportation, or even urban living environments. Furthermore, beyond the initial metric of computing, such as "efficiency" and "productivity", the benefits that people (users)...

By enabling driver mode on a smartphone, the smartphone can autonomously detect driving activities and suppress any incoming notifications so that drivers can focus on their attention on driving activities. However, the current driving mode implementations are limited in the fact that they frequently misjudge driving activities and also only have t...

Several recent mobile operating systems allow users to configure the smartphone into a "driving mode". This mode suppresses the smartphone's incoming SMS/call notifications so that it does not distract driving activities. However, currently available driving mode implementations keep all notifications from being delivered, which decreases its pract...

Certificate-based Public Key Infrastructure (PKI) schemes are used to authenticate the identity of distinct nodes on the Internet. Using certificates for the Internet of Things (IoT) can allow many privacy sensitive applications to be trusted over the larger Internet architecture. However, since IoT devices are typically resource limited, full size...

With the introduction of various advanced deep learning algorithms, initiatives for image classification systems have transitioned over from traditional machine learning algorithms (e.g., SVM) to Convolutional Neural Networks (CNNs) using deep learning software tools. A prerequisite in applying CNN to real world applications is a system that collec...

Capsule endoscopy identifies damaged areas in a patient’s small intestine but often outputs poor-quality images or misses lesions, leading to either misdiagnosis or repetition of the lengthy procedure. The authors propose applying deep-learning models to automatically process the captured images and identify lesions in real time, enabling the capsu...

With the exponential improvement of software technology during the past decade, many efforts have been made to design remote and personalized healthcare applications. Many of these applications are built on mobile devices connected to the cloud. Although appealing, however, prototyping and validating the feasibility of an application-level idea is...

Understanding the engagement levels players have with a game is a useful proxy for evaluating the game design and user experience. This is particularly important for mobile games as an alternative game is always just an easy download away. However, engagement is a subjective concept and usually requires fine-grained highly disruptive interviews or...

Recent advances in machine learning based data analytics are opening opportunities for designing effective clinical decision support systems (CDSS) which can become the "third-eye" in the current clinical procedures and diagnosis. However, common patient movements in hospital wards may lead to faulty measurements in physiological sensor readings, a...

This paper proposes an efficient architecture of HEVC in-loop filters (ILFs) with the target of providing effective multicore utilization for ultra-high definition video applications. While HEVC allows for a high level of parallelization, the issue of data dependencies at the ILF leads to inefficient parallel processing performance. The novel memor...

Objectives Biosignal data include important physiological information. For that reason, many devices and systems have been developed, but there has not been enough consideration of how to collect and integrate raw data from multiple systems. To overcome this limitation, we have developed a system for collecting and integrating biosignal data from t...

RPL is the IPv6 routing protocol for low-power and lossy networks (LLNs), standardized by IETF in 2012 as RFC6550. Specifically, RPL is designed to be a simple and inter-operable networking protocol for resource-constrained devices in industrial, home, and urban environments, intended to support the vision of the Internet of Things (IoT) with thous...

Many low-power wireless network system deployments are planned on a two-dimensional plane, while in reality, we live in a three-dimensional space. Therefore, although it is essential to well consider the impact of height on the overall wireless system performance, this aspect has often been overlooked if not neglected with simplifying assumptions....

With the advancement of technology in various domains, many efforts have been made to design advanced classification engines that aid the protection of civilians and their properties in different settings. In this work, we focus on a set of the population which is probably the most vulnerable: children. Specifically, we present ChildSafe, a classif...

Recent improvements in data learning techniques have catalyzed the development of various clinical learning systems. However, for clinical applications, training from noisy data can cause significant misleading results, directly leading to potentially dangerous clinical decisions. Given its importance, this work targets to present a preliminary eff...

For the past decade, wireless sensor networks have focused primarily on data collection. As a result the network topology for these systems was usually heavily centralized. However, for these networks to form a full system, the introduction of proper actuation units and decision-making intelligence is inevitable. Such a new wireless sensor and actu...

While a number of studies reveal the performance and effectiveness of applying wireless systems to various smart city applications, surprisingly, a market environment, in which we rely on a daily or weekly basis for purchasing essential goods, is still understudied. A wireless system in a market, along with rapidly growing IoT technology, can enabl...

With the wide-distribution of smart wearables, it seems as though ubiquitous healthcare can finally permeate into our everyday lives, opening the possibility to realize clinical-grade applications. However, given that clinical applications require reliable sensing, there is a need to understand how accurate healthcare sensors on wearable devices (e...

Recent advancesinhardware and software had led the smartphonesto becomean attractive candidate as mobile gateways and data-users for wireless sensor networks that have the capability to monitor the physical phenomena in various dimensions. Together, wireless sensing systems and smartphones enable services that allow users to gather and process fine...

Bio-signals can be crucial evidence in detecting urgent clinical events. However, until now, access to this data was limited. We aim to construct and provide a new open bio-signal repository with data gathered from more than 40 intensive care unit (ICU) beds. For doing so, we completed the interfacing system with the patient monitors at the target...

Recent advances in smartphone processing power have opened the possibilities for them to act as the processing component of software-defined radios (SDRs)For low-power sensor network systems using various communication protocols, this means that smartphones, when equipped with an SDR, can be their system management end-devices, (potentially) withou...

Near Field Communication (NFC) is a wireless communication technology using 13.56 MHz to support 2-way communications between two devices within ~10 cm. Such a short communication range may be considered as a shortcoming, but at the same time, this enables a secure data transfer within the connectivity region when compared to Bluetooth Low Energy (...

Smart watches are increasingly being used in various applications to monitor heart rate for exercise and health care purposes. It is crucial that the readings from these devices are accurate so that users can take proper actions according to the intensity of the heart rate. Taking actions from inaccurate readings can negatively impact the health of...

Analyzing large quantities of bio-signal data can lead to new findings in patient status diagnosis and medical emergency event prediction. Specifically, improvements in machine learning schemes suggest that by inputting clinical waveforms, designing mechanisms to predict medical emergencies, such as ventricular arrhythmia or sepsis, can soon be pos...

Bluetooth Low Energy (BLE) and the iBeacons have recently gained large interest for enabling various proximity-based application services. Given the ubiquitously deployed nature of Bluetooth devices including mobile smartphones, using BLE and iBeacon technologies seemed to be a promising future to come. This work started off with the belief that th...

In low-power wireless networks, maintaining multihop connectivity is considered effective in constructing communication routes between individual nodes to a gateway. Since sensor networks are typically used for data collection, multihop routing protocols are designed to find routes optimal in upward directions. As sensor networks become widely appl...

In the next generation wireless communication systems, an energy harvesting from radio frequency signals is considered as a method to solve the lack of power supply problem for sensors. In this paper, we try to propose an efficient algorithm for simultaneous wireless information and power transfer in energy harvesting networks with channel estimati...

The improvement in hardware capabilities of mobile devices has led to the active use of processor-heavy contents, such as multimedia files, on resource-limited platforms. In addition to simply enjoying such contents on mobile devices, recently commercialized protocols allow the real-time sharing of multimedia (or a mobile device’s screen contents)...