• 한국멀티미디어학회논문지
• /
• 제19권8호
• /
• pp.1415-1423
• /
• 2016

Interference in device-to-device (D2D) communication underlaying cellular network needs to be elaborately investigated because of channel sharing. The objective is to improve the quality of D2D communications while maintaining high performance for cellular users. In this paper, we solve the above problem by jointly considering channel allocation and power control using coalition formation game. Our cooperative game theoric approach allows to enhance network-wide performance. We design a merge-and-split algorithm to deal with the complexity of the combinatorial structure in coalition formation problem. The analytical and numerical results show that our algorithm converges to a stable point which achieves high network performance.

• 한국멀티미디어학회논문지
• /
• 제22권7호
• /
• pp.749-758
• /
• 2019

This paper proposes a data augmentation algorithm to improve the performance of DNN(Deep Neural Network) based speech enhancement. Many deep learning models are exploring algorithms to maximize the performance in limited amount of data. The most commonly used algorithm is the data augmentation which is the technique artificially increases the amount of data. For the effective data augmentation algorithm, we used a formant enhancement method that assign the different weights to the formant frequencies. The DNN model which is trained using the proposed data augmentation algorithm was evaluated in various noise environments. The speech enhancement performance of the DNN model with the proposed data augmentation algorithm was compared with the algorithms which are the DNN model with the conventional data augmentation and without the data augmentation. As a result, the proposed data augmentation algorithm showed the higher speech enhancement performance than the other algorithms.

• 한국멀티미디어학회논문지
• /
• 제21권12호
• /
• pp.1387-1395
• /
• 2018

Core algorithm of deep learning Convolutional Neural Network(CNN) shows better performance than other machine learning algorithms. However, if there is not sufficient data, CNN can not achieve satisfactory performance even if the classifier is excellent. In this situation, it has been proven that the use of transfer learning can have a great effect. In this paper, we apply two transition learning methods(freezing, retraining) to three CNN models(ResNet-50, Inception-V3, DenseNet-121) and compare and analyze how the classification performance of CNN changes according to the methods. As a result of statistical significance test using various evaluation indicators, ResNet-50, Inception-V3, and DenseNet-121 differed by 1.18 times, 1.09 times, and 1.17 times, respectively. Based on this, we concluded that the retraining method may be more effective than the freezing method in case of transition learning in image classification problem.

• Journal of Multimedia Information System
• /
• 제6권2호
• /
• pp.75-80
• /
• 2019

In machine learning, the performance of the system depends upon the nature of input data. The efficiency of the system improves when the behavior of the input data changes from un-normalized to normalized form. This paper experimentally demonstrated the performance of KNN, SVM, LDA and NB on Alzheimer's dataset. The dataset undertaken for the study consisted of 3 classes, i.e. Demented, Converted and Non-Demented. Analysis shows that LDA and NB gave an accuracy of 89.83% and 88.19% respectively in both the cases whereas the accuracy of KNN and SVM improved from 46.87% to 82.80% and 53.40% to 88.75% respectively when input data changed from un-normalized to normalized state. From the above results it was observed that KNN and SVM show significant improvement in classification accuracy on normalized data as compared to un-normalized data, whereas LDA and NB reflect no such change in their performance.

• 한국멀티미디어학회논문지
• /
• 제22권1호
• /
• pp.27-34
• /
• 2019

In this paper, a new type of lip feature is proposed as distance metric in CIELUV color system. The performance of the proposed feature was tested on face image database, Helen dataset from University of Illinois. The test processes consists of three steps. The first step is feature extraction and second step is principal component analysis for the optimal projection of a feature vector. The final step is Otsu's threshold for a two-class problem. The performance of the proposed feature was better than conventional features. Performance metrics for the evaluation are OverLap and Segmentation Error. Best performance for the proposed feature was OverLap of 65% and 59 % of segmentation error. Conventional methods shows 80~95% for OverLap and 5~15% of segmentation error usually. In conventional cases, the face database is well calibrated and adjusted with the same background and illumination for the scene. The Helen dataset used in this paper is not calibrated or adjusted at all. These images are gathered from internet and therefore, there are no calibration and adjustment.

• 한국멀티미디어학회논문지
• /
• 제18권11호
• /
• pp.1367-1374
• /
• 2015

In this paper, we propose a model which can prioritize the performance improvement work by analyzing the major risks and their influence, which can cause performance degradation in the system and show an example of a performance improvement using this model. In presentation-tier, as a result of log data analysis before and after the performance improvement of key processes which handle financial transactions, this model brought the CPU utilization and memory enhancement in the performance improvement work of the financial system which was carried out by applying the proposed model. It has been confirmed that the entire end-user can be accommodated. In the web-tier, the available memory increased by 200MB and we were able to improve the server restart(Recycling) that was sustained in the existing system. In the business logic-tier, we have been able to see better figures after performance improvements through the graph which analyzes the log collected with the key performance counters such as CPU%, Batch Requests/sec. In the data-tier, it has been confirmed that CPU usage and standby operation were reduced and the throughput was found to increase.

• 정보와 통신
• /
• 제11권6호
• /
• pp.37-54
• /
• 1994

멀티미디어 기술의 획기적 발전으로 이를 응용한 다양한 서비스가 개발되고 있다. 본 논문에서는 멀티미디어 응용 서비스를 TV, 컴퓨터, 통신, 주변기기, 소프트웨어 계열로 분류하고 설명하였다. 이중 컴퓨터 부분에서 멀티미디어 서비스를 중점적으로 설명하며 이러한 서비스가 제공되기 위한 시스템의 주요 서브시스템으로서 멀티미디어 데이터 처리를 위한 초고속 통신망, 운영체제, 지능형 에이젼트에 대한 논의를 하였다. 이중 특히 멀티미디어 운영체제 기술은 종합 멀티미디어 서비스 제공을 위한 하부 구조로서 최근 가장 활발히 발전되고 잇는 분야이다. 이에 본 논문에서는 일반적인 멀티미디어 운영체제의 발전방향 및 현재 개발 완료된 COSMOS 멀티미디어 그룹표현용 운영체제에 관한 설명을 한다. COSMOS의 특성, 모델, 추상적 데이터 구조에 관한 논의와 더불어 COSMOS상에서 개발된 3자간 영상회의시스템의 성능분석을 통하여 COSMOS와 같은 운영체제를 통한 멀티미디어 운영체제의 제공이 멀티미디어 서비스 환경변화를 주도함을 설명하였다.

• 융합신호처리학회논문지
• /
• 제9권3호
• /
• pp.213-218
• /
• 2008

본 논문은 양방향 위성 멀티미디어 통신시스템의 멀티미디어 STB (Multimedia Set-Top-Box)을 위한 하드웨어 기반의 고속 멀티미디어데이터 재조합 프로세서 설계 및 구현에 관한 것이다. 기존의 위성 멀티미디어 STB에서는 수신된 멀티미디어 데이터 재조합 기능을 STB의 CPU 소프트웨어 기반으로 처리하였다. 광대역 멀티미디어 서비스가 증대됨에 따라 STB의 CPU 부하가 증대되어 수신되는 멀티미디어 데이터 재조합 처리 성능이 제한되는 현상이 발생한다. 이러한 문제점을 해결하여 다양한 광대역 멀티미디어 서비스를 원활하게 처리할 수 있는 하드웨어 기반의 고속 멀티미디어 데이터 재조합 프로세서를 제안하였다. 구현된 멀티미디어 데이터 재조합 프로세서는 상용 FPGA, PCI 인터페이스 칩, 램 메모리 등으로 구현되었으며 위성 멀티미디어 시스템의 멀티미디어 STB에 실장하여 기능과 성능을 검증하였다. 제시된 요구기능을 모두 만족하였으며 최대 116 Mbps 처리 성능과 실용성을 확인하였다.

• 한국통신학회논문지
• /
• 제18권7호
• /
• pp.980-988
• /
• 1993

본 논문에서는 멀티미디어 트래픽올 갖는 B-ISBN의 스위치/멀티플렉서 등에서 중요한 파라메타로 결정되는 버퍼 크기에 대한 트래픽의 성능을 분석하였다. 멀티미디어 트래픽은 포아송 분포 트래픽, 그리고 기하분포와 지수 분포의 ON 시간 주기 (버스트 분포)를 갖는 버 스트 트래픽이 혼합되는 트래픽으로 모델링 하였다. 멀티미디어 트래픽으로 모델링한 트래픽에 대해서 성능을 분석하고 양질의 서비스를 제공할 수 있는 셀손실률에 대한 버퍼의 크기를 제시 하였다. 시뮬레이션은 PC SIMSCRIPT II.5를 이용하였고 시뮬레이션 프로그램은 PREAMBLE, MAIN, INITIAL, ARRIVAL., DEPARTURE및 STOP·SIM의 모듈로 구성하였다. 특히 혼합된 트래픽의 시뮬레이션에서는 ARRIVAL모듈을 I, II로 나누고 셀을 각각의 모듈에서 독립적으로 발생시켜 시뮬레이션을 수행하였다. 시뮬레이션 모델링은 이벤트 스케줄링 (Event Scheduling) 방식을 적용하여 시행하였다.

• International Journal of Contents
• /
• 제8권2호
• /
• pp.36-42
• /
• 2012

This study was conducted to develop Web-based multimedia content that assists undergraduate students in a clinical practicum on adult nursing. The study examined whether students in the intervention group could obtain clinical knowledge and perform more effectively when encouraged to learn Web content as compared with students in the conventional group. Web-based multimedia content consisting of 13 learning modules was developed based on real patients' scenarios through collaboration among college professors. A total of 120 nursing students (74 for the intervention and 46 for the conventional groups) from two universities in G-city, who engaged in a 3-week long clinical practicum in the digestive and respiratory units of a university hospital, participated in the study. Students' knowledge, self-directed learning, and clinical performance ability were measured using self-administered questionnaires. Data for pre- and posttests were collected over a 2-month period, between May and June of 2009. Clinical knowledge and self-reported clinical performance scores were significantly higher in students using the Web-enhanced clinical practicum than in those in the conventional group. However, there was no significant difference in self-directed learning ability between the 2 groups. These results demonstrate that Web-based multimedia content can be an effective educational tool for enhancing students' clinical knowledge and performance.