• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.14 no.2
• /
• pp.96-103
• /
• 2005

The early detection of the bearing defects in rotating machinery is very important since the critical failure of bearing causes a machinery shutdown. However it is not easy to detect the vibration signal caused by the initial defects of bearing because of the high level of random noise. A signal processing technique, called the adaptive line enhancer(ALE) as one of adaptive filter, is used in this study. This technique is to eliminate random noise with little a prior knowledge of the noise and signal characteristics. Also we propose the optimal methods fir selecting the three main ALE parameters such as correlation length filter order and adaptation constant. Vibration signals f3r three abnormal bearings, including inner and outer raceways and ball defects, were acquired by Anderon(angular derivative of radius on) meter. The experimental results showed that ALE is very useful f3r detecting the bearing defective signals masked by random noise.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.660-661
• /
• 2020

본 논문에서는 기존 인공 신경망 기반 시계열 학습 기법인 Recurrent Neural Network (RNN)의 많은 연산량 및 고 사양 시스템 요구를 개선하기 위해 랜덤 포레스트 (Random Forest)기반의 새로운 시계열 학습 기법을 제안한다. 기존의 RNN 기반 방법들은 복잡한 연산을 통해 높은 성능을 달성하는 데 집중하고 있다. 이러한 방법들은 학습에 많은 파라미터가 필요할 뿐만 아니라 대규모의 연산을 요구하므로 실시간 시스템에 적용하는데 어려움이 있다. 따라서 본 논문에서는, 효율적이면서 빠르게 동작할 수 있는 시계열 다층 랜덤 포레스트(Time series Multilayered Random Forest)를 제안하고 산불 조기 탐지에 적용해 기존 RNN 계열의 방법들과 성능을 비교하였다. 다양한 산불화재 실험데이터에 알고리즘을 적용해본 결과 GPU 상에서 방대한 연산을 수행하는 RNN 기반 방법들과 비교해 성능적인 한계가 존재했지만 CPU 에서도 빠르게 동작 가능하므로 성능의 개선을 통해 다양한 임베디드 시스템에 적용 가능하다.

• Journal of Sensor Science and Technology
• /
• v.32 no.6
• /
• pp.378-385
• /
• 2023

Human biosignals provide essential information for diagnosing diseases such as dementia and Parkinson's disease. Owing to the shortcomings of current clinical assessments, noninvasive solutions are required. Machine learning (ML) on wearable sensor data is a promising method for the real-time monitoring and early detection of abnormalities. ML facilitates disease identification, severity measurement, and remote rehabilitation by providing continuous feedback. In the context of wearable sensor technology, ML involves training on observed data for tasks such as classification and regression with applications in clinical metrics. Although supervised ML presents challenges in clinical settings, unsupervised learning, which focuses on tasks such as cluster identification and anomaly detection, has emerged as a useful alternative. This review examines and discusses a variety of ML algorithms such as Support Vector Machines (SVM), Random Forests (RF), Decision Trees (DT), Neural Networks (NN), and Deep Learning for the analysis of complex clinical data.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04a
• /
• pp.280-282
• /
• 2002

기존의 네트워크에서는 혼잡상황이 감지된 이후에 네트워크 성능이 급격하게 저하된다. 이러한 문제를 해결하고자 RED(Random Early Detection)기법이 소개되어 게이트웨이에서 혼잡상황에 대하여 능동적으로 대처할 수 있는 알고리즘이 제시되었다. 하지만, RED는 매개변수 설정이라는 문제가 남아있다. 그리하여, 잘못된 변수값 설정으로 인한 네트워크 성능 저하가 현저하게 발생한다. 본 논문에서는 기존의 RED를 개선한 Effective RED를 제안한다. Effective RED는 RED 알고리즘의 문제점을 개선하여 네트워크의 상황에 맞추어 동적으로 매개 변수 값을 조정하는 알고리즘이다. 그리고, ns를 이용하여 Effective RED의 성능을 검증하였다.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2003.05a
• /
• pp.417-424
• /
• 2003

In this paper, we introduce and analyze a feedback control model of TCP/AQM dynamics. Then, we propose the Pro-active Queue Management (PAQM) mechanism, which can provide proactive congestion avoidance and control using an adaptive congestion indicator and a control function for wide range of traffic environments. The PAQM stabilizes the queue length around a desired level while giving smooth and low packet loss rates independent of the traffic load level under a wide range of traffic environment. The PAQM outperforms other AQM algorithms such as Random Early Detection (RED) [1] and PI-controller [2]

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10c
• /
• pp.427-429
• /
• 2001

최근 라우터에서는 정체를 회피하고 전송률을 향상시키기 위한 능동적 큐 관리와 패킷 스케줄링에 대한 많은 논의가 이루어지고 있다. 본 논문은 라우터에서의 전송률 향상을 위한 Random Early Detection (RED) 알고리즘과 최근가지 변형된 RED알고리즘들의 특징을 살펴보고, RED라우터에 적용하여 실제로 종단 호스트(End-to-end)에서 전송 받는 패킷의 양을 창상하기 위한 선별적 처리 알고리즘을 제안한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10c
• /
• pp.490-492
• /
• 2000

기존의 인터넷 라우터는 Drop tail 방식으로 패킷을 관리한다. 따라서 네트워크 트래픽의 지수적인 증가로 인한 혼잡 상황으로 발생하는 패킷 손실을 해결할 수 없다. 이 문제를 해결하기 위해 IETF(Internet Engineering Task Force)에서는 RED(Random Early Detection)와 같은 능동적인 큐 관리 알고리즘을 제시하였다. 하지만 RED는 동적으로 변화하는 인터넷 트래픽에 대하여 단지 큐 크기의 변화 정보를 얻어 혼잡 상황을 제어하기 때문에 성능에 있어는 매우 비효율적이다. 본 논문에서는 기존의 RED를 개선한 MRED를 제안했다. MRED는 RED에 비하여 휴리스틱한 방법을 이용하여 폐기 확률 값을 계산하고, 이를 실험을 통하여 MRED의 성능을 검증하였다.

• The KIPS Transactions:PartC
• /
• v.10C no.1
• /
• pp.39-44
• /
• 2003

The network performance gets down during congestion periods to solve the problem effectively. A RED(Random Earl Detection) algorithm of the queue management algorithm is proposed and IETF recommends it as a queue management. A RED algorithm controls a congestion aspect dynamically. In analyzing parameters when static value of parameter is set in the gateway cannot be handled the status of current network traffic properly We propose the Effective RED algorithm to solve with the weakness of RED In this algorithm the maximum drop probability decides to accept or drop the interning packets, is adjusted dynamically on the current queue state for controlling the congestion phase effectively in the gateway. This algorithm is confirmed by computer simulation using the NS(Network Simulator)-2.

• Journal of Advanced Navigation Technology
• /
• v.14 no.3
• /
• pp.358-366
• /
• 2010

In the wireless network the congestion and delay occurs mainly when there are too many packets for the network to process or the sender transmits more packets than the receiver can accept. The congestion and delay is the reason of packet loss which degrades the performance of multimedia streaming. This paper proposes a novel transmission rate monitoring-based optimization mechanism to optimize packet loss and to improve QoS. The proposed scheme is based on the trade-off relationship between transmission rate monitoring and overhead monitoring. For this purpose this paper processes a source rate control-based optimization which optimizes congestion and delay. Performance evaluated RED, TFRC, and the proposed mechanism. The simulation results show that the proposed mechanism is more efficient than REC(Random Early Detection) mechanism and TFRC(TCP-friendly Rate Control) mechanism in packet loss rate, throughput rate, and average response rate.

• Journal of Advanced Navigation Technology
• /
• v.18 no.5
• /
• pp.461-467
• /
• 2014

Active queue management (AQM) algorithms run on routers and detect incipient congestion by typically monitoring the instantaneous or average queue size. When the average queue size exceeds a certain threshold, AQM algorithms infer congestion on the link and notify the end systems to back off by proactively dropping some of the packets arriving at a router or marking the packets to reduce transmission rate at the sender. Among the existing AQM algorithms, random early detection (RED) is well known as the representative queue-based management scheme by randomizing packet dropping. To reduce the number of timeouts in TCP and queuing delay, maintain high link utilization, and remove bursty traffic biases, the RED considers an average queue size as a degree of congestions. However, RED do not well in the specified networks conditions due to the fixed parameters($P_{max}$ and $TH_{min}$) of RED. This paper addresses a extended RED to be adapted in various networks conditions. By sensing network state, $P_{max}$ and $TH_{min}$ can be automatically changed to proper value and then RED do well in various networks conditions.