• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2286-2304
• /
• 2022

Blockchain is a distributed ledger that combines technologies such as cryptography, consensus mechanism, peer-to-peer transmission, and time stamping. The rapid development of blockchain has attracted attention from all walks of life, but storage scalability issues have hindered the application of blockchain. In this paper, a scalable blockchain storage model based on Distributed Hash Table (DHT) and the InterPlanetary File System (IPFS) was proposed. This paper introduces the current research status of the scalable blockchain storage model, as well as the basic principles of DHT and the InterPlanetary File System. The model construction and workflow are explained in detail. At the same time, the DHT network construction mechanism, block heat identification mechanism, new node initialization mechanism, and block data read and write mechanism in the model are described in detail. Experimental results show that this model can reduce the storage burden of nodes, and at the same time, the blockchain network can accommodate more local blocks under the same block height.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1295-1301
• /
• 2019

In this paper, we propose a novel method of denoising human motion using a bidirectional recurrent neural network (BRNN) with an attention mechanism. The corrupted motion captured from a single 3D depth sensor camera is automatically fixed in the well-established smooth motion manifold. Incorporating an attention mechanism into BRNN achieves better optimization results and higher accuracy than other deep learning frameworks because a higher weight value is selectively given to a more important input pose at a specific frame for encoding the input motion. Experimental results show that our approach effectively handles various types of motion and noise, and we believe that our method can sufficiently be used in motion capture applications as a post-processing step after capturing human motion.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.345-356
• /
• 2023

Target signal of passive sonar shows narrow band harmonic characteristic with a variation in intensity within a few seconds and long term frequency variation due to the Lloyd's mirror effect. We propose a signal classification algorithm based on Gated Recurrent Unit (GRU) that learns local and global time series features. The algorithm proposed implements a multi layer network using GRU and extracts local and global time series features via dilated connections. We learns attention mechanism to weight time series features and classify passive sonar signals. In experiments using public underwater acoustic data, the proposed network showed superior classification accuracy of 96.50 %. This result is 4.17 % higher classification accuracy compared to existing skip connected GRU network.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.11
• /
• pp.2924-2944
• /
• 2023

Alzheimer's disease (AD) is a neurological condition that is recognized as one of the primary causes of memory loss. AD currently has no cure. Therefore, the need to develop an efficient model with high precision for timely detection of the disease is very essential. When AD is detected early, treatment would be most likely successful. The most often utilized indicators for AD identification are the Mini-mental state examination (MMSE), and the clinical dementia. However, the use of these indicators as ground truth marking could be imprecise for AD detection. Researchers have proposed several computer-aided frameworks and lately, the supervised model is mostly used. In this study, we propose a novel 3D Convolutional Neural Network Multilayer Perceptron (3D CNN-MLP) based model for AD classification. The model uses Attention Mechanism to automatically extract relevant features from Magnetic Resonance Images (MRI) to generate probability maps which serves as input for the MLP classifier. Three MRI scan categories were considered, thus AD dementia patients, Mild Cognitive Impairment patients (MCI), and Normal Control (NC) or healthy patients. The performance of the model is assessed by comparing basic CNN, VGG16, DenseNet models, and other state of the art works. The models were adjusted to fit the 3D images before the comparison was done. Our model exhibited excellent classification performance, with an accuracy of 91.27% for AD and NC, 80.85% for MCI and NC, and 87.34% for AD and MCI.

• Smart Media Journal
• /
• v.12 no.11
• /
• pp.103-112
• /
• 2023

Recently, the need to prevent battery fires and accidents has emerged, as the use of lithium-ion batteries has increased. In order to prevent accidents, it is necessary to predict the state of health (SOH) and check the replacement timing of the battery with a lot of degradation. This paper proposes a model for predicting the degradation state of a battery by using four battery degradation indicators: maximum voltage arrival time, current change time, maximum temperature arrival time, and incremental capacity (IC) that can be obtained in the battery charging process, and LSTM using an attention mechanism. The performance of the proposed model was measured using the NASA battery data set, and the predictive performance was improved compared to that of the general LSTM model, especially in the SOH 90-70% section, which is close to the battery replacement cycle.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.544-551
• /
• 2023

Speech enhancement used to improve the perceptual quality and intelligibility of noise speech has been studied as a method using a complex-valued spectrum that can improve both magnitude and phase in a method using a magnitude spectrum. In this paper, a study was conducted on how to apply attention mechanism to complex-valued spectrum-based speech enhancement systems to further improve the intelligibility and quality of noise speech. The attention is performed based on additive attention and allows the attention weight to be calculated in consideration of the complex-valued spectrum. In addition, the global average pooling was used to consider the importance of the feature map. Complex-valued spectrum-based speech enhancement was performed based on the Deep Complex U-Net (DCUNET) model, and additive attention was conducted based on the proposed method in the Attention U-Net model. The results of the experiments on noise speech in a living room environment showed that the proposed method is improved performance over the baseline model according to evaluation metrics such as Source to Distortion Ratio (SDR), Perceptual Evaluation of Speech Quality (PESQ), and Short Time Object Intelligence (STOI), and consistently improved performance across various background noise environments and low Signal-to-Noise Ratio (SNR) conditions. Through this, the proposed speech enhancement system demonstrated its effectiveness in improving the intelligibility and quality of noisy speech.

• Proceeding of EDISON Challenge
• /
• 2016.03a
• /
• pp.455-462
• /
• 2016

Recently, followed by rapid growth of robotics field, multi-linkage mechanism which can even pass by rough road is getting lots of attention. In this paper, I focused on Jansen mechanism. It's a kinematics object which is named after Dutch artist Theo jansen. Jansen mechanism embraces structure and mechanism which creates locomotion with the combination of the power and simple structure. Theo jansen suggests a 'Holy number'. It's an ideal ratio of leg components length. However, if there's desired gait locomotion, you have to adjust the ratio and the length. But even slight change of the length could cause a big change at the end-point. To solve this problem, I suggest a reverse engineering method to get a ratio of each links by nonlinear optimization with pre-set desired trajectory. First, we converted a movement of the joint of Jansen mechanism to vectors by kinematics analysis of multi-linkage structure. And we showed the trajectory at the end-point. After that, we set desired trajectory which we found most ideal. Then we got the length of the leg components which draws a trajectory as same as trajectory we set, using Multi-objective genetic algorithm toolbox in MATLAB. Result is verified by Edison designer and mSketch. And we analyzed if it could pass through the obstruction which is set dynamically.

• Journal of Communications and Networks
• /
• v.6 no.1
• /
• pp.60-67
• /
• 2004

With the fast growth of Internet and a new widespread interest in optical networks, the unparalleled potential of Multi-Protocol Label Switching (MPLS) is leading to further research and development efforts. One of those areas of research is Path Protection Mechanism. It is widely accepted that layer three protection and recovery mechanisms are too slow for today’s reliability requirements. Failure recovery latencies ranging from several seconds to minutes, for layer three routing protocols, have been widely reported. For this reason, a recovery mechanism at the MPLS layer capable of recovering from failed paths in 10’s of milliseconds has been sought. In light of this, several MPLS based protection mechanisms have been proposed, such as end-to-end path protection and local repair mechanism. Those mechanisms are designed for intra-domain recoveries and little or no attention has been given to the case of non-homogenous independent inter-domains. This paper presents a novel solution for the setup and maintenance of independent protection mechanisms within individual domains and merged at the domain boundaries. This innovative solution offers significant advantages including fast recovery across multiple nonhomogeneous domains and high scalability. Detailed setup and operation procedures are described. Finally, simulation results using OPNET are presented showing recovery times of a few milliseconds.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.10
• /
• pp.4977-4996
• /
• 2016

The existence of excessively large and too filled network buffers, known as bufferbloat, has recently gained attention as a major performance problem for delay-sensitive applications. Researchers have made three types of suggestions to solve the bufferbloat problem. One is End to End (E2E) congestion control, second is deployment of Active Queue Management (AQM) techniques and third is the combination of above two. However, these solutions either seem impractical or could not obtain good bandwidth utilization. In this paper, we propose a Transmission Control Protocol（TCP）delayed window update mechanism which uses a congestion detection approach to predict the congestion level of networks. When detecting the network congestion is coming, a delayed window update control strategy is adopted to maintain good protocol performance. If the network is non-congested, the mechanism stops work and congestion window is updated based on the original protocol. The simulation experiments are conducted on both high bandwidth and long delay scenario and low bandwidth and short delay scenario. Experiment results show that TCP delayed window update mechanism can effectively improve the performance of the original protocol, decreasing packet losses and queuing delay while guaranteeing transmission efficiency of the whole network. In addition, it can perform good fairness and TCP friendliness.

• The Journal of the Korea Contents Association
• /
• v.7 no.2
• /
• pp.63-72
• /
• 2007

The traffic sign on the road should be easy to distinguishable even from far, and should be recognized in a short time. As traffic sign is a very important object which provides important information for the drivers to enhance safety, it has to attract human's attention among any other objects on the road. This paper proposes a new method of detecting the area of traffic sign, which uses attention module on the assumption that we attention our gaze on the traffic sign at first among other objects when we drive a car. In this paper, we analyze the previous studies of psycophysical and physiological results to get what kind of features are used in the process of human's object recognition, especially color processing, and with these results we detected the area of traffic sign. Various kinds of traffic sign images were tested, and the results showed good quality(average 97.8% success).