• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권9호
• /
• pp.3823-3840
• /
• 2020

Recently, auto-encoder has emerged as the most popular method in convolutional neural network (CNN) based image compression and has achieved impressive performance. In the traditional auto-encoder based image compression model, the encoder simply sends the features of last layer to the decoder, which cannot allocate bits over different spatial regions in an efficient way. Besides, these methods do not fully exploit the contextual information under different receptive fields for better reconstruction performance. In this paper, to solve these issues, a novel auto-encoder model is designed for image compression, which can effectively transmit the hierarchical features of the encoder to the decoder. Specifically, we first propose an adaptive bit-allocation strategy, which can adaptively select an importance channel. Then, we conduct the multiply operation on the generated importance mask and the features of the last layer in our proposed encoder to achieve efficient bit allocation. Moreover, we present an additional novel perceptual loss function for more accurate image details. Extensive experiments demonstrated that the proposed model can achieve significant superiority compared with JPEG and JPEG2000 both in both subjective and objective quality. Besides, our model shows better performance than the state-of-the-art convolutional neural network (CNN)-based image compression methods in terms of PSNR.

• Journal of Power Electronics
• /
• 제15권5호
• /
• pp.1348-1357
• /
• 2015

Under traditional unified power quality conditioner (UPQC) control, a UPQC series converter (SC) is mainly used to handle grid-side power quality problems while its parallel converter (PC) is mainly used to handle load-side power quality problems. The SC and PC are relatively independent. The SC is usually in standby mode and it only runs when the grid voltage abruptly changes. In this paper, novel UPQC coordinated control strategies are proposed which use the SC to share the reactive power compensation function of the PC especially without grid-side power quality problems. However, in some cases, there will be a circulating current between the SC and the PC, which will probably influence the compensation fashion, the compensation capacity, or the normal work of the UPQC. Through an active power circulation analysis, strategies with and without a circulating current are presented which fuses the reactive power allocation strategy of the SC and the PC, the composite control strategy of the SC and the compensation strategy of the DC storage unit. Both of the strategies effectively solve the SC long term idle problem, limit the influence of the circulating current, optimize all of the UPQC units and reduce the production cost. An analysis, along with simulation andexperimental results, is presented to verify the feasibility and effectiveness of the proposed control strategies.

• ETRI Journal
• /
• 제36권3호
• /
• pp.385-395
• /
• 2014

In this paper, we consider a resource allocation scheme for millimeter wave-based wireless personal area networks using directional antennas. This scheme involves scheduling the reservation period of medium access control for IEEE 802.15.3c. Objective functions are considered to minimize the average delay and maximize throughput; and two scheduling algorithms-namely, MInMax concurrent transmission and MAxMin concurrent transmission-are proposed to provide a suboptimal solution to each objective function. These are based on an exclusive region and two decision rules that determine the length of reservation times and the transmission order of groups. Each group consists of flows that are concurrently transmittable via spatial reuse. The algorithms appropriately apply two decision rules according to their objectives. A real video trace is used for the numerical results, which show that the proposed algorithms satisfy their objectives. They outperform other schemes on a range of measures, showing the effect of using a directional antenna. The proposed scheme efficiently supports variable bit rate traffic during the reservation period, reducing resource waste.

• 대한산업공학회지
• /
• 제30권2호
• /
• pp.93-106
• /
• 2004

This paper studied a maximal covering location problem with cost restrictions, to maximize level of service within predetermined cost. It is assumed that all demand have to be met. If the demand node is located within a given range, then its demand is assumed to be covered, but if it is not, then its demand is assumed to be uncovered. An uncovered demand is received a service but at an unsatisfactory level. The objective function is to maximize the sum of covered demand, Two heuristics based on the Lagrangean relaxation of allocation and decoupling are presented and tested. Upper bounds are found through a subgradient optimization and lower bounds are by a cutting algorithm suggested in this paper. The cutting algorithm enables the Lagrangean relaxation to be proceeded continually by allowing infeasible solution temporarily when the feasible solution is not easy to find through iterations. The performances are evaluated through computational experiments. It is shown that both heuristics are able to find the optimal solution in a relatively short computational time for the most instances, and that decoupling relaxation outperformed allocation relaxation.

• 한국전자거래학회지
• /
• 제12권1호
• /
• pp.73-87
• /
• 2007

세계 주요 기업들을 중심으로 비즈니스 프로세스를 보다 표준화하고, 명확화하려는 움직임이 어느 때보다 두드러지고 있다. BPM은 이러한 시도를 구현할 수 있는 적합한 도구로서 주목받고 있다. 최근 BPM으로 구현된 프로세스를 보다 효율적으로 실행하는 것이 이슈로 대두되고 있다. 본 논문에서는 프로세스 효율성을 크게 프로세스 엔진과 업무담당자 관점에서 이를 해결하고자 한다. 프로세스 엔진 관점에서는, 엔진이 사용자에게 업무를 얼마나 효율적으로 전달하는지 바라본다. 업무담당자 관점에서는, 전달된 업무를 각 업무담당자가 어떻게 처리하는지에 주목한다. 본 연구는 각 단계에서만 효율성을 고려해왔던 과거 연구와는 달리, 두 단계에서 모두 효율적일 수 있는 방법론을 제안하며, 이를 실험을 통해 유효성을 검증한다.

• 한국컴퓨터정보학회논문지
• /
• 제3권2호
• /
• pp.7-18
• /
• 1998

본 연구는 소프트웨어의 사용자 품질수준 극대화를 위한 세 가지 자원할당 모델을 제시한다. 제시되는 모델들은 소프트웨어 개발 특성상 가장 중요시되는 인적자원과 각 품질요소의 관리적 제약수준을 고려하고 있다. 각 모델의 최적 자원할당 해법은 할당된 자원량과 제품품질 수준의 관계를 규정하는 효용함수의 형태에 따라 다르다. 본 연구는 효용함수의 형태를 선형과 비선형 모델로 구분하며 또한 선형 모델은 기본모델과 확장모델로 구별하여 제시한다. 각 모델에 따른 최적 자원해법을 제시하고 비선형 모델의 경우 최적해임을 수리적으로 증명한다.

• 한국정밀공학회지
• /
• 제28권10호
• /
• pp.1196-1203
• /
• 2011

Statistical optimum methodology of table of orthogonal array, ANOM, ANOVA and RSM are applied to formulate optimum allocation design with design variables. It can be minimized average SPL of control volume, the objective function in closed system by optimal allocated positions of absorbing material. Structural natural frequency and acoustic natural frequency of cavity are analyzed by FEM and BEM in the closed system. Using BEM, average SPL of specific control volume is calculated according to the condition before using absorbing material and after using it. It is shown that noise is reduced by $5.02dB_{RMS}$ by absorbing material located at optimal position and minimum $1.83dB_{RMS}$ and maximum $3.47dB_{RMS}$ by the table of orthogonal array.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제5권12호
• /
• pp.2294-2314
• /
• 2011

The key issue in providing fast and reliable access on cloud services is the effective management of resources in a cloud system. However, the high variation in cloud service access rates affects the system performance considerably when there are no default routines to handle this type of occurrence. Adaptive techniques are used in resource management to support robust systems and maintain well-balanced loads within the servers. This paper presents an adaptive resource management for cloud systems which supports the integration of intelligent methods to promote quality of service (QoS) in provisioning of cloud services. A technique of dynamically assigning cloud services to a group of cloud servers is proposed for the adaptive resource management. Initially, cloud services are collected based on the excess cloud services load and then these are deployed to the assigned cloud servers. The assignment function uses the proposed proportional ordering which efficiently assigns cloud services based on its resource consumption. The difference in resource consumption rate in all nodes is analyzed periodically which decides the execution of service assignment. Performance evaluation showed that the proposed dynamic service assignment (DSA) performed best in throughput performance compared to other resource allocation algorithms.

• 한국통신학회논문지
• /
• 제23권6호
• /
• pp.1461-1470
• /
• 1998

비트 할당 문제는 허용 비트량을 초과하지 않는 상황에서 복호된 영상의 왜곡을 최소화하는 방법을 찾는 것으로 영상 손실 부호화에 있어서 가장 기본적으로 요구되는 점이라 할 수 있다. 이 문제의 최적 해는 Lagrangian 방식을 이용하여 계산량을 줄이면서도 최적해에 근접하는 복원 영상을 얻을 수 있는 비트 할당 기법을 제안한다. 이를 위하여 기존의 활성도 측정 방식들을 비교 검토하였으며, 활성도 값과 할당 비트량과의 관계를 파악하였다. MPEG-1 Intra 부호화 방법에 제안 기법을 적용하여 실험 영상을 부호화 하는 경우, 적은 계산량만으로도 최적 해에 근접하는 결과를 얻을 수 있었다.

• 수산해양교육연구
• /
• 제17권2호
• /
• pp.180-192
• /
• 2005

In this paper I tried quantitative measures of the rent-seeking for budgetary allocations in Korean fisheries. There is a great deal of ignorance about the level of rent-seeking that takes place as a consequence of government expenditures and the potential for influencing their allocation, and so the attempt by Kats and Rosenberg(1989) to measure this in terms of the extent of annual changes in these allocations was a significant step forward. The opportunity to concentrate net public expenditure benefits within the fiscal function of the state gives rise to rent-seeking by special interest taxpayer-citizens through government budget allocations, special interest groups and coalitions of special interest groups. There are several important implications of budgetary rent-seeking. Budgetary rent-seeking is composed of 'income' and substitution effects, the budgetary 'income' effect is the general increase in government expenditures from one period to another. According to result of this analysis, average expenditures were 17.0% of total fisheries budget in 1966-2005. Therefore, distribution of fisheries budget should be converted by indirect support systems to reduce these rent-seeking behaviors.