• 제목/요약/키워드: MFSC

검색결과 6건 처리시간 0.022초

무선 홈 네트워킹 서비스를 위한 스마트 캐싱 기법 (A Smart Caching Scheme for Wireless Home Networking Services)

  • 이종득
    • 디지털융복합연구
    • /
    • 제17권9호
    • /
    • pp.153-161
    • /
    • 2019
  • 무선 홈 프록시에서 미디어 객체 세그먼트들에 대한 식별은 캐싱 지연에 중요한 영향을 미치게 되며, 캐싱 지연은 프록시의 성능을 떨어뜨리는 원인을 제공한다. 본 논문에서는 홈 네트워크의 프록시 성능을 향상시키고 미디어 객체 세그먼트들에 대한 캐싱 성능을 향상시키기 위하여 SFSC(Single Fetching Smart Caching)전략과 MFSC(Multi-Fetching Smart Caching)전략을 제안한다. SFSC전략은 홈 노드가 요청한 객체 세그먼트를 한 번에 하나씩 순차적으로 패칭하여 캐싱을 수행하는 기법으로서 보다 빠른 캐시 히트율을 보장해 주는 기법이며, MFSC전략은 홈 노드가 요청한 객체 세그먼트를 한 번에 여러 개씩 block화하여 캐시하는 기법으로서 처리율을 향상시키기 위한 기법이다. 시뮬레이션 결과 SFSC기법은 MFSC기법에 비해서 캐시 히트율과 캐싱 지연이 보다 효율적임을 알 수 있었으며, 반대로 객체 세그먼트에 대한 처리율은 MFSC기법이 SFSC기법에 비해서 보다 효율적임을 알 수 있었다.

스펙트럼 분석과 신경망을 이용한 음성/음악 분류 (Speech/Music Discrimination Using Spectrum Analysis and Neural Network)

  • 금지수;임성길;이현수
    • 한국음향학회지
    • /
    • 제26권5호
    • /
    • pp.207-213
    • /
    • 2007
  • 본 연구에서는 스펙트럼 분석과 신경망을 이용한 효과적인 음성/음악 분류 방법을 제안한다. 제안하는 방법은 스펙트럼을 분석하여 스펙트럴 피크 트랙에서 지속성 특징 파라미터인 MSDF(Maximum Spectral Duration Feature)를 추출하고 기존의 특징 파라미터인 MFSC(Mel Frequency Spectral Coefficients)와 결합하여 음성/음악 분류기의 특징으로 사용한다. 그리고 신경망을 음성/음악 분류기로 사용하였으며, 제안하는 방법의 성능 평가를 위해 학습 패턴 선별과 양, 신경망 구성에 따른 다양한 성능 평가를 수행하였다. 음성/음악 분류 결과 기존의 방법에 비해 성능 향상과 학습 패턴의 선별과 모델 구성에 따른 안정성을 확인할 수 있었다. MSDF와 MFSC를 특징 파라미터로 사용하고 50초 이상의 학습 패턴을 사용할 때 음성에 대해서는 94.97%, 음악에 대해서는 92.38%의 분류율을 얻었으며, MFSC만 사용할 때보다 음성은 1.25%, 음악은 1.69%의 향상된 성능을 얻었다.

다문화가족지원센터 종사자의 가족영역 지원 사업 인식 및 가족생활주기별 다문화가족 프로그램 필요도에 관한 연구 (Multicultural Family Support Center Staff's Perceptions on Family Programs and Needs of Multicultural Family Programs based on the Family Life Cycle)

  • 전미경;강복정;손서희;이은주
    • 가정과삶의질연구
    • /
    • 제34권3호
    • /
    • pp.147-163
    • /
    • 2016
  • Multicultural Family Support Centers (MFSC) have provided five program areas including family, gender equality, human rights, social integration, and counseling since 2015. It is important to examine the MFSC staff's perceptions of the family programs and program needs for multicultural families to improve the effectiveness of these programs. In this study, we aim to explore the staff's perceptions of the family programs and to assess multicultural family program needs based on the family life cycle. A total of 130 MFSC staff were recruited through 128 MFSCs across Korea. Descriptive statistics were conducted for data analysis. Our findings revealed that MFSC staff have increasingly recognized that their target audiences are not only marriage immigrants and their families but also foreign workers' families and families from North Korea. In addition, the MFSC staff identified the importance of family programs instead of only programs for individual family members and multicultural families' different program needs based on their family life cycle. Contrary to the positive perception of the family programs, they described challenges of the family programs including a shortage of funds, difficulty recruiting family member combinations (e.g., couples, parent-child) for family programs, a lack of a program manual, and a heavy workload. This study provides insights into the family programs including their development and delivery.

다문화가족지원센터 상담자의 직무스트레스와 심리적 소진과의 관계 (The Relationship between Job-stress and Psychological-exhaustion of Counselors at Multicultural Family Support Center)

  • 선국진
    • 한국컴퓨터정보학회논문지
    • /
    • 제18권7호
    • /
    • pp.157-164
    • /
    • 2013
  • 본 연구는 다문화가족지원센터 상담자들의 직무스트레스가 그들의 심리적 소진에 미치는 영향에 대해서 분석하였다. 분석결과를 통하여 다문화가족지원센터 상담자들의 직무스트레스로 인한 심리적 소진의 원인을 규명하고, 직무스트레스를 완화시켜 다문화가족지원센터 상담자들에 대한 효율적인 인적자원관리의 방안을 제시하는 데 연구의 목적을 두었다. 분석결과, 다문화가족지원센터 상담자들의 직무스트레스는 심리적 소진의 정서적 고갈, 비인격화, 자아성취감 저하요인에 영향을 미치는 것으로 밝혀졌다. 이러한 결과를 반영하여 본 연구에서는 다문화가족지원센터 상담자들의 직무스트레스와 심리적 소진을 예방하기 위한 현실적인 방안마련의 필요성을 제시하였다.

임신중 피로에 관한 조사연구 (A Study on Pregnant Women's Fatigue)

  • 박영숙;이혜경;이은희;엄영란;강남미;이숙희;허명행;김혜원
    • 여성건강간호학회지
    • /
    • 제4권3호
    • /
    • pp.429-440
    • /
    • 1998
  • A survey was conducted to examine the fatigue and the related factors in pregnancy. 426 pregnant women were recruited from antenatal clinic at 8 hospital from November 1, 1997 to January 1, 1998. Fatigue was measured by the Modification of the Fatigue Symptoms Checklist(MFSC) developed by the Industrial Fatigue Research Committee of the Japanese Association of Industrial health(Yoshitake, 1978). The MFSC consisted of 34 items with 4-point Likert scale. Minor discomfort during pregnancy was measured using a 14-item checklist with 4-point Likert scale. The data were analyzed by percentage, t-test, ANOVA and Pearson correlation. The major findings of study were as follows : 1. The average fatigue score during pregnacy was 63.02 and the average minor discomfort score was 26.92. 2. The fatigue score of nullipara was significantly lower than that of mutipara. There was a significant difference of fatigue score between pregnant women with conjugal family and those with extended family. The fatigue score during 1st trimester was the highest in comparison with that of 2nd and 3rd trimester of pregnancy. 3. The fatigue score of pregnant women who worked for house-keeping only was significantly lower than that of pregnant women who worked partially or did not work for house-keeping. 4. There were significant correlations among each type of measure of fatigue and discomfort. Fatigue was the most common symptom during pregnancy and related to type of family, house-keeping work, and parity. Nursing intervention such as self-car, relaxation, and nutrition would be developed to manage fatigue problem in pregnancy.

  • PDF

MFSC: Mean-Field-Theory and Spreading-Coefficient Based Degree Distribution Analysis in Social Network

  • Lin, Chongze;Zheng, Yi
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제12권8호
    • /
    • pp.3630-3656
    • /
    • 2018
  • Degree distribution can provide basic information for structural characteristics and internal relationship in social network. It is a critical procedure for social network topology analysis. In this paper, based on the mean-field theory, we study a special type of social network with exponential distribution of time intervals. First of all, in order to improve the accuracy of analysis, we propose a spreading coefficient algorithm based on intimate relationship, which determines the number of the joined members through the intimacy among members. Then, simulation show that the degree distribution of follows the power-law distribution and has small-world characteristics. Finally, we compare the performance of our algorithm with the existing algorithms, and find that our algorithm improves the accuracy of degree distribution as well as reducing the time complexity significantly, which can complete 29.04% higher precision and 40.94% lower implementation time.