• Journal of KIISE:Computer Systems and Theory
• /
• v.34 no.9
• /
• pp.459-466
• /
• 2007

TCP/IP processing imposes a heavy load on the host CPU when it is processed by the host CPU on a very high-speed network. Recently the TCP/IP Offload Engine (TOE), which processes TCP/IP on a network adapter instead of the host CPU, has become an attractive solution to reduce the load in the host CPU. There have been two approaches to implement TOE. One is the software TOE in which TCP/IP is processed by an embedded processor and the other is the hardware TOE in which TCP/IP is processed by a dedicated ASIC. The software TOE has poor performance and the hardware TOE is neither flexible nor expandable enough to add new features. In this paper we designed and implemented a hybrid TOE architecture, in which TCP/IP is processed by cooperation of hardware and software, based on an FPGA that has two embedded processor cores. The hybrid TOE can have high performance by processing time-critical operations such as making and processing data packets in hardware. The software based on the embedded Linux performs operations that are not time-critical such as connection establishment, flow control and congestions, thus the hybrid TOE can have enough flexibility and expandability. To improve the performance of the hybrid TOE, we developed a hardware-based transmission/reception accelerator that processes important operations such as creating data packets. In the experiments the hybrid TOE shows the minimum latency of about $19{\mu}s$. The CPU utilization of the hybrid TOE is below 6 % and the maximum bandwidth of the hybrid TOE is about 675 Mbps.

• The Journal of Society for e-Business Studies
• /
• v.22 no.1
• /
• pp.107-122
• /
• 2017

Nowadays, digital forensic experts are not only computer experts who restore and find deleted files, but also general experts who posses various capabilities including knowledge about processes/laws, communication skills, and ethics. However, there have been few studies about qualifications or competencies required for digital forensic experts comparing with their importance. Therefore, in this study, AHP questionnaires were distributed to digital forensic experts and analyzed to derive priorities of competencies; the first-tier questions which consisted of knowledge, technology, and attitude, and the second-tier ones which have 20 items. Research findings showed that the most important competency was knowledge, followed by technology and attitude but no significant difference was found. Among 20 items of the second-tier competencies, the most important competency was "digital forensics equipment/tool program utilization skill" and it was followed by "data extraction and imaging skill from storage devices." Attitude such as "judgment," "morality," "communication skill," "concentration" were subsequently followed. The least critical one was "substantial law related to actual cases." Previous studies on training/education for digital forensics experts focused on law, IT knowledge, and usage of analytic tools while attitude-related competencies have not given proper attention. We hope this study can provide helpful implications to design curriculum and qualifying exam to foster digital forensic experts.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.12
• /
• pp.427-436
• /
• 2016

Due to disasters such as earthquakes and landslides in urban areas, persons have been buried inside collapsed buildings and structures. Rescuers have mainly utilized detection equipment by applying sound, video and electric waves, but these are expensive and due to the directional approaches onto the collapsed site, secondary collapse risk can arise. In addition, due to poor utilization of such equipment, new human detection technology with quick and high reliability has not been utilized. To address these issues, this study develops a wireless signal-based human detection module that can be loaded into an Unmanned Aerial Vehicle (UAV). The human detection module searches for the 3D location for buried persons by collecting Wi-Fi signal and barometer sensors data transmitted from the mobile phones. This module can gain diverse information from mobile phones for buried persons in real time. We present a development framework of the module that provides 3D location data with more reliable information by delivering the collected data into a local computer in the ground. This study verified the application feasibility of the developed module in a real collapsed area. Therefore, it is expected that these results can be used as a core technology for the quick detection of buried persons' location and for relieving them after disasters that induce building collapses.

• Journal of Cadastre & Land InformatiX
• /
• v.50 no.1
• /
• pp.215-229
• /
• 2020

In response to the recent 4th industrial revolution, the demand for 3D object models in the latest fields of digital twin, autonomous driving, and VR/AR, as well as the existing fields such as city, construction, transportation, and energy has increased significantly. It is expected that the demand for 3D object models with various precision from LOD1 to LOD4 will increase more and more in various industry fields. However, the Ministry of Land, Infrastructure and Transport, and the local government and the private sector have partially built 3D object models of different precisions for some specific regions because of the huge cost. Therefore, this study proposes a feasible plan that can solve the cost problem in generating 3D object models for the whole territory. For our purpose, we first analyzed usage, demand, generation technology and generation cost for 3D object models. Afterwards, we proposed LOD3 model generation plan for all territory using automatic 3D object model generation technology based on image matching. Additionally, we supplemented the proposed plan by using LOD4 generation plan for landmarks and LOD2 generation plan non-urban area. In the near future, we expect this would be a great help in establishing a feasible and effective 3D object model generation plan for the whole country.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.141-148
• /
• 2022

Recently, reinforcement learning combined with deep neural network technology has achieved remarkable success in various fields such as board games such as Go and chess, computer games such as Atari and StartCraft, and robot object manipulation tasks. However, such deep reinforcement learning describes states, actions, and policies in vector representation. Therefore, the existing deep reinforcement learning has some limitations in generality and interpretability of the learned policy, and it is difficult to effectively incorporate domain knowledge into policy learning. On the other hand, dNL-RRL, a new relational reinforcement learning framework proposed to solve these problems, uses a kind of vector representation for sensor input data and lower-level motion control as in the existing deep reinforcement learning. However, for states, actions, and learned policies, It uses a relational representation with logic predicates and rules. In this paper, we present dNL-RRL-based policy learning for transportation mobile robots in a manufacturing environment. In particular, this study proposes a effective method to utilize the prior domain knowledge of human experts to improve the efficiency of relational reinforcement learning. Through various experiments, we demonstrate the performance improvement of the relational reinforcement learning by using domain knowledge as proposed in this paper.

• KIPS Transactions on Computer and Communication Systems
• /
• v.7 no.2
• /
• pp.33-40
• /
• 2018

In an environment with a large network bandwidth, maximizing bandwidth utilization is an important issue to increase transmission efficiency. End-to-end transfer efficiency is significantly influenced by factors such as network, data transfer nodes, and intranet network security policies. Science DMZ is an innovative network architecture that maximizes transfer performance through optimal solution of these complex components. Among these, the data transfer node is a key factor that greatly affects the transfer performance depending on storage, network interface, operating system, and transfer application tool. However, tuning parameters constituting a data transfer node must be performed to provide high transfer efficiency. In this paper, we propose a method to enhance performance through tuning parameters of 100Gbps data transfer node. With experiment result, we confirmed that the transmission efficiency can be improved greatly in 100Gbps network environment through the tuning of Jumbo frame and CPU governor. The network performance test through Iperf showed improvement of 300% compared to the default state and NVMe SSD showed 140% performance improvement compared to hard disk.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.9 no.1
• /
• pp.93-99
• /
• 2008

In this paper, we propose new modulation schemes using the conventional CACB modulation with constant amplitude property. Also the proposed modulation schemes supports high transmission data rate by increasing the spectral efficiency. In order to obtain the high spectral efficiency, the $Q^2$PSK and CA-$Q^2$PSK are used. We explain the simplest combining modulation scheme of CACB and $Q^2$PSK (i.e., CACB-$Q^2$PSK). However, this modulation scheme cannot support the constant amplitude property. Hence the first CACB-CA-$Q^2$PSK (or CACB-CA-$Q^2$PSK I) modulation scheme is proposed for the constant amplitude property. In the modulation scheme, the redundant constant amplitude encoding (spectral efficiency decrease) is required. Therefore, the second CACB-CA-$Q^2$PSK (or CACB-CA-$Q^2$PSK II) modulation scheme is proposed retaining the constant amplitude and the spectral efficiency. Computer simulations show that the proposed CACB-CA-$Q^2$PSK II is the efficient modulation scheme.

• Journal of Chest Surgery
• /
• v.35 no.3
• /
• pp.209-216
• /
• 2002

Background: Valvular heart disease is still the most common health problem in Korea. By the end of the year 1999, there has been 94,586 cases of open heart surgery since the first case in 1958. Among them, 36,247 cases were acquired heart diseases and 20,704 of those had valvular heart disease. But there was no database system and every surgeon and physician had great difficulties in analysing and utilizing those tremendous medical resources. Therefore, we developed a valve registry database program and utilize it for risk factor analysis and so on. Material and Method: Personal computer-based multiuser database program was created using Microsoft AccessTM. That consisted of relational database structure with fine-tuned compact field variables and server-client architecture. Simple graphic user interface showed easy-to-use accessability and comprehensibility. User-oriented modular structure enabled easier modification through native AccessTM functions. Infinite application of query function aided users to extract, summarize, analyse and report the study result promptly. Result: About three-thousand cases of valve replacement procedure were performed in our hospital from 1968 to 1999. Total number of prosthesis replaced was 3,700. The numbers of cases for mitral, aortic and tricuspid valve replacement were 1600, 584, 76, respectively. Among them, 700 patients received prosthesis in more than two positions. Bioprosthesis or mechanical prosthesis were used in 1,280 and 1,500 patients respectively Redo valve replacements were performed in 460 patients totally and 40 patients annually Conclusion: Database program for registry of valvular heart disease was successfully developed and used in personal computer-based multiuser environment. This revealed promising results and perspectives in database management and utilization system.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.89-106
• /
• 2022

Recently, the amount of video data collected from smartphones, CCTVs, black boxes, and high-definition cameras has increased rapidly. According to the increasing video data, the requirements for analysis and utilization are increasing. Due to the lack of skilled manpower to analyze videos in many industries, machine learning and artificial intelligence are actively used to assist manpower. In this situation, the demand for various computer vision technologies such as object detection and tracking, action detection, emotion detection, and Re-ID also increased rapidly. However, the object detection and tracking technology has many difficulties that degrade performance, such as re-appearance after the object's departure from the video recording location, and occlusion. Accordingly, action and emotion detection models based on object detection and tracking models also have difficulties in extracting data for each object. In addition, deep learning architectures consist of various models suffer from performance degradation due to bottlenects and lack of optimization. In this study, we propose an video analysis system consists of YOLOv5 based DeepSORT object tracking model, SlowFast based action recognition model, Torchreid based Re-ID model, and AWS Rekognition which is emotion recognition service. Proposed model uses single-linkage hierarchical clustering based Re-ID and some processing method which maximize hardware throughput. It has higher accuracy than the performance of the re-identification model using simple metrics, near real-time processing performance, and prevents tracking failure due to object departure and re-emergence, occlusion, etc. By continuously linking the action and facial emotion detection results of each object to the same object, it is possible to efficiently analyze videos. The re-identification model extracts a feature vector from the bounding box of object image detected by the object tracking model for each frame, and applies the single-linkage hierarchical clustering from the past frame using the extracted feature vectors to identify the same object that failed to track. Through the above process, it is possible to re-track the same object that has failed to tracking in the case of re-appearance or occlusion after leaving the video location. As a result, action and facial emotion detection results of the newly recognized object due to the tracking fails can be linked to those of the object that appeared in the past. On the other hand, as a way to improve processing performance, we introduce Bounding Box Queue by Object and Feature Queue method that can reduce RAM memory requirements while maximizing GPU memory throughput. Also we introduce the IoF(Intersection over Face) algorithm that allows facial emotion recognized through AWS Rekognition to be linked with object tracking information. The academic significance of this study is that the two-stage re-identification model can have real-time performance even in a high-cost environment that performs action and facial emotion detection according to processing techniques without reducing the accuracy by using simple metrics to achieve real-time performance. The practical implication of this study is that in various industrial fields that require action and facial emotion detection but have many difficulties due to the fails in object tracking can analyze videos effectively through proposed model. Proposed model which has high accuracy of retrace and processing performance can be used in various fields such as intelligent monitoring, observation services and behavioral or psychological analysis services where the integration of tracking information and extracted metadata creates greate industrial and business value. In the future, in order to measure the object tracking performance more precisely, there is a need to conduct an experiment using the MOT Challenge dataset, which is data used by many international conferences. We will investigate the problem that the IoF algorithm cannot solve to develop an additional complementary algorithm. In addition, we plan to conduct additional research to apply this model to various fields' dataset related to intelligent video analysis.

• Journal of Internet Computing and Services
• /
• v.16 no.2
• /
• pp.109-115
• /
• 2015

The new medical device technologies for bio-signal information and medical information which developed in various forms have been increasing. Information gathering techniques and the increasing of the bio-signal information device are being used as the main information of the medical service in everyday life. Hence, there is increasing in utilization of the various bio-signals, but it has a problem that does not account for security reasons. Furthermore, the medical image information and bio-signal of the patient in medical field is generated by the individual device, that make the situation cannot be managed and integrated. In order to solve that problem, in this paper we integrated the QR code signal associated with the medial image information including the finding of the doctor and the bio-signal information. bio-signal. System implementation environment for medical imaging devices and bio-signal acquisition was configured through bio-signal measurement, smart device and PC. For the ROI extraction of bio-signal and the receiving of image information that transfer from the medical equipment or bio-signal measurement, .NET Framework was used to operate the QR server module on Window Server 2008 operating system. The main function of the QR server module is to parse the DICOM file generated from the medical imaging device and extract the identified ROI information to store and manage in the database. Additionally, EMR, patient health information such as OCS, extracted ROI information needed for basic information and emergency situation is managed by QR code. QR code and ROI management and the bio-signal information file also store and manage depending on the size of receiving the bio-singnal information case with a PID (patient identification) to be used by the bio-signal device. If the receiving of information is not less than the maximum size to be converted into a QR code, the QR code and the URL information can access the bio-signal information through the server. Likewise, .Net Framework is installed to provide the information in the form of the QR code, so the client can check and find the relevant information through PC and android-based smart device. Finally, the existing medical imaging information, bio-signal information and the health information of the patient are integrated over the result of executing the application service in order to provide a medical information service which is suitable in medical field.