• 한국전자통신학회논문지
• /
• 제15권4호
• /
• pp.725-732
• /
• 2020

본 논문은 소외계층 아동의 운동학습프로그램에서 체력 활동 중 나를 잘 따른다(0-9), 마음의 결정을 내리는데 많은 시간이 걸린다(0-9), 맥빠진(0-9) 등을 변수로 사용하여 '성별', '체육교실', 나이의 '상중하'를 분류하고 스포츠 재활치료를 통한 자아 탄력(ego-resiliency)과 자아 통제(self-control)의 변화를 관찰하여 정신 건강 변화를 알아본다. 이를 위해 취득한 데이터를 병합하고 Label encoder와 One-hot encoding을 사용하여 숫자의 크고 작음의 특성을 제거한 후 MLP, SVM, Dicesion tree, RNN, LSTM의 각각의 알고리즘을 적용하여 성능을 평가하기 위해 Train, Test 데이터를 75%, 25% 스플릿 한 뒤 Train 데이터로 알고리즘을 학습하고 Test 데이터로 알고리즘의 정확성을 측정한다. 측정 결과 성별에서는 LSTM, 체육 교실은 MLP와 LSTM, 나이는 SVM이 가장 우수한 결과를 보임을 확인하였다.

• 한국통신학회논문지
• /
• 제28권4B호
• /
• pp.298-305
• /
• 2003

본 논문에서는 ATM 기반 레이블 에지 라우터의 IP 패킷 포워딩 엔진을 위한 고성능 룩업 제어기를 제안하였다. 제안한 룩업 제어기는 IP 패킷에 대한 Best Effort 서비스 뿐만 아니라 MPLS(Multiprotocol Label Switching), VPN(Virtual Private Network), ELL(Emulated Leased Line), RT(Real Time) 서비스 등의 차별화된 서비스들을 제공할 수 있도록 설계하였다. 고속의 IP 주소 검색을 위하여 소프트웨어 기반의 알고리즘 방식을 사용하지 않고 TCAM(Ternary Content Addressable Memory) 기반의 하드웨어 방식을 적용하였으며, IP 패킷 헤더 처리 및 룩업 제어 기능을 고속으로 수행하기 위하여 룩업 제어 기능을 FPGA(Field Programmable Gate Array)로 구현하였다. 룩업 제어기는 사용자의 차별화된 다양한 요구를 수용할 수 있으며, 성능 향상을 위하여 파이프라인 기법으로 처리되도록 설계하였다. 또한 패킷 헤더의 여러 영역을 조합한 비교키에 대한 룩업 기능을 수행하기 위하여 2단계 검색 메카니즘을 가지며, 시뮬레이션을 통하여 제안한 룩업 제어기는 약 16Mpps의 성능을 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권2호
• /
• pp.311-326
• /
• 2024

The rapid development of neural network technology promotes the neural network model driven by big data to overcome the texture effect of complex objects. Due to the limitations in complex scenes, it is necessary to establish custom template matching and apply it to the research of many fields of computational vision technology. The dependence on high-quality small label sample database data is not very strong, and the machine learning system of deep feature connection to complete the task of texture effect inference and speculation is relatively poor. The style transfer algorithm based on neural network collects and preserves the data of patterns, extracts and modernizes their features. Through the algorithm model, it is easier to present the texture color of patterns and display them digitally. In this paper, according to the texture effect reasoning of custom template matching, the 3D visualization of the target is transformed into a 3D model. The high similarity between the scene to be inferred and the user-defined template is calculated by the user-defined template of the multi-dimensional external feature label. The convolutional neural network is adopted to optimize the external area of the object to improve the sampling quality and computational performance of the sample pyramid structure. The results indicate that the proposed algorithm can accurately capture the significant target, achieve more ablation noise, and improve the visualization results. The proposed deep convolutional neural network optimization algorithm has good rapidity, data accuracy and robustness. The proposed algorithm can adapt to the calculation of more task scenes, display the redundant vision-related information of image conversion, enhance the powerful computing power, and further improve the computational efficiency and accuracy of convolutional networks, which has a high research significance for the study of image information conversion.

• Clinics in Shoulder and Elbow
• /
• 제27권2호
• /
• pp.237-246
• /
• 2024

Proximal humeral fractures (PHFs) are a common injury among the older population. An ideal therapeutic protocol has yet to be developed, and numerous clinical trials are being conducted to find the best therapeutic approach. The purpose of this study is to evaluate the current body of knowledge available via interventional clinical trials. In December 2022, interventional clinical trials relating to PHFs on Clinicaltrials.gov were screened. Trial characteristics included duration, status, intervention, phase, outcomes, location, and study design. Publications associated with each trial were searched on PubMed/Medline using the ClinicalTrials.gov registry number. The final dataset comprised 64 trials. The most common trial status was completed (36%). The majority did not have a Food and Drug Administration-defined phase (67%), was randomized (81%), involved a single facility (72%), used a parallel assignment intervention model (80%), and used an open-label approach (45%). Eleven trials were associated with a publication, and the publication rate was 17%. Average enrollment was 86 participants, and mean trial duration was 51.4 months. Europe/UK/Russia/Turkey participated in the most trials (70%). Most of the trials were initiated after 2010 (87.5%). Procedure-related interventions (55%) were most common. Disability/function was the most common primary outcome assessed (61%). The low publication rate and the multitude of trials conducted after 2010 highlight the urgency and need for trial results to be published to establish an ideal therapeutic protocol. Since the majority of the trials involved a single institution and an open-label approach, reinforcing blinding and establishing multi-centered trials can improve the validity of the clinical trial results.

• ETRI Journal
• /
• 제24권1호
• /
• pp.43-55
• /
• 2002

This paper proposes a flow-based admission control algorithm through an Asynchronous Transfer Mode (ATM) based Multi-Protocol Label Switching (MPLS) network for multiple service class environments of Integrated Service (IntServ) and Differentiated Service (DiffServ). We propose the Integrated Packet Scheduler to accommodate IntServ and Best Effort traffic through the DiffServ-aware MPLS core network. The numerical results of the proposed algorithm achieve reliable delay-bounded Quality of Service (QoS) performance and reduce the blocking probability of high priority service in the DiffServ model. We show the performance behaviors of IntServ traffic negotiated by end users when their packets are delivered through the DiffServ-aware MPLS core network. We also show that ATM shortcut connections are well tuned with guaranteed QoS service. We validate the proposed method by numerical analysis of its performance in such areas as throughput, end-to-end delay and path utilization.

• 한국통신학회논문지
• /
• 제28권2B호
• /
• pp.144-156
• /
• 2003

Internet/Intranet has been continuously enhanced by new emerging IP technologies such as differentiate service(DiffServ), IPSec(IP Security) and MPLS(Multi-protocol Label Switching) traffic engineering. According to the increased demands of various real-time multimedia services, ISP(Internet Service Provider) should provide enhanced end-to-end QoS(quality of service) and security features. Therefore, Internet and Intranet need the management functionality of sophisticated traffic engineering functions. In this paper, we design and implement the performance management functionality for the guaranteed end-to-end QoS provisioning on MPLS-based VPLS(Virtual Private LAN Service). We propose VPLS OAM(Operation, Administration and Maintenance) for efficient performance management. We focus on a scheme of QoS management and measurement of QoS parameters(such as delay, jitter, loss, etc.) using VPLS OAM functions. The proposed performance management system also supports performance tuning to enhance the provided QoS by re-adjusting the bandwidth of LSPs for VPLS. We present the experimental results of performance monitoring and analysis using a network simulator.

• 제어로봇시스템학회논문지
• /
• 제6권8호
• /
• pp.666-671
• /
• 2000

With the current advances in optical WDM (Wavelength Division Multiplexing) networking technologies and the increasing demand for network bandwidth the Next Generation Internet is expected to be a network that runs IP(Internet Protocol) directly over WDM-based optical networks. The network control architecture for the IP over WDM networks is different from that of traditional Internet since the underlying WDM devices have more constraints than electronic IP routers such as the lack of optical buffers and wavelength continuity property etc. In this paper we introduce several architectural models for implementing IP over WDM networks and propose an integrated network control framework for the IP over WDM networks. This framework leverages the traffic engineering control architecture for the MPLS (Multi-Protocol Label Switching) networks and is mainly developed for the IP over packet-switched WDM networks. We also report several preliminary simulation results of contention resolution schemes in the packet-switched WDM networks.

• 한국정보처리학회논문지
• /
• 제6권11S호
• /
• pp.3190-3198
• /
• 1999

Because of the rapid growth of the mobile communication, the need for the mobile internet access has grown up as well. since the current mobile communication network, however, is optimized for a voice communication system, which exclusively occupies a channel for a given time, it is not suitable for variable rate packet data. In order to support the mobile internet access, it is essential do design a reasonable packet switching network which supports the mobility. Since mobile packet network has longer latency, high speed switching and QoS are required to meet the user's requirements. In this paper, we suggest an resonable way to construct a network and its operation procedures utilizing GPRS(General Packet Radio Service) network and MPLS(Multi Protocol Label Switching) to provide a high speed switching and QoS mobile internet access. GPRS is used as a network which supports the mobility and MPLS guarantees the QoS and high speed IP protocol transmission based on the ATM switching technology.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2017년도 춘계학술발표대회
• /
• pp.181-182
• /
• 2017

관성 센서 기반 행위인지는 스마트폰과 웨어러블 밴드 등의 출현으로 보다 간편한 방법으로 행위인지가 가능해졌다. 현재 대부분의 행위인지 서비스나 연구들은 단일 행위의 결론만을 도출하고 있으나, 이러한 방식은 한 행위에서 한 가지 동작밖에 취할 수 없는 경우에는 문제가 없지만 두 가지 이상의 동작이 합쳐진 경우에 어떤 행위를 최종 결론으로 도출해야 하는지에 대한 문제점을 내포한다. 따라서 본 논문에서는 이러한 문제점을 해결하기 위해 세 개의 센서 기기 (스마트폰, 스마트워치, 웨어러블 센서)를 이용한 멀티 레이블 행위인지를 제안한다. 스마트폰은 신체 전반적인 움직임 탐지를 위하여 소지위치가 정해지지 않은 비고정식 센서의 보조적인 역할을 수행한다. 스마트워치는 사용자가 주로 사용하는 손의 손목, 그리고 웨어러블 센서는 사용자의 허벅지에 부착되어 각각 상하체의 움직임을 파악한다. 이후 각 기기에서 도출된 결론에 Majority Weighted Voting 기법을 적용하여 단일 혹은 멀티 레이블의 최종 행위를 도출한다.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2003년도 가을 학술발표논문집 Vol.30 No.2 (3)
• /
• pp.4-6
• /
• 2003

IP망을 활용한 VPN(Virtual Private Network)에서의 QoS 보장 메커니즘과 광대역폭 제공에 대한 해결책으로 차세대 광 인터넷을 통한 OVPN(Optical VPN) 기술이 제시되고 있다. 차세대 광 인터넷의 구현이 IP/GMPLS(Generalized Multi­Protocol Label Switching) over DWDM(Dense Wavelength Division Multiplexing) 프로토콜 프레임워크로 표준화되고 있는 현실에 비추어, IP/GMPLS over DWDM 백본망을 통한 OVPN은 차세대 가상사설망으로써 멀티미디어 서비스 제공을 위한 최적의 방안이다. 이러한 차세대 멀티미디어 서비스 제공을 위한 OVPN에서는 서비스 클래스별로 차등화된 광 QoS 제공 및 보장 기술이 절실히 요구되고 있으므로, 본 논문에서는 차등화된 광 QoS 서비스(DOQS:Differetiated Optical QoS Service)를 제공하는 OVPN모델을 제시하고 제안된 DOQS­OVPN 모델의 QoS 지원 광경로상에서의 QoS Failure에 대한 서비스별 차등화된 Recovery 능력을 갖는 QoS 보장 프로토콜 프레임워크를 제안한다.