• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10b
• /
• pp.340-344
• /
• 2007

유비쿼터스 지능공간에서 사람의 행동과 위치를 모니터링하는 것은 서비스 제공을 위해 기본적이며 필수적인 단계이다. 본 논문에서는 스마트 펜던트(SmartPendant)의 카메라 영상과 GPS위치 정보를 이용한 새로운 웨어러블 컴퓨터를 제안한다. 우선, 행동 인식를 위해 영상간에 특정 픽셀 값 차와 옵티컬 플로우를 사용하였으며, 인식이 가능한 행동으로는 걷기, 멈춤, 방향전환이다. 또한, GPS를 이용한 사용자의 위치 정보는 위도와 경도에 대한 스트링값을 패킷값으로 변환하여 지능형 상황인지 서버에 전달된다.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.4
• /
• pp.487-494
• /
• 2023

In this study, performance comparison was performed on R and TensorFlow, which are free deep learning tools. In the experiment, six types of deep neural networks were built using each tool, and the neural networks were trained using the 10-year Korean temperature dataset. The number of nodes in the input layer of the constructed neural network was set to 10, the number of output layers was set to 5, and the hidden layer was set to 5, 10, and 20 to conduct experiments. The dataset includes 3600 temperature data collected from Gangnam-gu, Seoul from March 1, 2013 to March 29, 2023. For performance comparison, the future temperature was predicted for 5 days using the trained neural network, and the root mean square error (RMSE) value was measured using the predicted value and the actual value. Experiment results shows that when there was one hidden layer, the learning error of R was 0.04731176, and TensorFlow was measured at 0.06677193, and when there were two hidden layers, R was measured at 0.04782134 and TensorFlow was measured at 0.05799060. Overall, R was measured to have better performance. We tried to solve the difficulties in tool selection by providing quantitative performance information on the two tools to users who are new to machine learning.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.01a
• /
• pp.255-256
• /
• 2019

본 논문에서는 물체를 집기(picking) 위해 필요한 깊이 값을 특수카메라인 리얼센스를 사용하여 받아와서 2D 카메라로는 하지 못하는 로봇 팔 피킹 시스템을 구현하였다. 객체 인식은 텐서플로우 객체 검출 라이브러리를 사용하여 정확도를 높였고, ROS 기반의 rviz, moveit, gazebo 등의 패키지를 사용하여 아두이노와 통신하며 로봇팔 하드웨어로 인식된 객체를 피킹하는 시스템을 구현하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.219-220
• /
• 2021

본 논문에서는 마이크로컨트롤러에서 6축 IMU 센서를 사용한 제스쳐를 인식하기 위한 최적화된 학습 방법을 제안한다. 6축 센서값을 119번 샘플링할 경우 특징 차원이 매우 크기 때문에 다층 신경망을 이용할 경우 학습파라미터가 마이크로컨트롤러의 메모리 허용량을 초과하게 된다. 본 논문은 성능은 유지하며 학습 파라미터 개수를 효과적으로 줄이기 위한 마이크로컨트롤러에 최적화된 1D CNN을 제안한다.

• KIPS Transactions on Computer and Communication Systems
• /
• v.8 no.4
• /
• pp.79-86
• /
• 2019

This paper describes a time-series data prediction based on artificial neural networks (ANN). In this study, a batch based ANN model and a stochastic ANN model have been implemented using TensorFlow libraries. Each model are evaluated by comparing training and testing errors that are measured through experiment. To train and test each model, tax dataset was used that are collected from the government website of indiana state budget agency in USA from 2001 to 2018. The dataset includes tax incomes of individual, product sales, company, and total tax incomes. The experimental results show that batch model reveals better performance than stochastic model. Using the batch scheme, we have conducted a prediction experiment. In the experiment, total taxes are predicted during next seven months, and compared with actual collected total taxes. The results shows that predicted data are almost same with the actual data.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.4
• /
• pp.687-697
• /
• 2023

Botnets, whose attack patterns are becoming more sophisticated and diverse, are recognized as one of the most serious cybersecurity threats today. This paper revisits the experimental results of botnet detection using autoencoder, a semi-supervised deep learning model, for UGR and CTU-13 data sets. To prepare the input vectors of autoencoder, we create data points by grouping the NetFlow records into sliding windows based on source IP address and aggregating them to form features. In particular, we discover a simple power-law; that is the number of data points that have some flow-degree is proportional to the number of NetFlow records aggregated in them. Moreover, we show that our power-law fits the real data very well resulting in correlation coefficients of 97% or higher. We also show that this power-law has an impact on the learning of autoencoder and, as a result, influences the performance of botnet detection. Furthermore, we evaluate the performance of autoencoder using the area under the Receiver Operating Characteristic (ROC) curve.

• Journal of KIISE:Information Networking
• /
• v.28 no.4
• /
• pp.477-488
• /
• 2001

In wireless networking environments, supporting guaranteed quality of service to mobile hosts is difficult due to the facts that wireless networks have limited bandwidth and mobile hosts frequently move in and out of cells. In spite of the characteristics of wireless communications, the quality of some types of services, i.e., real-time services, must be guaranteed at a certain level. When a mobile host moves into another cell, service rates for mobile hosts in wireless networks may be adjusted since wireless networks have limited bandwidths. In this paper, we propose two resource allocation algorithms in wireless mobile networks, using quality of service (QoS) specifications. For efficient use of resources of wireless networks, the proposed algorithms dynamically allocate rates of flows in proportion to QoS with limited resources.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.9
• /
• pp.782-788
• /
• 2014

In this paper, we propose a scheme for the support of flow mobility according to a QoS condition when a mobile node with multiple wireless interface moves and attaches to a different access network. For this purpose, we extend the fields of the PBU/PBA message and add an additional request procedures. Through the proposed scheme, we can provide the best service in according to the network conditions of the next access network based on the measured QoS values by LMA. Finally we consider the state transition diagram of our proposed scheme and confirm its operation.

• Journal of Digital Contents Society
• /
• v.16 no.3
• /
• pp.381-388
• /
• 2015

TCP, which performs congestion control in congestion condition, is able to help a reliable transmission. However, packet loss can be increased because congestion window is increased by the time the packet is dropped in the process of congestion avoidance. In this paper, to solve the above problem, we propose a new congestion estimation based TCP congestion control scheme using the weighted average value of the RTT. After measuring a SRTT, which means the weighted average value of RTTs, at this point of time when a buffer overflow is occurred by an overloaded packet, the proposed scheme estimates the time, when the same SRTT is made in packet transmission, as a congestion time and then decreases the congestion window. The simulation results show that the proposed schem has a good performance in terms of packet loss rate and throughput when the packet loss due to buffer overflow is larger than that due to wireless channel.

• Resources Recycling
• /
• v.22 no.6
• /
• pp.12-18
• /
• 2013

In this Study, it was fabricated that shrinkage reducing agent and mortar used $C_{12}A_7$-based slag enhanced the shrinkage reduction and compressive strength. To reduce cement content, setting time, flow and compressive strength of mortar with varying content of fly ash and blast furnace slag were experimented. The flow increased and setting time delayed as the increase of fly ash and blast furnace slag content. And early strength was lower and long age strength was higher than that of mortar with low content of admixture.