• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 하계종합학술대회논문집
• /
• pp.46-49
• /
• 1998

In this paper, we introduce the notion of intelligent communication by introducing a novel intelligent receiver model. This receiver is continually evolving and learns and improves in performance as it compiles its experience over time. In digital communication context, in a typical training mode, it jearns the concept of "1" as is deteriorated by arbitrary (not necessarily additive as is typically assumed) disturbance and /or modulation. After learning "1", in test mode, it classifies the received signal "1" and "0" almost completely. The intelligent receiver as implemented is grounded on the recently introduced Stochastic Morphological Sampling Theorem(SMST), a distribution-free result which gives theoretical bounds on the sample complexity(training size) needed for the required performance parameters such as accuracy($\varepsilon$) and confidence($\delta$). Based on this theorem, we demonstrate --almost irrespective of channel and modulation model-- the number of samples needed to learn the concept of "1" is not too "large" and the resulting universal receiver structure, that corresponding to classical Nearest Neighbor rule in Pattern Recognition Theory, is trivial. We check the surprising efficiency and validity of this model through some simple simulations. and validity of this model through some simple simulations.

• International Journal of Internet, Broadcasting and Communication
• /
• 제12권2호
• /
• pp.120-126
• /
• 2020

Studies are being conducted regarding falling recognition using sensors on smartphonesto recognize falling in human daily life. These studies use a number of sensors, mostly acceleration sensors, gyro sensors, motion sensors, etc. Falling recognition system processes the values of sensor data by using a falling recognition algorithm and classifies behavior based on thresholds. If the threshold is ambiguous, the accuracy will be reduced. To solve this problem, Deep learning was introduced in the behavioral recognition system. Deep learning is a kind of machine learning technique that computers process and categorize input data rather than processing it by man-made algorithms. Thus, in this paper, we propose a falling recognition application system using deep learning based on smartphones. The proposed system is powered by apps on smartphones. It also consists of three layers and uses DataBase as a Service (DBaaS) to handle big data and address data heterogeneity. The proposed system uses deep learning to recognize the user's behavior, it can expect higher accuracy compared to the system in the general rule base.

• 산업기술연구
• /
• 제3권
• /
• pp.69-74
• /
• 1983

For the determination of a reservoir capacity Rippl's mass-curve method has long been used with the past river flow data assuming the same flow records will be repeated in the future. This study aims to find out a better method for determining the reservoir capacity by employing the analytical theory based on the stochastic process. For the present study the synthetic generation methods of Thomas-Fiering type was used to synthetically generate 50 years of monthly river inflows to three single-purpose reservoirs and three multi-purpose reservoirs. The generated sequences of monthly flows were analyzed based on the range concept. With the optimum operation rule of the reservoirs as the one which maximizes the water-use downstream the waterrelease from the reservoir was determined and with due consideration to the mean inflows and the range of monthly flows the required reservoirs capacity was stochastically determined. It was possible to repersent the so-determined reservoir capacity in terms of the mean monthly inflows and the number of subseries in the determination of ranges. It is suggested that the result obtained in this study would be applied to approximately estimate, in the stage of preliminary design, the required capacity of a reservoir in question with the limited information such as the mean monthly inflow and the period of reservoir operation.

• 디지털콘텐츠학회 논문지
• /
• 제9권4호
• /
• pp.785-790
• /
• 2008

디지털 캐로절은 사용자들에게 미디어 동기화 메카니즘을 통하여 미디어 객체 공유를 가능하게 한다. 본 시스템은 공동 작업에 참여한 사용자들이 다른 참여자들에게 같은 뷰로써 공유된 미디어 또는 오류 객체들을 참조할 수 있도록 구축한다. 본 논문에서는 결함 허용을 통하여 신뢰성을 향상시키는 방법에 대해서 기술한다. ER은 분산 멀티미디어에서 하나의 소프트웨어 오류를 자동적으로 복구할 수 있는 시스템이다. 본 논문은 규칙-기반 DEVS(Discrete Event System Specification) 모델링과 시뮬레이션 기법을 사용하면서 분산 멀티미디어 상에서의 오류 복구 시스템의 성능 분석을 설명한다. DEVS에서 하나의 시스템은 시간, 입력, 상태, 출력 및 함수들을 가지고 있다.

• 한국정밀공학회지
• /
• 제31권12호
• /
• pp.1127-1132
• /
• 2014

Fast tool servo (FTS) is an enabling technology to fabricate various shapes of functional surface geometries in a precise and controllable manner. FTS can be also employed as a straightforward and efficient surface treatment way of making such products more durable. In this work, process characteristics using high-precision FTS-based surface texturing were qualitatively and quantitatively investigated to provide a class of surface design rule. The morphologies of surfaces processed with different conditions were first examined by observing the resultant 2D/3D surface profiles. In addition, the effects of the surface treatment using FTS on hardness and wear properties were characterized and compared to those without treatment.

• 대한기계학회논문집A
• /
• 제40권7호
• /
• pp.629-635
• /
• 2016

이 논문은 꽃잎들, 나비나 민들레 씨앗들과 같은 생물체 시스템의 불규칙한 운동을 파티클 필터링 이론에 근거하여 예측하고 추적하는 유용한 방법을 제안한다. 생물체 모사 시스템 설계에 있어서, 생체 시스템의 운동에 대한 관측과 생체 시스템 운동학에 대한 새로운 설계원리가 어떻게 자연스럽게 운동하는가에 대한 인상을 얻는데 중요하다. 공기 중에서 비행하는 꽃잎에 대한 시스템 모델링이 베이지안 확률 규칙을 사용하여 수행되었다. 실험결과는 제안된 방법이 공기의 난류로부터 유도된 랜덤한 외란이 있는 경우에도 잘 예측함을 보여준다.

• 디지털콘텐츠학회 논문지
• /
• 제17권5호
• /
• pp.359-366
• /
• 2016

유비쿼터스 환경에서 사용자에게 유용한 서비스를 제공하기 위해서는 시간과 공간을 기반으로 사용자의 행동과 선호 패턴을 고려하여 가장 적합한 데이터를 처리할 수 있는 방법이 필요하다. 실세계에서 사용자의 관심은 시간이 지남에 따라 변화할 수 있다. 그러므로 서비스 관심도의 변화를 중요도에 반영하여 정보를 추출할 수 있는 방법이 필요하다. 이 논문에서는 사용자에게 필요한 서비스 정보를 온톨로지로 설계하고 시간에 따라 동적으로 변화하는 사용자의 서비스 이용 패턴이나 데이터의 중요도를 동적 가중치로 표현하여 빈발 패턴을 찾는 방법을 제안한다. 이 논문에서 제안하는 동적 가중치를 고려하는 빈발 서비스 패턴 마이닝 기법은 시간의 변화에 따라 필요로 하는 사용자의 관심을 서비스의 중요도로 반영하므로 실시간의 최적화된 서비스 제공이 가능하다.

• 정보처리학회논문지D
• /
• 제8D권5호
• /
• pp.623-630
• /
• 2001

최근 인터넷 솔루션 시장은 전자상거래의 필요성과 효과에 대한 논의를 벗어나 전자상거래 내에서의 경쟁에 초점이 맞추어져 있다. 본 논문에서는 전자상거래 시장을 활성화에 기여할 수 있는 정책지향 매칭 에이전트(policy-oriented matching agent) 시스템을 제안한다. 제안된 정책지향적 솔루션은 정책을 객체화하였기 때문에 다양한 프로모션을 구현할 수 있으며, 기존의 룰 기반 시스템의 장점에 정책이 실기되는 공간 개념(release post)을 추가하였기 때문에 정책의 수립, 실시, 추적 평가 등을 일관되게 처리할 수 있다. 정책지향 솔루션을 구현함에 있어 전자상거래를 위한 기초 플랫폼에 적합한 컴포넌트 기반 구조를 채택하고, 기존의 여타 시스템에 연동되어 마케팅 활동을 쉽게 지원할 수 있도록 한다. 또한 정책 담당자가 정책을 직관적으로 편집할 수 있는 인터페이스를 갖도록 한다.

• Structural Engineering and Mechanics
• /
• 제45권2호
• /
• pp.233-257
• /
• 2013

Given the yield shear of a single-story inelastic structure with simple eccentricity, the problem of strength distribution among the resisting elements is investigated, with respect to minimize its torsional response during a ground motion. Making the hypothesis that the peak accelerations, of both modes of vibration, are determined from the inelastic acceleration spectrum, and assuming further that a peak response quantity is obtained by an appropriate combination rule (square root of sum of squares-SRSS or complete quadratic combination-CQC), the first aim of this study is to present an interaction relationship between the yield shear and the maximum torque that may be developed in such systems. It is shown that this torque may be developed, with equal probability, in both directions (clockwise and anticlockwise), but as it is not concurrent with the yield shear, a rational design should be based on a combination of the yield shear with a fraction of the peak torque. The second aim is to examine the response of such model structures under characteristic ground motions. These models provide a rather small peak rotation and code provisions that are based on such principles (NBCC-1995, UBC-1994, EAK-2000, NZS-1992) are superiors to EC8 (1993) and to systems with a stiffness proportional strength distribution.

• 한국조명전기설비학회:학술대회논문집
• /
• 한국조명전기설비학회 2005년도 학술대회 논문집
• /
• pp.315-320
• /
• 2005

This paper is proposed adaptive fuzzy-neuro controller for high performance of induction motor drive. The design of this algorithm based on fuzzy-neural network controller that is implemented using fuzzy control and neural network. This controller uses fuzzy rule as training patterns of a neural network. Also, this controller uses the back-propagation method to adjust the weights between the neurons of neural network in order to minimize the error between the command output and actual output. A model reference adaptive scheme is proposed in which the adaptation mechanism is executed by fuzzy logic based on the error and change of nor measured between the motor speed and output of a reference model. The control performance of the adaptive fuzy-neuro controller is evaluated by analysis for various operating conditions. The results of experiment prove that the proposed control system has strong high performance and robustness to parameter variation, and steady-state accuracy and transient response.