• 정보통신설비학회논문지
• /
• 제4권2호
• /
• pp.37-43
• /
• 2005

A simple scheduler satisfying the proportional fairness (PF) was introduced in wireless access networks and revealed that it can achieve a good compromise between total throughput and user fairness. Though it has received much attention for some time, its application was mainly restricted to the single channel systems. In this paper, we study how to implement the PF scheduler in the multi-channel environments such as OFDMA/TDMA. Besides the traditional PF-SC scheme, we propose a new PF-OPT scheme that is the genuine PF scheduler in a sense of maximizing the total log-utility of users. The simulation results show that PF-OPT gives large throughput under the heterogeneous subchannel statistics.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(1)
• /
• pp.217-220
• /
• 2001

Turbo codes of long block sizes have been known to show very good performance in an AWGN channel and the turbo code has been strongly recommended as error correction code for IMT-2000 in 3GPP(3$^{rd}$ Ceneration Partnership Project). Recently, turbo codes of short block sizes suitable for real time communication systems have attracted a lot of attention. Thus in this paper we consider the turbo code of 1/3 code rate and short frame size of 192 bits in ITU-R channel model. We analyzed the performance of W-CDMA systems of 10MHz bandwidths employing RAKE receiver with not only MRC diversity but also turbo code..

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권12호
• /
• pp.4439-4455
• /
• 2021

Compared with two-stage object detection algorithms, one-stage algorithms provide a better trade-off between real-time performance and accuracy. However, these methods treat the intermediate features equally, which lacks the flexibility to emphasize meaningful information for classification and location. Besides, they ignore the interaction of contextual information from different scales, which is important for medium and small objects detection. To tackle these problems, we propose an image pyramid network based on dual attention mechanism (DAIPNet), which builds an image pyramid to enrich the spatial information while emphasizing multi-scale informative features based on dual attention mechanisms for one-stage object detection. Our framework utilizes a pre-trained backbone as standard detection network, where the designed image pyramid network (IPN) is used as auxiliary network to provide complementary information. Here, the dual attention mechanism is composed of the adaptive feature fusion module (AFFM) and the progressive attention fusion module (PAFM). AFFM is designed to automatically pay attention to the feature maps with different importance from the backbone and auxiliary network, while PAFM is utilized to adaptively learn the channel attentive information in the context transfer process. Furthermore, in the IPN, we build an image pyramid to extract scale-wise features from downsampled images of different scales, where the features are further fused at different states to enrich scale-wise information and learn more comprehensive feature representations. Experimental results are shown on MS COCO dataset. Our proposed detector with a 300 × 300 input achieves superior performance of 32.6% mAP on the MS COCO test-dev compared with state-of-the-art methods.

• 한국진공학회:학술대회논문집
• /
• 한국진공학회 2014년도 제46회 동계 정기학술대회 초록집
• /
• pp.434.2-434.2
• /
• 2014

We observe enhanced pH response of solution-gated field-effect transistors (SG-FET) having 1D-2D hybrid channel of vertical grown ZnO nanorods grown on CVD graphene (Gr). In recent years, SG-FET based on Gr has received a lot of attention for biochemical sensing applications, because Gr has outstanding properties such as high sensitivity, low detection limit, label-free electrical detection, and so on. However, low-defect CVD Gr has hardly pH responsive due to lack of hydroxyl group on Gr surface. On the other hand, ZnO, consists of stable wurtzite structure, has attracted much interest due to its unique properties and wide range of applications in optoelectronics, biosensors, medical sciences, etc. Especially, ZnO were easily grown as vertical nanorods by hydrothermal method and ZnO nanostructures have higher sensitivity to environments than planar structures due to plentiful hydroxyl group on their surface. We prepared for ZnO nanorods vertically grown on CVD Gr (ZnO nanorods/Gr hybrid channel) and to fabricate SG-FET subsequently. We have analyzed hybrid channel FETs showing transfer characteristics similar to that of pristine Gr FETs and charge neutrality point (CNP) shifts along proton concentration in solution, which can determine pH level of solution. Hybrid channel SG-FET sensors led to increase in pH sensitivity up to 500%, compared to pristine Gr SG-FET sensors. We confirmed plentiful hydroxyl groups on ZnO nanorod surface interact with protons in solution, which causes shifts of CNP. The morphology and electrical characteristics of hybrid channel SG-FET were characterized by FE-SEM and semiconductor parameter analyzer, respectively. Sensitivity and sensing mechanism of ZnO nanorods/Gr hybrid channel FET will be discussed in detail.

• Journal of Information Processing Systems
• /
• 제19권4호
• /
• pp.483-497
• /
• 2023

Depth estimation is one of the most complicated and difficult problems to deal with in the light field. In this paper, a compound attention convolutional neural network (CAttNet) is proposed to extract depth maps from light field images. To make more effective use of the sub-aperture images (SAIs) of light field and reduce the redundancy in SAIs, we use a compound attention mechanism to weigh the channel and space of the feature map after extracting the primary features, so it can more efficiently select the required view and the important area within the view. We modified various layers of feature extraction to make it more efficient and useful to extract features without adding parameters. By exploring the characteristics of light field, we increased the network depth and optimized the network structure to reduce the adverse impact of this change. CAttNet can efficiently utilize different SAIs correlations and features to generate a high-quality light field depth map. The experimental results show that CAttNet has advantages in both accuracy and time.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권7호
• /
• pp.2131-2153
• /
• 2022

To achieve accurate detection of tuberculosis (TB) areas in chest radiographs, we design a chest X-ray TB area detection algorithm. The algorithm consists of two stages: the chest X-ray TB classification network (CXTCNet) and the chest X-ray TB area detection network (CXTDNet). CXTCNet is used to judge the presence or absence of TB areas in chest X-ray images, thereby excluding the influence of other lung diseases on the detection of TB areas. It can reduce false positives in the detection network and improve the accuracy of detection results. In CXTCNet, we propose a channel attention mechanism (CAM) module and combine it with DenseNet. This module enables the network to learn more spatial and channel features information about chest X-ray images, thereby improving network performance. CXTDNet is a design based on a sparse object detection algorithm (Sparse R-CNN). A group of fixed learnable proposal boxes and learnable proposal features are using for classification and location. The predictions of the algorithm are output directly without non-maximal suppression post-processing. Furthermore, we use CLAHE to reduce image noise and improve image quality for data preprocessing. Experiments on dataset TBX11K show that the accuracy of the proposed CXTCNet is up to 99.10%, which is better than most current TB classification algorithms. Finally, our proposed chest X-ray TB detection algorithm could achieve AP of 45.35% and AP50 of 74.20%. We also establish a chest X-ray TB dataset with 304 sheets. And experiments on this dataset showed that the accuracy of the diagnosis was comparable to that of radiologists. We hope that our proposed algorithm and established dataset will advance the field of TB detection.

• 한국통신학회논문지
• /
• 제19권8호
• /
• pp.1518-1527
• /
• 1994

본 논문에서는 배전선을 통하여 전송된 BPSK 신호를 마이크로프로세서를 이용하여 복조하는 알고리즘을 개발하고 구현하였다. 배전선 BPSK 복조를 마이크로프로세서를 실현하기 위해서는 무엇보다 배전선을 통하여 수신된 BPSK 신호를 2진 신호로 바꿔주어야 한다. 그러므로, 본 논문에서는 먼저 전송된 BPSK 신호를 2진 신호로 바꿔주는 하드웨어를 설계하였다. 그런 다음 반송파의 주파수가 피변조 2진 신호 주파수의 우수배이면 변조점이 피변조 2진 신호의 상승에지(rising edge)와 하강에지(falling edge)에서 각각 다른 방향으로 나타난다는 사실과 배전선으로부터 수신된 BPSK 신호를 여과하고, 증폭하고, 크리핑하고, 정형하는 과정을 조절하면 변조점에서의 2진 신호 길이가 다른 점에서의 2진 신호 길이의 2배로 된다는 것에 착안하여 배전선 BPSK 신호를 복조하는 알고리즘을 개발하고, 이 알고리즘에 의한 마이크로프로세서 복조시스템을 실제로 구현하였다. 구현된 이 복조시스템은 실제의 배전선 복조에서 비트오류율(bit error rate)이 0.02% 이하이었다.

• 스마트미디어저널
• /
• 제12권5호
• /
• pp.28-35
• /
• 2023

본 연구에서는 음성감정인식의 적용 가능성과 실용성 향상을 위해 적은 수의 파라미터를 가지는 새로운 경량화 모델 RoutingConvNet(Routing Convolutional Neural Network)을 제안한다. 제안모델은 학습 가능한 매개변수를 줄이기 위해 양방향 MFCC(Mel-Frequency Cepstral Coefficient)를 채널 단위로 연결해 장기간의 감정 의존성을 학습하고 상황 특징을 추출한다. 저수준 특징 추출을 위해 경량심층 CNN을 구성하고, 음성신호에서의 채널 및 공간 신호에 대한 정보 확보를 위해 셀프어텐션(Self-attention)을 사용한다. 또한, 정확도 향상을 위해 동적 라우팅을 적용해 특징의 변형에 강인한 모델을 구성하였다. 제안모델은 음성감정 데이터셋(EMO-DB, RAVDESS, IEMOCAP)의 전반적인 실험에서 매개변수 감소와 정확도 향상을 보여주며 약 156,000개의 매개변수로 각각 87.86%, 83.44%, 66.06%의 정확도를 달성하였다. 본 연구에서는 경량화 대비 성능 평가를 위한 매개변수의 수, 정확도간 trade-off를 계산하는 지표를 제안하였다.

• 한국유체기계학회 논문집
• /
• 제17권5호
• /
• pp.19-26
• /
• 2014

The cross flow turbine attracts more and more attention for its relatively wide operating range and simple structure. In this study, a novel type of micro cross flow turbine is developed for application to a step in an irrigational channel. The head of the turbine is only H=4.3m and the turbine inlet channel is open ducted type, which has barely been studied. The efficiency of the turbine with inlet open duct channel is relatively low. Therefore, a guide nozzle on the turbine inlet is attached to improve the performance of the turbine. The guide nozzle shapes are investigated to find the best shape for the turbine. The guide nozzle plays an important role on directing flow at the runner entry, and it also decreases the negative torque loss by reducing the pressure difference in Region 1. There is 12.5% of efficiency improvement by attaching a well shaped guide nozzle on the turbine inlet.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권10호
• /
• pp.3409-3422
• /
• 2014

The frequency spectrum available for the wireless communication is extremely crowded. In order to improve the spectral efficiency, the two-way relay networks have aroused great attention. A simple $N^{th}$ best-relay selection criterion for the opportunistic two-way relay networks is proposed, which can be implemented easily by extending the distributed timer technique in practice, since the proposed criterion is mainly based on the channel gains. The outage performance of the proposed relay selection scheme is analyzed under the outdated channel state information (CSI), and a tight closed-form lower bound and asymptotic value of the outage probability over Rayleigh fading channels are obtained. Simulation results demonstrate that the tight closed-form lower bound of the outage probability very closely matches with simulated ones in the whole SNR region, and the asymptotic results provide good tight approximations to the simulation ones, especially in the high SNR region.