• 전기전자학회논문지
• /
• 제17권3호
• /
• pp.254-259
• /
• 2013

본 논문에서는 준감독 학습 알고리즘(Semi-Supervised Learning Algorithm)의 학습데이터에 필요한 소수의 레이블 데이터를 능동적으로 선택하기 위한 무감독경쟁학습 알고리즘인 VCNN(Vector Centroid Neural Network)을 제안한다. 준감독 학습 알고리즘에서 레이블 데이터의 선택은 학습 결과 큰 영향을 미치고, 레이블 데이터를 선택하는데 있어 많은 비용과 전문적인 지식이 필요하다. 본 논문에서 능동적이고 효율적인 레이블 데이터 선택을 검증하기 위하여 UCI database 와 caltech dataset 을 이용하여 실험한 결과, 기존의 레이블 데이터 선택 방법과 비교하여 안정된 분류 결과와 최소의 오차율을 나타냈다.

• 대한전자공학회논문지TC
• /
• 제47권3호
• /
• pp.31-38
• /
• 2010

본 논문에서는 MIMO-OFDM 시스템에 여러 MIMO 기법을 적용한 후 모의실험을 통하여 성능을 비교 및 분석하였고, 또한 Pre-coding, Antenna Subset Selection, AMC 기법을 적용하여 시스템을 분석하였다. 그리고 최종적으로 제안한 Adaptive-MCM(Adaptive-MIMO-Coding-Modulation) 기법의 성능을 분석하였다. 그 결과 4세대 이동통신 시스템에서 요구되고 있는 최대 전송률과 고품질 서비스의 Trade-off 관계를 개선할 수 있음을 확인하였다.

• 한국정보전자통신기술학회논문지
• /
• 제8권5호
• /
• pp.384-393
• /
• 2015

본 논문은 두 가지 속성 삭제 방법인 ReliefF와 SVM-REF를 조합하여 유전자 선택을 위한 속성 삭제에 기반을 둔 최적화된 분류법(OCFE)을 제안한다. ReliefF 알고리즘은 데이터의 중요도에 따라 데이터 순위를 매기고 필터(filter) 속성 선택 알고리즘이다. SVM-RFE 알고리즘은 속성의 가중치 기반으로 데이터 순위를 매기고 데이터를 감싸는 래퍼(wrapper) 속성 선택 알고리즘이다. 이러한 두 가지 기법을 조합함으로써, 우리는 SVM-RFE는 0.3096779이고 OCFE는 0.3016138으로 에러율 평균이 좀 더 낮게 나타났다. 또한, 제안된 기법은 SVM-RFE가 69%이고 OCFE는 70%으로 좀 더 정확한 것으로 나타났다.

• 보험의학회지
• /
• 제26권
• /
• pp.31-39
• /
• 2007

Background and main issue: In the Korean insurance market, an outstanding issue is the decrease of margin of risk ratio. This affects the solvency and profitability of insurance companies. Insurance medicine, which has been developed in Western countries, is so-called medical risk selection or medical underwriting. Medical risk selection is based on clinical follow-up study and mortality analysis methodology. Unfortunately, there have been few clinical follow-up studies, and no intercompany disease analysis system is available in the Korean insurance market. In practice, we use underwriting guidelines, which were developed by some global reinsurance companies. However, these guidelines were developed under clinical follow-up studies performed abroad. So, we cannot rule out underestimation of excess mortality factors such as mortality ratio, excess death rate, and life expectancy. It is necessary to perform medical assessment in claims administration. Comparing the insured's statement by medical records with products' benefit according to this procedure, we can make sound claim decisions and participate in the role of sound underwriting. We can call this scientific procedure as the verification of medical claims review. Another area of medical claims review is medical counsel for claims staff. Result: There is another insurance medicine in addition to medical risk selection. Independent medical assessment by medical records of insured is medical claims review. Medical claims review is composed of verification and counsel.

• 한국통신학회논문지
• /
• 제38A권3호
• /
• pp.249-257
• /
• 2013

최근 고속 데이터 서비스의 보급으로 데이터 사용량이 급증함에 따라 이를 수용하기 위한 주파수 자원의 효율적 사용 방법에 대한 연구가 주목받고 있다. 따라서 주파수 효율성을 향상시키기 위해 주파수 사용에 대한 우선권을 가진 사용자와 후순위에 있는 사용자가 해당 주파수 대역을 공유하는 기술인 인지 무선 기술(Cognitive Radio, CR)이 제안되었다. 본 논문에서는 CR 시스템에서 직교하는 채널 벡터를 이용하여 선순위 시스템에게 미치는 간섭을 줄이면서 후순위 시스템의 성능 향상을 동시에 고려한 사용자 선택 기법을 제안한다. 모의실험 결과 제안 기법을 적용했을 때 전체 시스템의 합 용량이 기존 기법보다 약 1.62bps/Hz의 이득을 가지는 것을 확인하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권12호
• /
• pp.4269-4292
• /
• 2014

Currently, increasing numbers of access points (AP) are being deployed in enterprise offices, campuses and municipal downtowns for flexible Internet connectivity, but most of these access points are idle or redundant most of the time, which causes significant energy waste. Therefore, with respect to power conservation, applying energy efficient strategies in WIFI networks is strongly advocated. One feasible method is dynamically managing network resources, particularly APs, by powering devices on or off. However, when an AP is powered on, the device is initialized through a long boot time, during which period clients cannot be associated with it; therefore, the network performance would be greatly impacted. In this paper, based on a global view of an entire WLAN, we propose an AP selection technology, known as Temporary Access Selection (TAS). The criterion of TAS is a fusion metric consisting of two evaluation indexes which are based on throughput and battery life, respectively. TAS is both service and clients' preference specific through balancing the data rate, battery life and packet size. TAS also works well independently in traditional WLANs in which no energy efficient strategy is deployed. Moreover, this paper demonstrates the feasibility and performance of TAS through experiments and simulations with Network Simulator version 3 (NS3).

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권4호
• /
• pp.1390-1405
• /
• 2014

This paper investigates optimal relay selection and power allocation under an aggregate power constraint for cooperative wireless sensor networks assisted by amplify-and-forward relay nodes. By considering both transmission power and circuit power consumptions, the received signal-to-noise ratio (SNR) at the destination node is calculated, based on which, a relay selection and power allocation scheme is developed. The core idea is to adaptively adjust the selected relays and their transmission power to maximize the received SNR according to the channel state information. The proposed scheme is derived by recasting the optimization problem into a three-layered problem-determining the number of relays to be activated, selecting the active relays, and performing power allocation among the selected relays. Monte Carlo simulation results demonstrate that the proposed scheme provides a higher received SNR and a lower bit error rate as compared to the average power allocation scheme.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권11호
• /
• pp.5209-5228
• /
• 2016

Device-to-device (D2D) communication underlaying cellular networks can bring significate benefits for improving the performance of mobile services. However, it hinges on elaborate resource sharing scheme to coordinate interference between cellular users and D2D pairs. We formulate a joint mode selection, link allocation and power control optimization problem for D2D communication sharing uplink resources in a multi-user cellular network and consider the efficiency and the fairness simultaneously. Due to the non-convex difficulty, we propose a three-step scheme: firstly, we conduct mode selection for D2D pairs based on a minimum distance metric after an admission control and obtain some cellular candidates for them. And then, a cellular candidate will be paired to each D2D pair based on fairness. Finally, we use Lagrangian Algorithm to formulate a joint power control strategy for D2D pairs and their reused cellular users and a closed-form of solution is derived. Simulation results demonstrate that our proposed algorithms converge in a short time. Moreover, both the sum rate of D2D pairs and the energy efficiency of cellular users are improved.

• 대한안전경영과학회:학술대회논문집
• /
• 대한안전경영과학회 2009년도 추계학술대회
• /
• pp.231-249
• /
• 2009

The 'Excellent Quality Competitive Enterprises' selection system began with the government's selection and announcement of 'Top 100 Excellent Quality Competitive Enterprises' in 1997 in accordance with Article 6 of the Quality Management and Safety Control of Industrial Products Act and Article 5 (Selection of excellent quality management enterprises, etc.) of the Act's Enforcement Decree. The content of the research and analysis results is summarized as follows: 1. The participation and confidence of enterprises in the differentiation of the assessment system for the excellent quality competitive enterprises selection system is high: from the initial year to the present, 10% of enterprises have participated in the system for over 10 years and since 2000 participating enterprises have been on a steady rise. 2. The distribution of participating enterprises is not limited to any specific region but is spread nationwide, while the participation scale and business types also are being evenly distributed. 3. From the perspective of management performance, as compared with the average of domestic manufacturing businesses or that of U.S. manufacturing businesses, an analysis found that among management results, the excellent quality competitive enterprises achieved far higher sales growth and operating profit rate on sales, however their own capital ratio and debt to equity ratio appeared to be slightly higher.

• 대한음성학회지:말소리
• /
• 제63호
• /
• pp.125-137
• /
• 2007

Recently, many techniques have been proposed to improve the noise robustness for speaker verification. In this paper, we consider the feature recombination technique in multi-band approach. In the conventional feature recombination for speaker verification, to compute the likelihoods of speaker models or universal background model, whole feature components are used. This computation method is not effective in a view point of multi-band approach. To deal with non-effectiveness of the conventional feature recombination technique, we introduce a subband likelihood computation, and propose a modified feature recombination using subband likelihoods. In decision step of speaker verification system in noise environments, a few very low likelihood scores of a speaker model or universal background model cause speaker verification system to make wrong decision. To overcome this problem, a reliable feature selection method is proposed. The low likelihood scores of unreliable feature are substituted by likelihood scores of the adaptive noise model. In here, this adaptive noise model is estimated by maximum a posteriori adaptation technique using noise features directly obtained from noisy test speech. The proposed method using subband-based reliable feature selection obtains better performance than conventional feature recombination system. The error reduction rate is more than 31 % compared with the feature recombination-based speaker verification system.