• Journal of Korea Multimedia Society
• /
• v.21 no.3
• /
• pp.391-399
• /
• 2018

In this paper, we construct and compare various guitar chord classification systems using perceptron neural network and convolutional neural network without pre-processing other than Fourier transform to identify the optimal chord classification system. Conventional guitar chord classification schemes use, for better feature extraction, computationally demanding pre-processing techniques such as stochastic analysis employing a hidden markov model or an acoustic data filtering and hence are burdensome for real-time chord classifications. For this reason, we construct various perceptron neural networks and convolutional neural networks that use only Fourier tranform for data pre-processing and compare them with dataset obtained by playing an electric guitar. According to our comparison, convolutional neural networks provide optimal performance considering both chord classification acurracy and fast processing time. In particular, convolutional neural networks exhibit robust performance even when only small fraction of low frequency components of the data are used.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.8B
• /
• pp.1150-1158
• /
• 2010

At present multimedia service composed of voice, data, and video service is prevalent in the Internet. As such, wide-scale penetration of Internet service imposes tremendous pressure to the network infrastructure such as the media servers and links as well as the nodes. In addition, users from a large-scale population require broad bandwidth and high level of QoS. This requires a network with reliable and scalable services to customers, which also necessitates a realistic method for the design of the a media server. In this work, we explore the influence of user's behavior about delay on the performance of media server that takes into account the system and user attributes in a realistic manner. By incorporating user's behavior about the delay-sensitivity, we present an analytic framework for the evaluation of the performance of the media server, via which we illustrate a meaningful intuition in the provision of Internet multimedia service.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.2 no.2
• /
• pp.129-135
• /
• 2007

In this paper, we present hybrid ARQ techniques to improve the throughput performance in high speed packet transmission such as Internet or multimedia services in Hospital. In order to evaluate the performance of the three different types of hybrid ARQ schemes, systemized link level simulations based on WWW traffic model are considered. In this paper, we also consider the simulated performance for an average link throughput and a normalized packet delay to compare the hybrid ARQ with link adaptation scheme in GPRS.

• Journal of Korea Multimedia Society
• /
• v.11 no.6
• /
• pp.828-834
• /
• 2008

This paper presents high-spread (HS) random interleavers based interleave-division multiple access (IDMA) scheme for the 4th generation mobile radio system. High-spread feature of a random interleaver offers high-spread characteristics to interleavers and thus produces extrinsic values with low correlation. As interleavers are key components for user separation in the IDMA, the HS interleavers with this feature are employed and analyzed in the IDMA for performance improvement. In addition, by replacing random interleavers with the HS interleavers in the IDMA, bandwidth efficiency is achieved by means of reducing the length of the repetition code. Performance evaluation shows that the proposed scheme of the IDMA with the HS interleavers provides superior BER performance and improved bandwidth efficiency.

• Journal of Korea Multimedia Society
• /
• v.24 no.9
• /
• pp.1211-1223
• /
• 2021

Internet of Things(IoT) services have increased the demand for connectivity among electronics devices. As a result, various types of novel wireless communication technologies have been standardized and developed. In this paper, we evaluate the performance of low power wireless communication technologies such as Bluetooth, IEEE 802.15.4, DASH 7, IEEE 802.15.4g, LoRa, and SigFox in various environments. This is the first experiment evaluating various low power wireless communication technologies in real testbed. We expect that the evaluation results will be useful data to other researchers in applying the IoT technology in the future.

• Journal of Multimedia Information System
• /
• v.5 no.4
• /
• pp.273-276
• /
• 2018

In video compression standards, the entropy coding is essential to the high performance compression because redundancy of data symbols is removed. Binary arithmetic coding is one of high performance entropy coding methods. However, the dependency between consecutive binary symbols prevents improving the throughput. For the throughput enhancement, a new probability model is proposed for encoding multi-symbols at one time. In the proposed method, multi-symbol encoder is implemented with only adders and shifters, and the multiplication table for interval subdivision of binary arithmetic coding is removed. Compared to the compression ratio of CABAC of H.264/AVC, the performance degradation on average is only 1.4% which is negligible.

• Journal of Integrative Natural Science
• /
• v.11 no.4
• /
• pp.204-208
• /
• 2018

Comparison of different optimizer performance in photovoltaic power modeling using artificial neural deep learning techniques is described in this paper. Six different deep learning optimizers are tested for Long-Short-Term Memory networks in this study. The optimizers are namely Adam, Stochastic Gradient Descent, Root Mean Square Propagation, Adaptive Gradient, and some variants such as Adamax and Nadam. For comparing the optimization techniques, high and low fluctuated photovoltaic power output are examined and the power output is real data obtained from the site at Mokpo university. Using Python Keras version, we have developed the prediction program for the performance evaluation of the optimizations. The prediction error results of each optimizer in both high and low power cases shows that the Adam has better performance compared to the other optimizers.

• Journal of Multimedia Information System
• /
• v.5 no.4
• /
• pp.291-297
• /
• 2018

Performance measurement is not an easy task. However, it is necessary to measure the expected value of the informatization project to be invested, to help the successful progress of the informatization project, and to confirm the result of the informatization project. This study describes the evaluation systems of the defense informatization project in Korea used after the end of the informatization project (post-implementation evaluation stage). In the post-implementation evaluation stage of proposed system, the informatization project is evaluated in terms of performance achievement, project plan compliance, project management, and economic validity. For each element, this study suggests metrics and practical measurement method. With metrics and their method, it can be evaluated which project is better than others. Moreover, the lessons learned can be developed. Under scarce budget condition, defense organizations can make better performance from IT project with the evaluation system suggested in this study.

• Journal of Korea Multimedia Society
• /
• v.22 no.6
• /
• pp.665-675
• /
• 2019

In this paper, we compare and analyze the classification performance of deep learning algorithm Convolutional Neural Network(CNN) ac cording to ensemble generation and combining techniques. We used several CNN models(VGG16, VGG19, DenseNet121, DenseNet169, DenseNet201, ResNet18, ResNet34, ResNet50, ResNet101, ResNet152, GoogLeNet) to create 10 ensemble generation combinations and applied 6 combine techniques(average, weighted average, maximum, minimum, median, product) to the optimal combination. Experimental results, DenseNet169-VGG16-GoogLeNet combination in ensemble generation, and the product rule in ensemble combination showed the best performance. Based on this, it was concluded that ensemble in different models of high benchmarking scores is another way to get good results.

• Journal of Korea Multimedia Society
• /
• v.24 no.1
• /
• pp.51-57
• /
• 2021

In this paper, we propose a task distribution scheme in fog computing environment considering opportunistic fog computing nodes. As latency is one of the important performance metric for IoT(Internet of Things) applications, there have been lots of researches on the fog computing system. However, since the load can be concentrated to the specific fog computing nodes due to the spatial and temporal IoT characteristics, the load distribution should be considered to prevent the performance degradation. Therefore, this paper proposes a task distribution scheme which considers the static as well as opportunistic fog computing nodes according to their mobility feature. Especially, based on the task requirements, the proposed scheme supports the delay sensitive task processing at the static fog node and delay in-sensitive tasks by means of the opportunistic fog nodes for the task distribution. Based on the performance evaluation, the proposed scheme shows low service response time compared to the conventional schemes.