• 융합신호처리학회논문지
• /
• 제6권1호
• /
• pp.39-44
• /
• 2005

스케줄링은 CDFG 내의 각 연산에 우선순위 관계를 유지하면서 연산이 수행될 제어스텝을 할당하는 과정으로 합성된 하드웨어의 성능에 직접적인 영향을 미치는 중요한 단계이다. 본 논문에서는 자원제한 스케줄링 알고리즘을 제안한다. 제안된 알고리즘은 주어진 그래프를 분석하여 연산유닛의 개수를 결정하고 이에 따라 각 연산을 제어스텝에 할당한다. 스케줄링 과정 중에 상대적으로 부족한 연산유닛과 여유 있는 연산유닛을 구별하여 연산유닛의 수를 조절한 후 반복적으로 성능개선을 시도하게 된다. 제안된 알고리즘의 성능을 평가하기 위하여 모의실험을 수행하였고 그 결과는 기존의 방법들에 비해 우수함을 알 수 있었다.

• Journal of Information Processing Systems
• /
• 제13권5호
• /
• pp.1183-1202
• /
• 2017

Image restoration has been carried out by texture synthesis mostly for large regions and inpainting algorithms for small cracks in images. In this paper, we propose a new approach that allows for the simultaneous fill-in of different structures and textures by processing in a wavelet domain. A combination of structure inpainting and patch-based texture synthesis is carried out, which is known as patch-based inpainting, for filling and updating the target region. The wavelet transform is used for its very good multiresolution capabilities. The proposed algorithm uses the wavelet domain subbands to resolve the structure and texture components in smooth approximation and high frequency structural details. The subbands are processed separately by the prioritized patch-based inpainting with isophote energy driven texture synthesis at the core. The algorithm automatically estimates the wavelet coefficients of the target regions of various subbands using optimized patches from the surrounding DWT coefficients. The suggested performance improvement drastically improves execution speed over the existing algorithm. The proposed patch optimization strategy improves the quality of the fill. The fill-in is done with higher priority to structures and isophotes arriving at target boundaries. The effectiveness of the algorithm is demonstrated with natural and textured images with varying textural complexions.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2017년도 하계학술대회
• /
• pp.83-84
• /
• 2017

We propose a novel high dynamic range (HDR) video synthesis algorithm from alternatively exposed low dynamic range (LDR) videos. We first estimate correspondences between input fames using an illumination invariant descriptor. Then, we synthesize an HDR frame with the weights computed to maximize detail preservation in the output HDR frame. Experimental results demonstrate that the proposed algorithm provides high-quality HDR videos without noticeable artifacts.

• 전자공학회논문지A
• /
• 제28A권1호
• /
• pp.74-84
• /
• 1991

This paper presents an algorithm for technology-dependent logic optimization and technology mapping, and describes a performance-driven logic synthesis system, SILOS, implemented based on the proposed algorithm. The system analyzes circuits and resynthesizes the critical sections such that generated circuit operates opertes within time constraints, using only gate types supported by library for direct implementation. Experimental results show that the system can be a viable tool in synthesizing high-performance logic circuits.

• International Journal of Computer Science & Network Security
• /
• 제21권3호
• /
• pp.304-310
• /
• 2021

The paper presents problems of optimization of the synthesis of embedded systems, in particular Pareto optimization. The model of such a system for its design for high-level of abstract is based on the classic approach known from the theory of task scheduling, but it is significantly extended, among others, by the characteristics of tasks and resources as well as additional criteria of optimal system in scope structure and operation. The metaheuristic algorithm operating according to this model introduces a new approach to system synthesis, in which parallelism of task scheduling and resources partition is applied. An algorithm based on a genetic approach with simulated annealing and Boltzmann tournaments, avoids local minima and generates optimized solutions. Such a synthesis is based on the implementation of task scheduling, resources identification and partition, allocation of tasks and resources and ultimately on the optimization of the designed system in accordance with the optimization criteria regarding cost of implementation, execution speed of processes and energy consumption by the system during operation. This paper presents examples and results for multi-criteria optimization, based on calculations for specifying non-dominated solutions and indicating a subset of Pareto solutions in the space of all solutions.

• 전자공학회논문지B
• /
• 제32B권7호
• /
• pp.1025-1035
• /
• 1995

In this paper, we realize a real time digital FM synthesizer based on genetic algorithm using a general purpose digital signal processor. Especially, we synthesize diverse music sounds nicely using a synthesis model consisting of a single modulator and multiple carriers. Also we present genetic algorithm-based technique which determines optimal parameters for reconstruction through FM synthesis of a sound after analyzing the spectrum of PCM data as a standard music sound using FFT. Using the suggested parameter extractiuon algorithm, we extract parameters of several instruments and then synthesize digital FM sounds. To verify the validity of the parameter extraction algorithm as well as realization of a real time digital music synthesizer, the evaluation is first done by listening the sound directly as subjective test. Secondly, to evaluate the synthesized sound objectively with an engineering sense, we compare the synthesized sound with an original one in a time domain and a frequency domain.

• 전자공학회논문지A
• /
• 제31A권7호
• /
• pp.117-124
• /
• 1994

In this pape, we propose a graph-theoretic approach for solving the allocation problem for the synthesis of datapaths based on multiport memories. An efficient algorithm is devised by using the weighted bipartite matching algorithm to assign variables to each port of memory modules. The proposed algorithm assigns program variables into a minimum number of multiport memory modules such that usage of memory elements and interconnection cost can be kept minimal. Experimental results show that the proposed algorithm generates the datapaths with fewer registers in memory modules and less interconnection cost for several benchmarks available from the literatures.

• 한국정밀공학회지
• /
• 제16권10호
• /
• pp.126-131
• /
• 1999

Up until now, it is said that no satisfactory computer solutions have been found for synthesizing four-bar linkage based on the prescribed coupler link curve. In our study, an algorithm has been developed to improve the design synthesis of four-bar linkage automatically generating prescribed path by using B-spline interpolation. The suggested algorithm generates the desired coupler curve by using B-spline interpolation, and hence the generated curve approximates as closely as to the desired curve representing coupler link trajectory. Also, when comparing each generated polygon with the control polygon, rapid comparison by applying convex hull concept. finally, optimization process using ADS is incorporated into the algorithm based on the 5 precision point method to reduce the total optimization process time. As for examples, three different four-bar linkages were tested and the results showed the effectiveness of the algorithm.

• 융합신호처리학회 학술대회논문집
• /
• 한국신호처리시스템학회 2001년도 하계 학술대회 논문집(KISPS SUMMER CONFERENCE 2001
• /
• pp.69-72
• /
• 2001

High-level synthesis는 주어진 동작과 면적, 성능, 전력 소모, 패키징, 테스팅 등의 주어진 제한을 만족하게 구현된 구조적 디자인을 생성한다. 즉 high-level syntehesis란 디지털 시스템의 알고리즘 레벨 서술로부터 레지스터 전달구조의 구현에 이르는 과정을 의미한다. 이러한 high-level synthesis의 과정은 컴파일, 분할, 스케줄링 등의 단계를 거쳐 디지털 시스템을 설계할 수 있다. 본 논문에서는 high-level synthesis의 단계중 분할 과정을 연구하고, 분할 알고리즘 중에서 min-cut 알고리즘과 simulated annealing 알고리즘을 사용하여 비교 분석하였다.

• 한국음향학회지
• /
• 제28권2호
• /
• pp.119-126
• /
• 2009

본 논문에서 제안한 줄 없는 기타는 크게 레이저 현과 프렛, 음 합성 알고리듬과 프로세서로 구성된다. 레이저 현은 레이저 모듈과 포토다이오드를 이용하여 스트로크와 아르페지오를 표현할 수 있도록 하였고, 프렛은 전압 분배기를 이용하여 구현하였다. 몸통은 물리적 모델링 합성법을 이용하였기 때문에 줄 없는 기차에는 울림통의 역학을 하는 물리적인 몸통이 없다. 제안한 기타의 프렛은 실제 프렛을 동일하게 표현할 수 있어 기존의 코드 글러브를 이용한 로드 표현뿐 아니라 실제 기타와 같은 솔로 연주도 가능하다. 해머링 온, 풀링 오프, 슬라이딩과 같이 프렛 변화가 있는 연주음은 전압 분배기로부터 받은 프렛의 정보를 파라미터로 사용하여 합성한다. 연구법에 따른 음의 피치 변화는 디지털 도파관 모델에서 파동의 전파 속도 변화로 표현하였다. 이 합성 모델은 동일 프렛에서 현의 장력을 변화시켜 연주하는 비브라토 음도 합성 할 수 있다. 레이저 현과 프렛으로부터 받아들인 정보를 합성 알고리듬의 파라미터로 변환하여 기타 음을 생성하고 이를 실시간으로 출력할 수 있도록 TMS320F2812를 사용하였다. 웹에 공개한 동영상에는 제안한 알고리듬과 인터페이스를 이용하여 실시간으로 합성한 '아리랑' 연주를 볼 수 있다. 제안한 알고리듬이 피치 변화를 표현하는 기타 솔로 연주법에 효과적이고 줄 없는 기타로 실시간 연주가 가능함을 확인할 수 있다.