• The Journal of the Acoustical Society of Korea
• /
• v.41 no.1
• /
• pp.30-37
• /
• 2022

This paper compares and evaluates model performance from two perspectives according to the learning target and network structure for training Deep Neural Network (DNN)-based speech enhancement models in the frequency domain. In this case, spectrum mapping and Time-Frequency (T-F) masking techniques were used as learning targets, and a real network and a complex network were used for the network structure. The performance of the speech enhancement model was evaluated through two objective evaluation metrics: Perceptual Evaluation of Speech Quality (PESQ) and Short-Time Objective Intelligibility (STOI) depending on the scale of the dataset. Test results show the appropriate size of the training data differs depending on the type of networks and the type of dataset. In addition, they show that, in some cases, using a real network may be a more realistic solution if the number of total parameters is considered because the real network shows relatively higher performance than the complex network depending on the size of the data and the learning target.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.24 no.4
• /
• pp.392-400
• /
• 2021

Unlike optical equipment, SAR(Synthetic Aperture Radar) has the advantage of obtaining images in all weather, and object detection in SAR images is an important issue. Generally, deep learning-based object detection was mainly performed in real-valued network using only amplitude of SAR image. Since the SAR image is complex data consist of amplitude and phase data, a complex-valued network is required. In this paper, a complex-valued ResNet network is proposed. SAR image object detection was performed by combining the ROI transformer detector specialized for aerial image detection and the proposed complex-valued ResNet. It was confirmed that higher accuracy was obtained in complex-valued network than in existing real-valued network.

• 한국신재생에너지학회:학술대회논문집
• /
• 2007.11a
• /
• pp.213-216
• /
• 2007

PEMFC의 전기화학적 반응은 촉매, 이오노머, 기공이 만나는 삼상계면에서만 일어나므로, 전극 구조의 최적화가 성능 향상 및 장기안정성 확보에 있어 매우 중요하다. 본 연구에서는 전극 미세구조를 실시간으로 분석하기 위해 임피던스 복소캐패시턴스법을 도입하고자 하였다. 즉, PEMFC의 양극에 질소를 공급하면 0.4 V 부근에서 전기이중층 형성 반응만이 일어나는 것을 확인하였으며, 이때 음극에는 수소를 공급하여 기준전극 및 반대전극으로 사용하였다. 측정된 임피던스를 복소캐패시턴스로 변환하고 허수부를 주파수에 대해 도시하면 피크 형태의 곡선이 얻어지는데, (1) 피크 면적은 전극/전해질의 계면면적, (2) 피크 위치는 이오노머 네트워크에 의한 수소이온 전도 특성, (3) 피크 폭은 다공성 구조의 균일도를 각각 나타내므로, 피팅 없이 직접적인 해석이 가능하다는 장점을 가진다. 반면, 기존의 Nyquist 도시법은 피팅에 의한 분석이 필요하며, 전극층의 불균일한 구조로 인해 단순한 등가회로 구성이 어려운 문제점을 가진다. 최종적으로, MEA 제작 조건 및 운전 조건을 변수로 하여 임피던스를 측정하고 복소캐패시턴스 분석을 수행하여, 퇴화 경로를 규명하고 운전 조건을 최적화하고자 하였다.

• Composites Research
• /
• v.27 no.5
• /
• pp.190-195
• /
• 2014

This paper predicts the complex permittivity of MWNT added epoxy depending on agglomeration by numerical analysis. 1wt% MWNT added epoxy specimen is prepared using 3-roll-mill method and its complex permittivity is measured in X-band (8.2~12.4 GHz) using freespace measurement system. The analytic model is comprised of cube epoxy and perfect sphere agglomeration. The complex permittivity of the agglomeration model is predicted by complex permittivity mixing rule using the measured complex permittivity of epoxy and 1 wt% MWNT added epoxy. Commercial electromagnetic analysis software, CST, is used to obtain S-parameter of the analytic model and MATLAB code is used to calculate complex permittivity from the S-parameter. It is confirmed that the complex permittivity increases when the agglomeration size decreases.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.4
• /
• pp.119-125
• /
• 2015

In this paper, we introduce a Hadamard matrix interstream transmission based blind channel estimation for multi-cells multiple-input and multiple-output (MIMO) networks. The proposed scheme is based on a network with mobile stations (MS) which are deployed with multi cells. We assume that the MS have the signals from both cells. The signal from near cell are considered as desired signal and the signals from the other cells are interference signal. Since the channel is blind, so that we transmit Hadamard matrix pattern pilot stream to estimate the channel; that gives easier and fast channel estimation for large scale MIMO channel. The computation of Hadamard based system takes only complex additions, and thus the complexity of which is much lower than the scheme with Fourier transform since complex multiplications are not needed. The numerical analysis will give perfection of proposed channel estimation.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.858-859
• /
• 2022

본 논문에서는 프린지 패턴을 생성하는 딥러닝 기반의 WGAN-GP 네트워크의 최적화 방법을 제안한다. 기존의 복소 프린지 패턴 생성을 위한 GAN 모델은 생성의 정확도뿐만 아니라 학습의 안정성이 다소 부족하였다. 이에 따라 WGAN-GP 등의 업그레이드 된 방법을 사용하였지만, 네트워크 구조 및 파라미터에 따른 최적화가 필요하다. 보다 정확도 높은 정확도를 가진 프린지 패턴 생성을 위해 learning rate decay 사용하여 학습된 결과를 epoch 별 그래프로 최적화 전의 결과와 비교하고, 홀로그램과 복원 결과에 대한 PSNR 을 비교한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.05a
• /
• pp.117-120
• /
• 2008

In this paper, we developed the EM wave absorber for eliminating false images in collision-avoidance radar. First of all, we fabricated some samples in different composition ratio of $TiO_2$ and CPE. And the relative permittivities of samples are calculated from S-parameter of samples by using 1-port method. We designed and fabricated the EM wave absorber by using the calculated relative permittivity. As a result, the EM wave absorber with composition of $TiO_2$:CPE=70:30 wt% has thickness of 1.85 mm and absorption ability higher than 20 dB in the frequency range 76-77 GHz.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.176-182
• /
• 2021

This paper evaluates and compares the performance of the Deep Nerual Network (DNN)-based speech enhancement models according to various loss functions. We used a complex network that can consider the phase information of speech as a baseline model. As the loss function, we consider two types of basic loss functions; the Mean Squared Error (MSE) and the Scale-Invariant Source-to-Noise Ratio (SI-SNR), and two types of perceptual-based loss functions, including the Perceptual Metric for Speech Quality Evaluation (PMSQE) and the Log Mel Spectra (LMS). The performance comparison was performed through objective evaluation and listening tests with outputs obtained using various combinations of the loss functions. Test results show that when a perceptual-based loss function was combined with MSE or SI-SNR, the overall performance is improved, and the perceptual-based loss functions, even exhibiting lower objective scores showed better performance in the listening test.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.2A
• /
• pp.188-194
• /
• 2007

In order to become ubiquitous world, the compatibility of wireless machines has become the significant characteristic of a communication terminal. Thus, SDR is the most necessary technology and standard. However, among the environment which has different communication protocol, it's difficult to make a terminal with only hardware using ASIC or SoC. This paper suggests the processor that can accelerate several communication protocol. It can be connected with main-processor, and it is specialized PHY layer of network The C-program that is modeled with the wireless protocol IEEE802.11a and IEEE802.11b which are based on widely used modulation way; OFDM and CDM is compiled with ARM cross compiler and done simulation and profiling with Simplescalar-Arm version. The result of profiling, most operations were Viterbi operations and complex floating point operations. According to this result we suggested a co-processor which can accelerate Viterbi operations and complex floating point operations and added instructions. These instructions are simulated with Simplescalar-Arm version. The result of this simulation, comparing with computing only one ARM core, the operations of Viterbi improved as fast as 4.5 times. And the operations of complex floating point improved as fast as twice. The operations of IEEE802.11a are 3 times faster, and the operations of IEEE802.11b are 1.5 times faster.

• Polymer(Korea)
• /
• v.35 no.5
• /
• pp.451-457
• /
• 2011

The effect of latex particle size on rheological and electrical properties of latex-blended polystyrene (PS)/multi-walled carbon nanotube (MWCNT) nanocomposites was investigated. Mono-dispersed PS particles synthesized either by emulsifier-free emulsion polymerization or by dispersion polymerization were mixed with MWCNTs under ultrasonication, and freeze-dried to prepare the nanocomposites. As the MWCNT content increased, storage modulus, complex viscosity and electrical conductivity were substantially increased. The increase of storage modulus and complex viscosity was higher for larger PS particles. The effect of particle size on electrical properties was different depending on MWCNT content. With lower MWCNT content, the nanocomposite prepared by smaller PS particles showed higher electrical conductivity, but the opposite result was given as the content increased.