• ETRI Journal
• /
• v.36 no.6
• /
• pp.960-967
• /
• 2014

In this paper, an analytical framework is proposed for the optimization of network performance through joint congestion control, channel allocation, rate allocation, power control, scheduling, and routing with the consideration of fairness in multi-channel wireless multihop networks. More specifically, the framework models the network by a generalized network utility maximization (NUM) problem under an elastic link data rate and power constraints. Using the dual decomposition technique, the NUM problem is decomposed into four subproblems - flow control; next-hop routing; rate allocation and scheduling; power control; and channel allocation - and finally solved by a low-complexity distributed method. Simulation results show that the proposed distributed algorithm significantly improves the network throughput and energy efficiency compared with previous algorithms.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.537-539
• /
• 2019

구문 분석은 문장의 구조를 분석하는 자연어처리 분야로 그래프 기반 방법과 전이 기반 방법으로 나뉘어 연구되어 왔다. 본 논문에서는 그래프 기반 방식에서 높은 성능을 보이고 있는 Deep Biaffine 어텐션 모델에 별도의 High-Order 정보 추출 없이 Graph Neural Network(GNNs)을 이용하여 High-Order 정보를 학습할 수 있도록 확장한 Deep Biaffine 어텐션 GNNs을 적용하여 한국어 세종 구문 분석 셋에서 UAS : 94.44%, LAS : 92.55%의 성능을 달성하였으며 Dual Decomposition을 통해 전이 기반 한국어 구문 분석 모델과 결합하여 추가적인 성능 향상을 보였다.

• Journal of radiological science and technology
• /
• v.40 no.2
• /
• pp.253-259
• /
• 2017

The study examines changes in calcium volume on born by comparing two figures; one is measured by dual energy computed tomography(DECT) followed by applying variation in monochromatic energy selection(keV), material decomposition(MD), and material suppressed iodine(MSI) analysis, and the other is measured by conventional single source computed tomography(CSCT). For this study, based on CSCT images taken by using human mimicked phantom, 70, 100, 140 keV and MSI, MD material calcium weighting(MCW) and MD material iodine weighting(MIW) of DECT were applied respectively. Then calculated calcium volume was converted to Agatston score for comparison. Volume of human mimicked phantom was in inverse proportion to keV. The volume decreased while keV increased(p<0.05). The most similar DECT volumes were reconstructed at 70 keV, the difference was showed $35.8{\pm}12.2$ for rib, femur ($16.1{\pm}24.1$), pelvis($13.7{\pm}18.8$), and spine($179.0{\pm}61.8$). However, the volume of MSI was down for each organ; the volume of rib was 5.55%, femur(76.34%), pelvis(55.16%) and spine(87.58%). The volume of MSI decreased 55.9% for rib, femur(80.7%), pelvis(69.6%) and spine(54.2%) while MD MIW reduced for rib(83.51%), femur(87.68%), pelvis(86.64%), and spine(82.62%). With the results, the study found that outcomes were affected by the method which examiners employed. When using DECT, calcium volume of born dropped with keV increased. It also found that the most similar DECT images were reconstructed at 70 keV. The results of experiments implied that the users of MSI and MD should be cautious of errors as there are big differences in scores between those two methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.10
• /
• pp.1517-1524
• /
• 2022

In this paper, we consider a heterogeneous network (HetNet) consisting of one macro base station and multiple small base stations, and assume the coordinated multi-point transmission between the base stations. In addition, we assume that the channel between the base station and the user consists of path loss and Rayleigh fading. Under these assumptions, we present the energy efficiency (EE) achievable by the user for a given base station and we formulate an optimization problem of dynamic cell selection and transmit power allocation to maximize the total EE of the HetNet. In this paper, we propose an unsupervised deep learning method to solve the optimization problem. The proposed deep learning-based scheme can provide high EE while having low complexity compared to the conventional iterative convergence methods. Through the simulation, we show that the proposed dynamic cell selection scheme provides higher EE performance than the maximum signal-to-interference-plus-noise ratio scheme and the Lagrangian dual decomposition scheme, and the proposed transmit power allocation scheme provides the similar performance to the trust region interior point method which can achieve the maximum EE.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2006.05a
• /
• pp.347-350
• /
• 2006

The rocket grade hydrogen peroxide has been widely used as a monopropellant in propulsion systems. In the present paper, we described an experimental study of a catalytic reactor that employs two stage catalyst beds to enhance the low temperature performance of the reactor. $K_2MnO_4$ was chosen as the catalyst for the initial stage of the reactor bed for its superior behavior in the low temperature regime. LSC was used for the catalyst of the second stage of the reactor. The gas generator with combined catalyst beds was built and tested to exhibit high decomposition efficiency over 90% and successful cold-start characteristics.

• Journal of Korean Society for Atmospheric Environment
• /
• v.33 no.6
• /
• pp.605-615
• /
• 2017

Efficient simultaneous reduction conditions for $NO_x$ and $NH_3$-slip was investigated in SCR (Selective Catalytic Reduction) process with load variation by applying dual catalysts (SCR catalyst, $NH_3$ decomposition catalyst) system. $N_2O$ formation characteristics were analyzed to look into possible undesirable reaction pathways. In the experiments of catalyst characteristics, various operational variables were tested for the combined catalytic system, such as $NH_3/NO_x$ ratio, temperature, oxygen concentration and $H_2O$. The reaction characteristics of $NO_x$, $NH_3$ and $N_2O$ were analyzed and optimal conditions could be evaluated for the combustion facility with varied load. In terms of $NO_x/NH_3$ simultaneous reduction and $N_2O$ formation suppression, optimal condition was considered NSR 1.2 and temperature $300^{\circ}C$. At this operational condition, $NO_x$ conversion was 98%, $NH_3$ reduction efficiency was 95%, generated $N_2O$ concentration 9.5 ppm with inlet $NO_x$ concentration of 100 ppm. In $NH_3-SCR$ process with $NH_3$ decomposition catalyst, $NO_x$ and $NH_3$ can be considered to be reduced simultaneously at limited conditions. The results of this study may be utilized as basic data at facilities requiring simultaneous $NO_x$ and $NH_3$ reduction for facilities with load variation.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.3
• /
• pp.87-92
• /
• 2006

The rocket grade hydrogen peroxide has been widely used as a monopropellant in propulsion systems. In the present paper, we described an experimental study of a catalytic reactor that employs two stage catalyst beds to enhance the low temperature performance of the reactor inlet. $K_2MnO_4$ was chosen as the catalyst for the initial stage of the reactor bed for its superior behavior in the low temperature regime. Alumina sol-gel method was successfully applied for coating $K_2MnO_4$ on a reactor bed of cordierite monolith. LSC was used for the catalyst of the second stage of the reactor. The reactor with combined catalyst beds was built and tested to exhibit superior performance in low temperature regime and high decomposition efficiency.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_2
• /
• pp.737-746
• /
• 2022

Utilizing weather independent SAR images along with machine learning based object detector is effective in robust vessel monitoring. While conventional SAR images often applied amplitude data from Single Look Complex, exploitation of polarimetric parameters acquired from multiple polarimetric SAR images was yet to be implemented to vessel detection utilizing machine learning. Hence, this study used four polarimetric parameters (H, p₁, DoP, DPRVI) retrieved from eigen-decomposition and two backscattering coefficients (γ₀, _VV, γ₀, _VH) from radiometric calibration; six bands in total were respectively exploited from 52 Sentinel-1 SAR images, accompanied by vessel training data extracted from AIS information which corresponds to acquisition time span of the SAR image. Evaluating different cases of combination, the use of polarimetric indexes along with amplitude values derived enhanced vessel detection performances than that of utilizing amplitude values exclusively.

• Journal of Satellite, Information and Communications
• /
• v.11 no.2
• /
• pp.55-59
• /
• 2016

Conventional RF energy harvesting systems can harvest energy and decode information from same source as an Hybirid-AP (H-AP). However, harvesting efficiency is seriously dependent on distance between users and H-AP. Therefore, in this paper, we proposed a transmission model for RF harvesting consisting of information and power source separately called Decoupled RF Energy harvesting networks. Main purpose of this paper is to maximize energy efficiency under various constraints of transmit power from H-AP and power beacon (PB), minimum quality of service and quality of harvested power of each users. To measure proposed model's performance, we proposed optimal time scheduling algorithms for energy efficiency (EE) maximization using Lagrangian dual decomposition theory that locally maximizes the EE by obtaining suboptimal values of three arguments : transmit power of H-AP, transmit power of PB, frame splitting factor. Experiment results show that the proposed energy-efficient algorithms converge within a few iterations with its optimality and greatly improve the EE compared to that of baseline schemes.

• Journal of Korean Society of Environmental Engineers
• /
• v.27 no.7
• /
• pp.712-718
• /
• 2005

Stepwise reductions of glucose and Orange II concentration were observed from the experiment of both white-rot fungi such as T. versicolor and P. chrysosporium. As a result, typical removal patterns in those dual substrate system were categorized through several distinctive steps: initial lag period, primary and secondary carbon consumption periods. Also, based on the total removal amounts of Orange II, COD and Color during the experimental period, similar removal extent were observed from both species experiments, within the maximal error range of 5%. However, it was refereed that the internal steps of Orange II removal on enzymatic level should be different between two species: Enzyme Lac showed good affinity for Orange II removal in T. versicolor, however in P. chrysosporium enzyme LiP represented more close affinity to the similar experimental condition. Thus, even though the superficial removal amount of calcitrant Orange II at different fungal species was merely similar, removal pathway of enzymatic levels and intermediates produced during the fungal decomposition would be different.