• 한국통신학회논문지
• /
• 제25권10B호
• /
• pp.1760-1768
• /
• 2000

비동기 DS-CDMA 다중 사용자 환경의 이동통신 시스템에서 채널의 효율을 높이기 위해 학습 신호열을 사용하지 않는 블라인드 등화 기법을 제안한다. 블라인드 등화 기법 중에서 가장 단순하면서 성능이 좋고 구현하기 쉬운 CMA(Constant Modulus Algorithm)는 근거리-원거리 효과(near-far effect)가 심하게 나타날 경우 전력이 약한 사용자들의 수렴 영역이 작아져 mis-convergence 위험성이 증가한다. 이런 경우에도 Newton 방식을 이용한 등화 기법은 자승 오차와 Eye-Pattern의 성능 비교를 통해 기존의 방법 보다 우수함을 확인할 수 있었다.

• 한국통신학회논문지
• /
• 제28권7C호
• /
• pp.705-711
• /
• 2003

This paper studies the application of a fuzzy-ARTMAP neural network to digital communications channel equalization. This approach provides new solutions for solving the problems, such as complexity and long training, which found when implementing the previously developed neural-basis equalizers. The proposed fuzzy-ARTMAP equalizer is fast and easy to train and includes capabilities not found in other neural network approaches; a small number of parameters, no requirements for the choice of initial weights, automatic increase of hidden units, no risk of getting trapped in local minima, and the capability of adding new data without retraining previously trained data. In simulation studies, binary signals were generated at random in a linear channel with Gaussian noise. The performance of the proposed equalizer is compared with other neural net basis equalizers, specifically MLP and RBF equalizers.

• 보건행정학회지
• /
• 제27권2호
• /
• pp.149-156
• /
• 2017

Background: This study was conducted to evaluate the performance of the Hierarchical Condition Category (HCC) model, identify potentially high-cost patients, and examine the effects of adding prior utilization to the risk model using Korean claims data. Methods: We incorporated 2 years of data from the National Health Insurance Services-National Sample Cohort. Five risk models were used to predict health expenditures: model 1 (age/sex groups), model 2 (the Center for Medicare and Medicaid Services-HCC with age/sex groups), model 3 (selected 54 HCCs with age/sex groups), model 4 (bed-days of care plus model 3), and model 5 (medication-days plus model 3). We evaluated model performance using $R^2$ at individual level, predictive positive value (PPV) of the top 5% of high-cost patients, and predictive ratio (PR) within subgroups. Results: The suitability of the model, including prior use, bed-days, and medication-days, was better than other models. $R^2$ values were 8%, 39%, 37%, 43%, and 57% with model 1, 2, 3, 4, and 5, respectively. After being removed the extreme values, the corresponding $R^2$ values were slightly improved in all models. PPVs were 16.4%, 25.2%, 25.1%, 33.8%, and 53.8%. Total expenditure was underpredicted for the highest expenditure group and overpredicted for the four other groups. PR had a tendency to decrease from younger group to older group in both female and male. Conclusion: The risk adjustment models are important in plan payment, reimbursement, profiling, and research. Combined prior use and diagnostic data are more powerful to predict health costs and to identify high-cost patients.

• 한국BIM학회 논문집
• /
• 제6권2호
• /
• pp.39-46
• /
• 2016

While institutional matters such as improvement on Basic Guidelines for Construction Safety are greatly concerned to reduce falling accidents at construction sites, there are short of studies on how to practically predict accident signs at construction sites and to preemptively prevent them. As one of existing accident prevention methods, it was attempted to build the early warning system based on standardized accident scenarios to control the situations. However, the investment cost was too high depending on the site situation, and it did not help construction workers directly since it was developed to mainly provide support operational work support to safety managers. In the long run, it would be possible to develop the augmented reality based accident prevention method from the worker perspective by extracting product information from BIM, visually rendering it along with site installation materials term and comparing it with the site situation. However, to make this method effective, the BIM model should be implemented first and the technology that can promptly process site situations should be introduced. Accordingly, it is necessary to identify risk signs through lightweight image processing to promptly respond only with currently available resources. In this study, it was intended to propose the system concept that identified potential risk factors of falling accidents by histogram equalization, which was known as the fastest image processing method presently, used visual words, which could enhance model classification by wording image records, to determine the risk factors and notified them to the work manager.

• International Journal of Internet, Broadcasting and Communication
• /
• 제14권2호
• /
• pp.47-54
• /
• 2022

For safe driving of autonomous vehicles, road damage detection is very important to lower the potential risk. In order to ensure safety while an autonomous vehicle is driving on the road, technology that can cope with various obstacles is required. Among them, technology that recognizes static obstacles such as poor road conditions as well as dynamic obstacles that may be encountered while driving, such as crosswalks, manholes, hollows, and speed bumps, is a priority. In this paper, we propose a method to extract similarity of images and find damaged road images using OpenCV image processing and CNN algorithm. To implement this, we trained a CNN model using 280 training datasheets and 70 test datasheets out of 350 image data. As a result of training, the object recognition processing speed and recognition speed of 100 images were tested, and the average processing speed was 45.9 ms, the average recognition speed was 66.78 ms, and the average object accuracy was 92%. In the future, it is expected that the driving safety of autonomous vehicles will be improved by using technology that detects road obstacles encountered while driving.

• 한국멀티미디어학회논문지
• /
• 제4권4호
• /
• pp.333-338
• /
• 2001

본 논문에서는 이전에 개발된 신경회로망 채널 등화기에서 볼 수 있었던 구조의 복잡성 및 많은 학습시간의 소요 등과 같은 단점을 극복하고자 퍼지-ARTMAP 신경망을 이용하여 채널 등화기를 구성하였다. 제안된 퍼지-ARTMAP 채널 등화기는 다른 형태의 신경망을 이용한 등화기에서는 찾아 볼 수 없는 빠르고 쉬운 학습 능력을 갖고 있다. 즉, 등화기 구성에 필요한 파라미터의 수가 적으며 지역적 최소값에 빠질 우려 없이 각 계층간의 초기 연결강도를 지정할 수 있을 뿐만 아니라 기존의 학습된 데이터를 재학습시킬 필요 없이 새로운 데이터를 단순히 추가 학습시킬 수 있는 장점 등을 가지고 있다. 본 연구의 시뮬레이션 과정에서는 선형채널에서 발생된 가우시안 잡음을 동반한 이진 신호를 대상으로 퍼지-ARTMAP 채널 등화기의 성능을 LMS 기반의 선형 등화기 및 MLP와 RBF 신경망 등화기와 비교하였으며 퍼지-ARTMAP 등화기가 상대적으로 간단한 구조와 빠른 처리속도를 가짐은 물론 선형등화기로 해결하지 못했던 비선형 문제들도 해결할 수 있음을 보였다.

• 한국콘텐츠학회논문지
• /
• 제13권5호
• /
• pp.303-310
• /
• 2013

초음파영상은 갑상선 질병에서 결절성 갑상선 질병을 진단하는 검사로서 결절의 위치, 크기, 개수, 내부 에코 특성에 대한 정보를 제공하여 암의 가능성이 높은 고위험 결절을 선별하며, 세침흡인 검사 시 정확한 유도를 가능하게 한다. 갑상선 결절 중 악성으로 진단되는 경우는 5% 미만이지만 초음파에서 감별진단이 중요하다. 그러므로 본 연구에서는 병리학적으로 갑상선 유두암으로 진단된 증례를 실험 대상으로 하며, 영역을 묘사하는 알고리즘으로 그 질감을 정량화하는 방법으로 질감특징 분석(TFA)를 적용하여 컴퓨터자동진단의 검출 효율을 실험하였다. 초음파영상에서 관심영역을 설정하여 $50{\times}50$ 픽셀 크기, 히스토그램 평활화로 전처리하여 실험영상을 획득하였다. 전체영상 70증례에서 갑상선 유두암의 영상 35증례를 테스트 영상으로 하고, 고유영상 생성의 정상영상 35증례를 학습영상으로 실험하였다. 질감특징 분석 알고리즘을 적용한 실험결과 GLavg, SKEW, UN, ENT 4개 파라미터의 질병 검출 효율이 91~100%로 높게 나타났다. 이는 갑상선 결절 질병을 감별하는 컴퓨터자동진단의 응용을 나타내며, 갑상선 질병의 감별진단에 전처리 자동진단 가능성을 나타낸다. 향후 추가적인 관련 알고리즘의 연구가 계속 진행된다면 갑상선 질병의 컴퓨터자동진단의 실용화기반을 마련할 수 있을 것이고, 다양한 초음파영상의 질병에 대한 적용이 가능할 것으로 사료된다.