About

Sangjin Hong is a Professor at the Department of Electrical and Computer Engineering at Stony Brook University. His research focuses on low-power VLSI design of multimedia wireless communications and digital signal processing systems, including system-on-chip (SOC) design methodology and optimization. His work involves developing efficient hardware architectures and design techniques to improve the performance and energy efficiency of communication and signal processing systems, contributing to advancements in wireless technology and multimedia applications.

Research topics

• Artificial Intelligence
• Computer Science
• Computer vision
• Real-time computing
• Mathematics education
• Psychology
• Data science
• World Wide Web

Selected publications

• The Role of Veto Players Theory on Decision Making Process of the US Troops’ Withdrawal from Afghanistan

  Journal of Northeast Asian Studies · 2024-06-30

  article1st authorCorresponding

  본 연구는 미국이 역사상 최장 기간인 20년 동안 아프가니스탄에서 병력을 철수하지 못한 원인을 밝히는데 목적이 있다. 아프가니스탄 전쟁이 장기화된 원인을 미국의 내부적인 요인을 중심으로 살펴보는데, 체벨리스(George Tsebelis)의 거부권 행사자 이론을 적용한다. 거부권 행사자 이론을 적용한 것은 기존 외교정책 분석이 아프가니스탄 철군 정책이 지연된 원인을 잘 설명하지 못한데서 비롯되었다. 외교정책 분석과는 별개로 국제 정치에서 자주 인용되는 구조적 현실주의(structural realism) 이론도 설명력에 한계가 있다. 이에따라, 거부권 행사자 이론을 적용하여 정책결정 과정에 참여한 행위자들의 상호 작용을 중심으로 설명하면서 정책의 안정성 문제를 도출한다. 정책의 안정성은 아프가니스탄 철군 정책이 장기간 지연된 현상을 규명한다. 본 연구의 결론은 외교정책 결정 과정을 설명하는 데 있어서 행위자 중심 분석의 적용성을 보여주고, 향후 외교정책을 수립하는데 함의를 제공한다.

  PublisherDOI

• Development of a New PVC-Gel Actuator for Implementing Emotional Vibration

  SAE technical papers on CD-ROM/SAE technical paper series · 2023-04-06

  article

  <div class="section abstract"><div class="htmlview paragraph">Electro-hydraulic actuators, a type of soft actuators, can provide soft-touch vibrations due to their structural characteristics, but some problems need to be improved to apply them to vehicles. That is, it is necessary to increase excitation force, expand frequency band, lower driving voltage, and increase durability. This research aims to design a new type based on electro-hydraulic actuator and improve problems with its performance to develop a product that generates emotional vibration in vehicles. First, a new mechanism and design of an electro-hydraulic actuator called a PVC-gel film actuator are proposed. This actuator uses PVC-gel as a film which covers a dielectric liquid and uses carbon nanotube as a cathode material. In addition, a method of manufacturing an actuator with improved performance has been proposed by creating and testing prototypes with different sizes and material properties. It has been verified that the proposed actuator improves excitation force, frequency band, driving voltage, and durability compared to the existing actuator. Next, as a result of investigating the object that this actuator can exert the most effect on a vehicle based on body sensitivity, a device has been developed that built into a neck pillow in front of the vehicle seat and linked to the music played on the vehicle to generate a comfortable vibration in the neck. In addition, four vibration scenarios have been constructed to generate vibrations according to various genres of music. Finally, as a result of testing on a vehicle with a large number of people, it has been verified that the vibration generating device developed in this study can provide a new and emotional vibration experience for users.</div></div>

  PublisherDOI

• Real-time realizable mobile imaging photoplethysmography

  Scientific Reports · 2022 · 16 citations

  • Computer Science
  • Artificial Intelligence
  • Computer Science

  Photoplethysmography imaging (PPGI) sensors have attracted a significant amount of attention as they enable the remote monitoring of heart rates (HRs) and thus do not require any additional devices to be worn on fingers or wrists. In this study, we mounted PPGI sensors on a robot for active and autonomous HR (R-AAH) estimation. We proposed an algorithm that provides accurate HR estimation, which can be performed in real time using vision and robot manipulation algorithms. By simplifying the extraction of facial skin images using saturation (S) values in the HSV color space, and selecting pixels based on the most frequent S value within the face image, we achieved a reliable HR assessment. The results of the proposed algorithm using the R-AAH method were evaluated by rigorous comparison with the results of existing algorithms on the UBFC-RPPG dataset (n = 42). The proposed algorithm yielded an average absolute error (AAE) of 0.71 beats per minute (bpm). The developed algorithm is simple, with a processing time of less than 1 s (275 ms for an 8-s window). The algorithm was further validated on our own dataset (BAMI-RPPG dataset [n = 14]) with an AAE of 0.82 bpm.

  PublisherOA PDFDOI

• Dynamic Diagnosis of the Progress and Shortcomings of Student Learning using Machine Learning based on Cognitive, Social, and Emotional Features

  arXiv (Cornell University) · 2022 · 2 citations

  • Computer Science
  • Artificial Intelligence
  • Computer Science

  Student diversity, like academic background, learning styles, career and life goals, ethnicity, age, social and emotional characteristics, course load and work schedule, offers unique opportunities in education, like learning new skills, peer mentoring and example setting. But student diversity can be challenging too as it adds variability in the way in which students learn and progress over time. A single teaching approach is likely to be ineffective and result in students not meeting their potential. Automated support could address limitations of traditional teaching by continuously assessing student learning and implementing needed interventions. This paper discusses a novel methodology based on data analytics and Machine Learning to measure and causally diagnose the progress and shortcomings of student learning, and then utilizes the insight gained on individuals to optimize learning. Diagnosis pertains to dynamic diagnostic formative assessment, which aims to uncover the causes of learning shortcomings. The methodology groups learning difficulties into four categories: recall from memory, concept adjustment, concept modification, and problem decomposition into sub-goals (sub-problems) and concept combination. Data models are predicting the occurrence of each of the four challenge types, as well as a student's learning trajectory. The models can be used to automatically create real-time, student-specific interventions (e.g., learning cues) to address less understood concepts. We envision that the system will enable new adaptive pedagogical approaches to unleash student learning potential through customization of the course material to the background, abilities, situation, and progress of each student; and leveraging diversity-related learning experiences.

  PublisherOA PDFDOI

• Quinol oxidases in the superfamily of heme-copper oxidoreductases and the cryoEM structure of the cytochrome bo3 ubiquinol oxidase from E. coli

  Biochimica et Biophysica Acta (BBA) - Bioenergetics · 2022-08-20

  article
  PublisherDOI

• Mobile Remote Photoplethysmography: A Real-Time Realization Perspective

  Research Square (Research Square) · 2021

  • Artificial Intelligence
  • Computer Science
  • Computer vision

  Abstract Remote photoplethysmography (rPPG) sensors have attracted a significant amount of attention as they enable the remote monitoring of instantaneous heart rates (HRs) and thus do not require any additional devices to be worn on fingers or wrists. In this study, we mounted rPPG sensors on a robot for active and autonomous instantaneous HR (R-AAIH) estimation. Subsequently, we proposed the algorithm providing accurate instantaneous HRs, which can be performed in real time with vision and robot manipulation algorithms. By simplifying the extraction of facial skin images using saturation (S) values in the HSV color space, and selecting pixels based on the most frequent S value on the face image, we achieved reliable HR assessment. The results of the proposed algorithm using the R-AAIH were evaluated by rigorous comparison with the results of existing algorithms on the UBFC-RPPG dataset (n = 42). Our algorithm exhibited an average absolute error (AAE) of 0.71 beats per minute (bpm). The developed algorithm is simple and the processing time is less than 1 s (275 ms for an 8-s window). The algorithm was further validated on our own dataset (BAMI-RPPG dataset [n = 14]) with an AAE of 0.82 bpm.

  PublisherOA PDFDOI

• A Secure IoT-Fog-Cloud Framework Using Blockchain Based on DAT for Mobile IoT

  2019-10-01 · 11 citations

  articleSenior author

  Internet of Things (IoT) devices are becoming important and familiar products in our everyday lives. Yet, many of these IoT devices have a multitude of vulnerabilities that create a security risk, despite being products that bring a lot of convenience to our lives. Blockchain (BC) technology is gaining popularity these days due to the importance of data security. However, this BC technology is not suitable for IoT devices, because it requires a lot of computing power also IoT devices uses a limited amount of memory and has a lower performing CPU for cost reasons. Also, IoT devices collect a large amount of data and transmit that data every hour and every day to a cloud system to analyze or process the collected data. For this reason, network latency is an important factor in a cloud computing system. In this paper, we propose a secure and distributed 10T-Fog-Cloud Framework using BC to compensate for the security weaknesses of IoT nodes, the delay aware tree construction (DATC) algorithm that considers the service delay for fog-cloud computing, and the mobility of mobile IoT nodes for BC to deal with the triangular routing problem. To verify this proposed framework, we performed security analysis and experiments for the effectiveness of the algorithm through simulation and confirmed that the overall service delay was reduced by choosing the minimum delay path by the DATC algorithm.

  PublisherDOI

• Patient-Provider Interaction System for Efficient Home-Based Cardiac Rehabilitation Exercise

  IEEE Access · 2019-01-01 · 14 citations

  articleOpen access

  This paper presents an efficient home-based cardiac rehabilitation (CR) exercise system for patient-healthcare provider interaction in cardiac management. The proposed system consists of a wearable device, a smartphone application, and a medical station. A CR specialist prescribes CR exercise for a patient through a medical station. The prescription information is sent to a patient's smartphone application with Internet's hypertext transfer protocol. The information is further transferred to a patient's wearable device via Bluetooth communication. A patient performs home-based CR exercise with the worn wearable device that measures instantaneous heart rates (HRs) accurately even during intensive physical exercise. During the CR exercise, the wearable device recommends appropriate exercise intensity in real time based on the prescribed target HR zone and the measured instantaneous HRs. Furthermore, the wearable device informs the current exercise stage (resting, walking, or running), allowing home-based CR exercise without the help of a medical staff. Upon completion of the CR exercise, exercise records are sent to a smartphone application so that a patient can monitor his or her own exercise records. Exercise records are also sent to the medical station where a CR specialist can monitor these records and provide new exercise prescription. In this paper, we have implemented and demonstrated the full functionality of the overall home-based CR exercise system consisting of a wearable device, a smartphone application, and a medical station. Moreover, we have verified the overall operation of the system and that the proposed system can be an efficient CR exercise tool for remotely sharing CR exercise prescription and exercise records between a patient and a healthcare provider.

  PublisherOA PDFDOI

• Dynamic Load Balancing Mechanism in Mobile Gateway with Heterogeneous Network

  2018-10-01 · 2 citations

  article

  Due to the increasing popularity and the need for cloud computing systems, new paradigms and models such as fog and edge computing have recently emerged. The researches on mobile edge computing, including 5G, are focused on virtual network function placement for communication performance. Mobility studies of heterogeneous networks have been carried out in network mobility (NEMO), vehicular ad hoc networks (VANET), and telematics technology. Mobile gateways play an important role in communication between things/mobile users and networks. However, mobile gateways have a difficulty on maintaining the communication link in a wireless environment due to obstacles and unstable connections with devices. In this paper proposes a dynamic load balancing approach in order that the multi-network mobile gateway (MNMG) distributes its traffic to one of several heterogeneous networks. For high utilization of networks, the MNMG distributes traffic dynamically by reflecting each network state. To evaluate the performance of the proposed approach, experiments is performed on real commercial networks consisting of the MNMG with the proposed load balancing were used in Korea, Thailand and India.

  PublisherDOI

• Finding Survivable Routes in Multi-Domain Optical Networks With Geographically Correlated Failures

  Journal of Optical Communications and Networking · 2018-07-12 · 8 citations

  article

  This work presents the problem of survivable path pair routing in multi-domain optical networks with geographically correlated failures. The objective is to minimize the risk of simultaneous failure of both the primary and backup paths. We develop a probabilistic model to calculate the simultaneous failure probability of both the paths under circular geographic failures with uniform and non-uniform distributions of epicenter location. We develop topology aggregation techniques and an inter-domain minimum overlapping area routing algorithm based on the aggregated information from each domain. Our algorithm is compared to Suurballe’s Algorithm (SUR) and an approach with full information shared amongst domains, and we show that our heuristic approach is effective in reducing the total probability of simultaneous failure.

  PublisherDOI

Frequent coauthors

• Jinseok Lee

  Kyung Hee University

  25 shared

• Shu-Shin Chin

  Stony Brook University

  21 shared

• Yunyoung Nam

  Soonchunhyang University

  18 shared

• We‐Duke Cho

  Ajou University

  18 shared

• Shung Han Cho

  Samsung (South Korea)

  17 shared

• Petar M. Djurić

  Stony Brook University

  16 shared

• Miodrag Bolić

  14 shared

• Nammee Moon

  12 shared

Labs

• Electrical and Computer Engineering, Stony Brook UniversityPI

Education

• Ph.D., Electrical and Computer Engineering

  Stony Brook University

  2000

• M.S., Electrical and Computer Engineering

  Stony Brook University

  1995

• B.S., Electrical Engineering

  Seoul National University

  1991