• The Journal of the Acoustical Society of Korea
• /
• v.37 no.6
• /
• pp.475-482
• /
• 2018

In this paper, we present a study on data augmentation methods for DNN (Deep Neural Network)-based audio tagging. In this system, an audio signal is converted into a mel-spectrogram and used as an input to the DNN for audio tagging. To cope with the problem associated with a small number of training data, we augment the training samples using time stretching, pitch shifting, dynamic range compression, and block mixing. In this paper, we derive optimal parameters and combinations for the augmentation methods through audio tagging simulations.

• Proceedings of the KIEE Conference
• /
• summer
• /
• pp.403-405
• /
• 2004

A service restoration is one of the most important missions in distribution system operation. This paper proposes a multi-agent system approach to distribution system restoration. Every relay is developed as an agent by adding its own intelligent, self-tuning and communication ability. Relay agent calculates and corrects its restoration index by itself through communication with neighboring agents and its own intelligence. The proposed algorithm is applied to a simple network to show how to calculate restoration index. Keywords Multi-Agent System, Service Restoration, Distribution Networks

• International Journal of Precision Engineering and Manufacturing
• /
• v.7 no.1
• /
• pp.35-41
• /
• 2006

In this paper, we develop anti-sway control in proposed techniques for an ATC system. The developed algorithm is to build the optimal path of container motion and to calculate an anti-collision path for collision avoidance in its movement to the finial coordinate. Moreover, in order to show the effectiveness in this research, we compared NNP PID controller to be tuning parameters of controller using NN with 2-DOF PID controller. The experimental results jar an ATC simulator show that the proposed control scheme guarantees performances, trolley position, sway angle, and settling time in NNP PID controller than other controller. As a result, the application of NNP PID controller is analyzed to have robustness about disturbance which is wind of fixed pattern in the yard.

• Proceedings of the IEEK Conference
• /
• 2003.07d
• /
• pp.1239-1242
• /
• 2003

The neural networks are a famous model to learn the nonlinear function or nonlinear system. The main point of neural network is that the difference actual output from desired output is used to update weights. Usually, the gradient descent method is used for the learning process. On training process, if learning rate is too large, neural networks hardly guarantee convergence of neural networks. On the other hand, if learning rate is too small, the training spends much time. Therefore, one major problem in use of neural networks are to decrease the teaming time while neural networks are guaranteed convergence. In this paper, we suggest the model of fuzzy logic to neural networks to calibrate learning rate. This method is to tune learning rate dynamically according to error and demonstrates the optimization of training.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.549-550
• /
• 2020

최근 등장한 자율주행, 스마트팩토리 및 IoT 등 다양한 기술은 기존 4G 네트워크를 활용하기에 부적합한 사항이 많았다. 따라서 5G 네트워크가 등장하였으며 네트워크 슬라이싱 기술을 통해 다양한 서비스에 각각의 네트워크 환경을 구성하여 제공하였다. 그러나 같은 네트워크 환경인 슬라이스 내에서도 특징이 다른 트래픽이 발생할 수 있으며 서비스의 종류로 고정된 슬라이스의 네트워크 환경은 트래픽 처리시간 증가 및 응답시간 증가를 유발할 수 있다. 따라서 본 논문에서는 트래픽의 특성을 고려하여 클러스터링을 시행하여 자동으로 네트워크 슬라이스를 관리하는 시스템을 제안한다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.10a
• /
• pp.324-329
• /
• 1998

In this paper, neural networks predictive PID (NNPPID) control system is proposed to reduce the vibration of structure. NNPPID control system is made up predictor, controller, and self-tuner to yield the optimal parameters of controller. The neural networks predictor forecasts the future outputs based on present input and output of structure. The controller is PID type whose parameters are yielded by neural networks self tuning algorithm. Computer simulations show displacements of multi-story structures applied to NNPPID system about environmental load-wind forces and earthquakes.

• KIISE Transactions on Computing Practices
• /
• v.24 no.1
• /
• pp.53-57
• /
• 2018

A poisoning accident is often caused by a situation in which people eat poisonous mushrooms because they cannot distinguish between edible mushrooms and poisonous mushrooms. In this paper, we propose an automatic mushroom recognition system by using the convolutional neural network. We collected 1478 mushroom images of 38 species using image crawling, and used the dataset for learning the convolutional neural network. A comparison experiment using AlexNet, VGGNet, and GoogLeNet was performed using the collected datasets, and a comparison experiment using a class number expansion and a fine-tuning technique for transfer learning were performed. As a result of our experiment, we achieve 82.63% top-1 accuracy and 96.84% top-5 accuracy on test set of our dataset.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.404-407
• /
• 2009

Bluetooth Video Distribution Profile (VDP) defines the protocol and procedures that realize the distribution of video content compressed in a specific format for the efficient use of the limited bandwidth. In this paper, we describe the design of VDP tester based on TTCN-2 (Tree and Tabular Combined Notation), a language standardized by ISO for the specification of tests for real-time and communicating systems. Our work was carried out as a part of supporting a new profile testing module for VDP in PTS (Profile Tuning Suite), a reference test system for Bluetooth interoperability testing. Test demonstration for the interoperability with various VDP solutions at the PTS session in UPF30 (Unplug Fest) showed the validity of the developed tester. Eventually, we introduce the PTS architecture, and show the design and implementation of VDP tester included in the released PTS 3.0 in this paper.

• ETRI Journal
• /
• v.34 no.5
• /
• pp.674-683
• /
• 2012

We propose a slot antenna consisting of a rectangular slot on the ground plane, fed by a microstrip line with a rectangular-ring-shaped tuning stub that can be deployed in ultra-wideband (UWB) communication systems to avoid interference with wireless local area network (WLAN) communication. Our antenna can achieve a single band-notched property from the 5 GHz frequency to the 6 GHz frequency owing to a controllable band notch that uses L- and J-shaped parasitic elements. The antenna characteristics can be modified to tune the band-notched property (4 GHz to 5 GHz or 6 GHz to 7 GHz) and the bandwidth of the band notch (1 GHz to 2 GHz). Furthermore, the shifted notch with enhanced width of the band notch from 1 GHz to 1.5 GHz is described in this paper. The UWB slot antenna and L- and J-shaped parasitic elements also provide the band-rejection function for reference in the WiMAX (3.5 GHz) and WLAN (5 GHz to 6 GHz) regions of the spectrum. Experiment results evidence the return loss performance, radiation patterns, and antenna gains at different operational frequencies.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2000.05a
• /
• pp.378-385
• /
• 2000

In this paper, we consider to apply of 2-DOF (Degree of Freedom) PID controller at D.C servo motor system. Many control system use I-PD , PID control system. but the position control system have difficulty in controling variable load and changing parameter. We propose neural network 2-DOF PID control system having feature for removal disturbrances and tracking function in the target value point. The back propagation algorithm of neural network used for tuning the 2-DOF parameter(${\alpha}$,${\beta}$,${\gamma}$,η). We investigate the 2-DOF PID control system in the position control system and verify the effectiveness of proposal method through the result of computer simulation.