• Smart Structures and Systems
• /
• v.30 no.4
• /
• pp.421-432
• /
• 2022

Regarding the high significance of correct pan evaporation modeling, this study introduces two novel neuro-metaheuristic approaches to improve the accuracy of prediction for this parameter. Vortex search algorithms (VSA), sunflower optimization (SFO), and stochastic fractal search (SFS) are integrated with a multilayer perceptron neural network to create the VSA-MLPNN, SFO-MLPNN, and SFS-MLPNN hybrids. The climate data of Arcata-Eureka station (operated by the US environmental protection agency) belonging to the years 1986-1989 and the year 1990 are used for training and testing the models, respectively. Trying different configurations revealed that the best performance of the VSA, SFO, and SFS is obtained for the population size of 400, 300, and 100, respectively. The results were compared with a conventionally trained MLPNN to examine the effect of the metaheuristic algorithms. Overall, all four models presented a very reliable simulation. However, the SFS-MLPNN (mean absolute error, MAE = 0.0997 and Pearson correlation coefficient, R_P = 0.9957) was the most accurate model, followed by the VSA-MLPNN (MAE = 0.1058 and R_P = 0.9945), conventional MLPNN (MAE = 0.1062 and R_P = 0.9944), and SFO-MLPNN (MAE = 0.1305 and R_P = 0.9914). The findings indicated that employing the VSA and SFS results in improving the accuracy of the neural network in the prediction of pan evaporation. Hence, the suggested models are recommended for future practical applications.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.1
• /
• pp.24-31
• /
• 2020

In this paper, the deep convolutional neural network for sound event detection is studied. Especially, the end-to-end neural network, which generates the detection results from the input audio waveform, is studied for weakly supervised problem that includes weakly-labeled and unlabeled dataset. The proposed system is based on the network structure that consists of deeply-stacked 1-dimensional convolutional neural networks, and enhanced by the skip connection and gating mechanism. Additionally, the proposed system is enhanced by the sound event detection and post processings, and the training step using the mean-teacher model is added to deal with the weakly supervised data. The proposed system was evaluated by the Detection and Classification of Acoustic Scenes and Events (DCASE) 2019 Task 4 dataset, and the result shows that the proposed system has F₁-scores of 54 % (segment-based) and 32 % (event-based).

• Journal of Sensor Science and Technology
• /
• v.30 no.3
• /
• pp.175-180
• /
• 2021

In this paper, methods for improving the sensitivity of gas sensors to NO₂ gas are presented. A gas sensor was fabricated based on an SnO₂ nanowire network using the vapor-phase-growth method. In the gas sensor, the Au electrode was replaced with a fluorinedoped tin oxide (FTO) electrode, to achieve high sensitivity at low temperatures and concentrations. The gas sensor with the FTO electrode was more sensitive to NO₂ gas than the sensor with the Au electrode: notably, both sensors were based on typical SnO₂ nanowire network. When the Au electrode was replaced by the FTO electrode, the sensitivity improved, as the contact resistance decreased and the surface-to-volume ratio increased. The morphological features of the fabricated gas sensor were characterized in detail via field-emission scanning electron microscopy and X-ray diffraction analysis.

• Smart Media Journal
• /
• v.4 no.3
• /
• pp.9-15
• /
• 2015

A pub(lish)/sub(scribe) system is a data forwarding system which forwards only interesting data among the whole published data, which is related to the subscriptions registered by end users. Classical pub/sub systems are realized by constructing a network of brokers which are responsible for storing or forwarding data. Along with the substantial increase of the population mobile users, it is required that the pub/sub system handles the subscriptions of user locations which changes continuously and frequently. In this paper, a new broker network based pub/sub system which efficiently handles the frequent changes of subscriptions related to user locations is proposed. In consideration of moving patterns of users and geographical property, the proposed pub/sub system categorize the entire data space into Slow Moving Region and Normal Moving Region, and manages the brokers which are responsible for these regions by using Q+R tree in order to handle user requests more efficiently. Through the extensive simulation, it is presented that the proposed Q+R tree based pub/sub system can reduce unnecessary needs of brokers and network traffic and can support the dynamic subscription related to user location.

• Geomechanics and Engineering
• /
• v.31 no.3
• /
• pp.291-304
• /
• 2022

Maximum shear modulus (G_max or G₀) is an important soil property useful for many engineering applications, such as the analysis of soil-structure interactions, soil stability, liquefaction evaluation, ground deformation and performance of seismic design. In the current study, bender element (BE) tests are used to evaluate the effect of the void ratio, effective confining pressure, grading characteristics (D₅₀, C_u and C_c), anisotropic consolidation and initial fabric anisotropy produced during specimen preparation on the G_max of sand-gravel mixtures. Based on the tests results, an empirical equation is proposed to predict G_max in granular soils, evaluated by the experimental data. The artificial neural network (ANN) and Adaptive Neuro Fuzzy Inference System (ANFIS) models were also applied. Coefficient of determination (R²) and Root Mean Square Error (RMSE) between predicted and measured values of G_max were calculated for the empirical equation, ANN and ANFIS. The results indicate that all methods accuracy is high; however, ANFIS achieves the highest accuracy amongst the presented methods.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.33 no.5
• /
• pp.181-186
• /
• 2023

In this study, novel Eu₂O₃-BaF₂-La₂O₃-B₂O₃ oxyfluoride glasses and glass ceramics were developed by a containerless processing. Differential thermal analysis (DTA) analysis was performed to analyze the thermophysical properties of oxyfluoride glasses doped with Eu₂O₃, and photoluminescence (PL) characteristics were analyzed to evaluate the luminous efficiency depending on the degree of crystallinity. The glass transition temperature decreased with increasing BaF₂ concentration since BaF₂ acts as a network modifier in this glass system. In addition, thermal stability which can be estimated by the difference between the glass transition temperature and the onset temperature of the crystallization decreased with increasing BaF₂ contents. The peak related to the BaF₂ crystal was confirmed after the crystallization by X-ray Diffraction (XRD) analysis. Photoluminescence intensity increased after the crystallization which indicates that the Eu³⁺ ions are sited in BaF₂ crystal. La 3d_5/2 x-ray photoelectron spectroscopy (XPS) and F1s XPS spectra were analyzed to precisely understand the behavior of the fluorine ion in the glass structure. Fluorine tends to bond with the network modifying cations such as La³⁺ and Ba²⁺ ions and after the crystallization the La-F bonds decreased because F^- ions used to form BaF₂ crystals.

• ETRI Journal
• /
• v.43 no.5
• /
• pp.799-811
• /
• 2021

This paper considers a challenging problem: to simultaneously optimize the cost and the quality of service in opaque wavelength division multiplexing (WDM) networks. An optimization problem is proposed that takes the information including network topology, traffic between end nodes, and the target level of congestion at each link/ node in WDM networks. The outputs of this problem include routing, link channel capacities, and the optimum number of switch ports locally added/dropped at all switch nodes. The total network cost is reduced to maintain a minimum congestion level on all links, which provides an efficient trade-off solution for the network design problem. The optimal information is utilized for dynamic traffic in WDM networks, which is shown to achieve the desired performance with the guaranteed quality of service in different networks. It was found that for an average link blocking probability equal to 0.015, the proposed model achieves a net channel gain in terms of wavelength channels (𝛾_w) equal to 35.72 %, 39.09 %, and 36.93 % compared to shortest path first routing and 𝛾_w equal to 29.41 %, 37.35 %, and 27.47 % compared to alternate routing in three different networks.

• Korean Journal of Materials Research
• /
• v.20 no.4
• /
• pp.223-227
• /
• 2010

Recently, one-dimensional semiconducting nanomaterials have attracted considerable interest for their potential as building blocks for fabricating various nanodevices. Among these semiconducting nanomaterials,, $SnO_2$ nanostructures including nanowires, nanorods, nanobelts, and nanotubes were successfully synthesized and their electrochemical properties were evaluated. Although $SnO_2$ nanowires and nanobelts exhibit fascinating gas sensing characteristics, there are still significant difficulties in using them for device applications. The crucial problem is the alignment of the nanowires. Each nanowire should be attached on each die using arduous e-beam or photolithography, which is quite an undesirable process in terms of mass production in the current semiconductor industry. In this study, a simple process for making sensitive $SnO_2$ nanowire-based gas sensors by using a standard semiconducting fabrication process was studied. The nanowires were aligned in-situ during nanowire synthesis by thermal CVD process and a nanowire network structure between the electrodes was obtained. The $SnO_2$ nanowire network was floated upon the Si substrate by separating an Au catalyst between the electrodes. As the electric current is transported along the networks of the nanowires, not along the surface layer on the substrate, the gas sensitivities could be maximized in this networked and floated structure. By varying the nanowire density and the distance between the electrodes, several types of nanowire network were fabricated. The $NO_2$ gas sensitivity was 30~200 when the $NO_2$ concentration was 5~20ppm. The response time was ca. 30~110 sec.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.6
• /
• pp.187-194
• /
• 2020

This paper proposes a website fingerprinting method using ensemble learning over a Tor network that guarantees client anonymity and personal information. We construct a training problem for website fingerprinting from the traffic packets collected in the Tor network, and compare the performance of the website fingerprinting system using tree-based ensemble models. A training feature vector is prepared from the general information, burst, cell sequence length, and cell order that are extracted from the traffic sequence, and the features of each website are represented with a fixed length. For experimental evaluation, we define four learning problems (Wang14, BW, CW_T, CW_H) according to the use of website fingerprinting, and compare the performance with the support vector machine model using CUMUL feature vectors. In the experimental evaluation, the proposed statistical-based training feature representation is superior to the CUMUL feature representation except for the BW case.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.8
• /
• pp.3841-3861
• /
• 2017

Virtualized small cell network is a promising architecture which can realize efficient utilization of the network resource. However, conventional full duplex self-backhauls lead to residual self-interference, which limits the network performance. To handle this issue, this paper proposes a virtual resource allocation, in which the residual self-interference is fully exploited by employing a physical-layer network coding (PNC) aided self-backhaul scheme. We formulate the features of PNC as time slot and information rate constraints, and based on that, the virtual resource allocation is formulated as a mixed combinatorial optimization problem. To solve the problem efficiently, it is decomposed into two sub problems, and a two-phase iteration algorithm is developed accordingly. In the algorithm, the first sub problem is approximated and transferred into a convex problem by utilizing the upper bound of the PNC rate constraint. On the basis of that, the convexity of the second sub problem is also proved. Simulation results show the advantages of the proposed scheme over conventional solution in both the profits of self-backhauls and utility of the network resource.