• 한국통신학회논문지
• /
• 제34권4C호
• /
• pp.403-413
• /
• 2009

본 논문에서는 OFDM 신호의 높은 PAPR과 전력 증폭기의 비선형성에 의한 신호의 왜곡과 스펙트럼의 확산을 방지하기 위한 전치 보상기의 설계 기법으로 디지털 영역에서 구현 가능한 p차 역필터를 이용한 방법, 간접 학습 구조를 이용한 방법 그리고 룩업 테이블을 이용한 방법 등 3가지 방식을 설명하고 각각의 성능을 비교 분석하였다. 앞의 두 방법은 다항식을 이용한 방법으로, 계수의 개수가 적어 많은 메모리가 필요 없고 수렴 속도가 빠르고, 진폭과 위상의 보상을 나누어서 구성하므로 복소 계산이 필요 없어 계산도 간단하다. 룩업 테이블 방법은 연산 과정이 간단하기 때문에 구현이 가장 쉬운 장점을 가지지만 위의 두 방식에 비해 많은 메모리를 필요로 하는 단점을 가진다. 모의 실험 결과 간접 학습 구조가 가장 좋은 성능을 가지지만 64QAM 변조 방식을 기준으로 $BER=10^{-4}$에서 최대 SNR 1 dB 정도의 차이를 가지므로 거의 같은 성능을 가진다고 볼 수 있다. 위의 세가지 전치보상기는 증폭기의 에이징(aging)과 환경 변화에 적응적으로 동작하며 구현 상의 요구에 따라 선택될 수 있다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1993년도 한국자동제어학술회의논문집(국내학술편); Seoul National University, Seoul; 20-22 Oct. 1993
• /
• pp.14-19
• /
• 1993

In this paper, we proposed an indirect learning and direct adaptive control schemes using neural networks, i.e., composite adaptive neural control, for a class of continuous nonlinear systems. With the indirect learning method, the neural network learns the nonlinear basis of the system inverse dynamics by a modified backpropagation learning rule. The basis spans the local vector space of inverse dynamics with the direct adaptation method when the indirect learning result is within a prescribed error tolerance, as such this method is closely related to the adaptive control methods. Also hash addressing technique, similar to the CMAC functional architecture, is introduced for partitioning network hidden nodes according to the system states, so global neuro control properties can be organized by the local ones. For uniform stability, the sliding mode control is introduced when the neural network has not sufficiently learned the system dynamics. With proper assumptions on the controlled system, global stability and tracking error convergence proof can be given. The performance of the proposed control scheme is demonstrated with the simulation results of a nonlinear system.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 추계학술대회 논문집 학회본부
• /
• pp.136-138
• /
• 1997

This paper proposes a design method to improve the performance of Indirect Adaptive Controllers using a CP net. This hybrid control architecture consists of Indirect Adaptive Controllers and CP net Controller. The performance of a single Adaptive Controller, multi Adaptive Controllers and the proposed model is compared by control problems. The simulation results show that the proposed model is superior to the others in most cases, in regard of not only learning speed but also control problems.

• 대한전자공학회논문지TC
• /
• 제49권5호
• /
• pp.27-36
• /
• 2012

본 논문에서는 스마트 안테나 시스템에 대해 경로 결함들의 보정과 전력 증폭기의 비선형성 보상을 결합하는 새로운 기술을 제안하였다. 배열 안테나의 각각의 경로들이 동일한 특성을 갖기 위한 보정과 보상을 위해 선형 항에 3차 항을 추가한 다항식과 간접 학습 구조를 사용하였다. 본 논문에서는 컴퓨터 모의실험을 통해 성능을 입증하였다. 모의실험 결과, 단 하나의 3차 항을 추가하여 선형 결함들뿐만 아니라 모든 비선형 효과까지 효율적으로 보상함을 확인하였다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제8권1호
• /
• pp.31-36
• /
• 2008

This paper presents adaptive learning data of evolvable neural networks (ENNs) for time series prediction of nonlinear dynamic systems. ENNs are a special class of neural networks that adopt the concept of biological evolution as a mechanism of adaptation or learning. ENNs can adapt to an environment as well as changes in the enviromuent. ENNs used in this paper are L-system and DNA coding based ENNs. The ENNs adopt the evolution of simultaneous network architecture and weights using indirect encoding. In general just previous data are used for training the predictor that predicts future data. However the characteristics of data and appropriate size of learning data are usually unknown. Therefore we propose adaptive change of learning data size to predict the future data effectively. In order to verify the effectiveness of our scheme, we apply it to chaotic time series predictions of Mackey-Glass data.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.920-924
• /
• 2005

This paper presents adaptive learning data of evolvable neural networks (ENNs) for time series prediction of nonlinear dynamic systems. ENNs are a special class of neural networks that adopt the concept of biological evolution as a mechanism of adaptation or learning. ENNs can adapt to an environment as well as changes in the environment. ENNs used in this paper are L-system and DNA coding based ENNs. The ENNs adopt the evolution of simultaneous network architecture and weights using indirect encoding. In general just previous data are used for training the predictor that predicts future data. However the characteristics of data and appropriate size of learning data are usually unknown. Therefore we propose adaptive change of learning data size to predict the future data effectively. In order to verify the effectiveness of our scheme, we apply it to chaotic time series predictions of Mackey-Glass data.

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2007년도 추계학술발표논문집
• /
• pp.81-84
• /
• 2007

In this thesis, we have designed the indirect adaptive controller using Dynamic Neural Units(DNU) for unknown nonlinear systems. Proposed indirect adaptive controller using Dynamic Neural Unit based upon the topology of a reverberating circuit in a neuronal pool of the central nervous system. In this thesis, we present a genetic DNU-control scheme for unknown nonlinear systems. Our method is different from those using supervised learning algorithms, such as the backpropagation (BP) algorithm, that needs training information in each step. The contributions of this thesis are the new approach to constructing neural network architecture and its training.

• 한국정보전자통신기술학회논문지
• /
• 제6권4호
• /
• pp.243-248
• /
• 2013

In this thesis, we have designed the indirect adaptive controller using Dynamic Neural Units(DNU) for unknown nonlinear systems. Proposed indirect adaptive controller using Dynamic Neural Unit based upon the topology of a reverberating circuit in a neuronal pool of the central nervous system. In this thesis, we present a genetic DNU-control scheme for unknown nonlinear systems. Our method is different from those using supervised learning algorithms, such as the backpropagation (BP) algorithm, that needs training information in each step. The contributions of this thesis are the new approach to constructing neural network architecture and its training.

• 한국융합학회논문지
• /
• 제7권3호
• /
• pp.77-85
• /
• 2016

음향 전력 증폭기의 출력단에서 발생하는 비선형성을 보상하는 디지털 신호 처리 기술을 제안하고 그 모의실험 결과를 제시한다. 음향 전력 증폭기에 사용되는 소자에 의한 비선형성을 간접학습구조와 적응형 필터로 구성되는 디지털 신호 처리 기술로 보상한다. 적응형 필터를 사용함으로써 증폭기의 비선형 특성이 시간적으로 변하더라도 이를 적응적으로 보상할 수 있다. 모의실험 결과 전치 보상기는 3 차의 다항식으로 구현할 수 있으며 홀수차 비선형성을 효과적으로 제거할 수 있음을 보였다. 짝수 차 비선형은 출력 신호에 존재하는 직류 옵셋이 가장 큰 부분을 차지하며 이는 제안하는 기술로는 제거가 어려우므로 바이어스 회로 설계 시 유의해야 한다. 제안하는 기술은 아날로그 시스템의 본질적 특성 결함을 디지털 신호 처리 기술로서 보상할 수 있음을 보여준다.

• 국제학술발표논문집
• /
• The 9th International Conference on Construction Engineering and Project Management
• /
• pp.1146-1153
• /
• 2022

As the population is aging, accidents involving elderly people are also increasing (2014:11,667 persons; 2018: 11,797 persons). In the case of the elderly population, falling accidents are the primary direct or indirect causes of death; in particular, they face an elevated risk of staircase falls. This study proposes a method of evaluating the safety of staircases using Building Information Modeling (BIM)-based virtual simulation. By making a virtual user with the behavioral characteristics of the elderly respond to a staircase in a BIM model, its safety performance can be evaluated. The evaluation criteria were derived from regulations, elements, and characteristics relevant to the safety of staircases. To validate the proposed method, safety evaluation tests were simulated on actual staircases. The evaluation result of the test simulation shows the safety scores of 1.97 points for the elderly user and 2.95 points for the average male adult user against a required safety score of a minimum of 2 points. That is, safety is relative to users as the safety of the same staircase can be different depending upon the different behavioral characteristics of users. The study suggests that the risk of staircase-related fall accidents to the elderly can be reduced by improving staircase designs through the proposed method.