• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.6
• /
• pp.2480-2496
• /
• 2020

Considering that high-dose X-ray radiation during CT scans may bring potential risks to patients, in the medical imaging industry there has been increasing emphasis on low-dose CT. Due to complex statistical characteristics of noise found in low-dose CT images, many traditional methods are difficult to preserve structural details effectively while suppressing noise and artifacts. Inspired by the deep learning techniques, we propose a densely connected residual network (DCRN) for low-dose CT image noise cancelation, which combines the ideas of dense connection with residual learning. On one hand, dense connection maximizes information flow between layers in the network, which is beneficial to maintain structural details when denoising images. On the other hand, residual learning paired with batch normalization would allow for decreased training speed and better noise reduction performance in images. The experiments are performed on the 100 CT images selected from a public medical dataset-TCIA(The Cancer Imaging Archive). Compared with the other three competitive denoising algorithms, both subjective visual effect and objective evaluation indexes which include PSNR, RMSE, MAE and SSIM show that the proposed network can improve LDCT images quality more effectively while maintaining a low computational cost. In the objective evaluation indexes, the highest PSNR 33.67, RMSE 5.659, MAE 1.965 and SSIM 0.9434 are achieved by the proposed method. Especially for RMSE, compare with the best performing algorithm in the comparison algorithms, the proposed network increases it by 7 percentage points.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.447-450
• /
• 2019

본 논문은 한국어 문장의 감성 분류를 위해 문장의 형태소, 음절, 자소를 입력으로 하는 합성곱층과 DenseNet 을 적용한 Text Multi-channel DenseNet 모델을 제안한다. 맞춤법 오류, 음소나 음절의 축약과 탈락, 은어나 비속어의 남용, 의태어 사용 등 문법적 규칙에 어긋나는 다양한 표현으로 인해 단어 기반 CNN 으로 추출 할 수 없는 특징들을 음절이나 자소에서 추출 할 수 있다. 한국어 감성분석에 형태소 기반 CNN 이 많이 쓰이고 있으나, 본 논문에서 제안한 Text Multi-channel DenseNet 모델은 형태소, 음절, 자소를 동시에 고려하고, DenseNet 에 정보를 밀집 전달하여 문장의 감성 분류의 정확도를 개선하였다. 네이버 영화 리뷰 데이터를 대상으로 실험한 결과 제안 모델은 85.96%의 정확도를 보여 Multi-channel CNN 에 비해 1.45% 더 정확하게 문장의 감성을 분류하였다.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.2
• /
• pp.101-111
• /
• 2011

A new MAC protocol, MT-MAC(Multi-Tier Medium Access Control) by name, is proposed for dense sensor networks. Depending on the density of nodes in a virtual cluster, the cluster header performs the splitting to several tiers in nodes of virtual cluster. MT-MAC split the tiers to use modfied-SYNC message after receiving the beacon message from the cluster header. Because only the sensor nodes in the same tier communicate each other, less power is consumed and longer network life time is guaranteed. By a simulation method with NS-2, we evaluated our protocol. In dense nodes environments, MT-MAC protocol shows better results than S-MAC in terms of packet delivery rates throughput and energy consumption.

• Journal of the Korean Society for information Management
• /
• v.40 no.2
• /
• pp.59-80
• /
• 2023

Traditional models for categorizing researcher types have mostly utilized research output metrics. This study proposes a new model that classifies researchers based on the characteristics of research collaboration. The model uses only research collaboration indicators and does not rely on citation data, taking into account that citation impact is related to collaborative research. The model categorizes researchers into four types based on their collaborative research pattern and scope: Sparse & Wide (SW) type, Dense & Wide (DW) type, Dense & Narrow (DN) type, Sparse & Narrow (SN) type. When applied to the quantum metrology field, the proposed model was statistically verified to show differences in citation indicators and co-author network indicators according to the classified researcher types. The proposed researcher type classification model does not require citation information. Therefore, it is expected to be widely used in research management policies and research support services.

• Journal of Internet Computing and Services
• /
• v.21 no.2
• /
• pp.121-130
• /
• 2020

This paper suggests evaluating the military closed network data as an image which is generated by Generative Adversarial Network (GAN), applying an image evaluation method such as the InceptionV3 model-based Inception Score (IS) and Frechet Inception Distance (FID). We employed the famous image classification models instead of the InceptionV3, added layers to those models, and converted the network data to an image in diverse ways. Experimental results show that the Densenet121 model with one added Dense Layer achieves the best performance in data converted using the arctangent algorithm and 8 * 8 size of the image.

• Journal of Communications and Networks
• /
• v.18 no.2
• /
• pp.135-149
• /
• 2016

Cloud radio access network (C-RAN) refers to the virtualization of base station functionalities by means of cloud computing. This results in a novel cellular architecture in which low-cost wireless access points, known as radio units or remote radio heads, are centrally managed by a reconfigurable centralized "cloud", or central, unit. C-RAN allows operators to reduce the capital and operating expenses needed to deploy and maintain dense heterogeneous networks. This critical advantage, along with spectral efficiency, statistical multiplexing and load balancing gains, make C-RAN well positioned to be one of the key technologies in the development of 5G systems. In this paper, a succinct overview is presented regarding the state of the art on the research on C-RAN with emphasis on fronthaul compression, baseband processing, medium access control, resource allocation, system-level considerations and standardization efforts.

• Journal of Korea Multimedia Society
• /
• v.21 no.12
• /
• 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.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.9
• /
• pp.2967-2981
• /
• 2014

Device-to-Device (D2D) communication is a technology component for long-term evolution-advanced (LTE-A). In D2D communication, users in close proximity to each other can communicate directly without going through a base station; such direct communication can improve spectral efficiency. Although D2D communication brings improvement in spectral efficiency, it also causes interference to the cellular network as a result of spectrum sharing. In particularly, D2D communication can generate interference for each D2D pair when the common wireless medium in a co-located limited area is accessed. Even though the interference management for between the D2D pair and cellular networks has been proposed, the interference reducing methods have still not been fully studied for the D2D pairs. In this paper, we investigate the problem of D2D pair coexistence in which interference is considered between D2D pairs. Using a signal to interference model for a target D2D pair, we provide an analysis of the aggregated throughput of a dense D2D network. For a target D2D pair, we assume that the desired signal and interference signals obey multipath fading and shadow fading. Through analysis, we demonstrate the effect of cluster size such as the number of D2D pairs and the size of the considered area on the network performance. The analytical results are compared with computer simulations. Our work can be used for a rough guideline for controlling the system throughput in a dense D2D network environment.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.3
• /
• pp.143-148
• /
• 2022

In this paper, we employ deep neural network (DNN) to discern Android user interaction data distribution from artificial data distribution. We utilize real Android user interaction trace dataset collected from [1] to evaluate our DNN design. In particular, we use sequential model with 4 dense hidden layers and 1 dense output layer in TensorFlow and Keras. We also deploy sigmoid activation function for a dense output layer with 1 neuron and ReLU activation function for each dense hidden layer with 32 neurons. Our evaluation shows that our DNN design fulfills high test accuracy of at least 0.9955 and low test loss of at most 0.0116 in all cases of artificial data distributions.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.05a
• /
• pp.612-615
• /
• 2012

Energy efficiency in wireless sensor network is a principal issue because wireless sensor network uses limited energy. In wireless sensor network, because nodes are placed randomly, they may be concentrated in certain area. This dense area causes shortening the life of the concentrated area, and furthermore reducing the life of the entire network. In this paper, we suggest a additional cluster head selection algorithm for reducing the overload of head node in dense area and shows simulation result using our algorithm with LEACH algorithm.