• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.10
• /
• pp.3458-3478
• /
• 2022

Automatic modulation recognition is the core algorithm in the field of modulation classification in communication systems. Our investigations show that deep learning (DL) based modulation recognition techniques have achieved effective progress for multiple-input multiple-output (MIMO) systems. However, network complexity is always an additional burden for high-accuracy classifications, which makes it impractical. Therefore, in this paper, we propose a low-complexity dimensional interactive lightweight network (DilNet) for MIMO systems. Specifically, the signals received by different antennas are cooperatively input into the network, and the network calculation amount is reduced through the depth-wise separable convolution. A two-dimensional interactive attention (TDIA) module is designed to extract interactive information of different dimensions, and improve the effectiveness of the cooperation features. In addition, the TDIA module ensures low complexity through compressing the convolution dimension, and the computational burden after inserting TDIA is also acceptable. Finally, the network is trained with a penalized statistical entropy loss function. Simulation results show that compared to existing modulation recognition methods, the proposed DilNet dramatically reduces the model complexity. The dimensional interactive lightweight network trained by penalized statistical entropy also performs better for recognition accuracy in MIMO systems.

• International Journal of Safety
• /
• v.2 no.1
• /
• pp.17-22
• /
• 2003

In order to analyze general three dimensional cracks in an infinite body, the symmetric Galerkin boundary element method formulated by Li and Mear is used. A crack is modelled as distribution of displacement discontinuities, and the governing equation is formulated as singularity-reduced integral equations. With the proposed method several example problems for three dimensional cracks in an infinite solid, as well as their growth under fatigue, are solved and the accuracy and efficiency of the method are demonstrated.

• Proceedings of the KSR Conference
• /
• 2009.05a
• /
• pp.1083-1087
• /
• 2009

We present an 1-dimensional annular disk element with which natural vibration of a circular plate can be analyzed accurately and facilely. The natural vibration characteristics of a circular plate with free outer boundary are analyzed by using the presented I-dimensional element. Its results are compared with the results obtained by utilizing 2-dimensional 8-node quadrilateral plane element and cyclic symmetry of the circular plate. And also, by comparing with the theoretical results of previous researchers, the accuracy and facility of the presented I-dimensional element are verified.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.8
• /
• pp.713-718
• /
• 2012

Lumber spine range of motion has been used to measure of physical and functional impairment by various tools from a ruler to 3D kinematic devices. However, pre-existing tools have problems in either movement or accuracy and reliability limitations. Accurate devices are limited by fixed space whereas simple devices are limited in measuring complex movements with less accuracy. In order to solve the location, movement and accuracy limitations at once, we have developed a novice measurement device equipped with accelerometer sensor and gyroscope sensor for getting three-dimensional information of motion. Furthermore, Kalman filter was applied to the algorithm to improve accuracy. In addition, RF wireless communication was added for the user to conveniently check measured data in real time. Finally, the measurement method was improved by considering the movement by a reference point. An experiment was conducted to test the accuracy and reliability of the device by conducting a test-retest reliability test. Further modification will be conducted to used the device in various joints range of motion in clinical settings in the future.

• Journal of the Optical Society of Korea
• /
• v.15 no.1
• /
• pp.30-37
• /
• 2011

Binocular-pair images obtained from two cameras can be used to calculate the three-dimensional (3D) world coordinate of a feature point. However, to apply this method, measurement accuracy of binocular vision depends on some structure factors. This paper presents an experimental study of measurement distance, baseline distance, and baseline direction. Their effects on camera reconstruction accuracy are investigated. The testing set for the binocular model consists of a series of feature points in stereo-pair images and corresponding 3D world coordinates. This paper discusses a method to increase the baseline distance of two cameras for enhancing the accuracy of a binocular vision system. Moreover, there is an inflexion point of the value and distribution of measurement errors when the baseline distance is increased. The accuracy benefit from increasing the baseline distance is not obvious, since the baseline distance exceeds 1000 mm in this experiment. Furthermore, it is observed that the direction errors deduced from the set-up are lower when the main measurement direction is similar to the baseline direction.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.26-26
• /
• 2023

General circulation models (GCMs) are widely used in hydrological prediction, however their coarse grids make them unsuitable for regional analysis, therefore a downscaling method is required to utilize them in hydrological assessment. As one of the downscaling methods, convolutional neural network (CNN)-based downscaling has been proposed in recent years. The aim of this study is to generate the process of dynamic downscaling using CNNs by applying GCM output as input and RCM output as label data output. Prediction accuracy is compared between different input datasets, and model structures. Several input datasets with key atmospheric variables such as precipitation, temperature, and humidity were tested with two different formats; one is two-dimensional data and the other one is three-dimensional data. And in the model structure, the hyperparameters were tested to check the effect on model accuracy. The results of the experiments on the input dataset showed that the accuracy was higher for the input dataset without precipitation than with precipitation. The results of the experiments on the model structure showed that substantially increasing the number of convolutions resulted in higher accuracy, however increasing the size of the receptive field did not necessarily lead to higher accuracy. Though further investigation is required for the application, this paper can contribute to the development of efficient downscaling method with CNNs.

• Journal of radiological science and technology
• /
• v.38 no.3
• /
• pp.213-220
• /
• 2015

This research was study the accuracy of three-dimensional stereotactic breast biopsy, using a core Needle Biopsy and to assess the accuracy of Stereotactic biopsy and Sono guided biopsy. Using Stereotactic QC phantom to measure the accuracy of the 3D sterotactic machine. CT Scan and equipment obtained in the measured X, Y, Z and compares the accuracy of the length. Using Agar power phantom compare the accuracy of the 3D sterotactic machine and 2D ultrasound machine. Z axis measured by the equipment to compare the accuracy and reliability. Check the accuracy by using visual inspection and Specimen Medical application phantom. The accuracy of the 3D sterotactic machine measured by Stereotactic QC phantom was 100%. Accuracy as compared to CT, all of X, Y, Z axis is p > 0.05. The accuracy of the two devices was 100% as measured by Agar powder phantom. There was no difference between t he t wo d evices as C T and p > 0.05. 3D sterotactic machine of the ICC was 0.954, 2D ultrasound machine was 0.785. 2D ultrasound machine was different according to the inspector. Medical application phantom experiments in 3D sterotactic machine could not find the Sliced boneless ham. 2D ultrasound machine has not been able to find a small chalk powder group. The reproducibility of the three-dimensional stereotactic breast biopsy was better than effect of Sono guided biopsy.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2008.10a
• /
• pp.337-340
• /
• 2008

The cross section shape of intermediate die is one of important parameters to obtain dimensional accuracy of final product in shaped drawing process. Until now it has been designed by the experience or trial and error of the expert. In this study, the cross section shape of intermediate die fur spline shape is determined by the electronic field analysis, shape factor method. The result of the electronic field analysis, shape factor method has been compared with that of the present method. The effects of cross section shape on the dimensional accuracy were investigated by using FE analysis. And then the multi-stage shaped drawing experiments were performed to verify the results of FE analysis. As a result, the cross section shape from the electronic field analysis had the good dimensional accuracy. The electronic field analysis can be used for the method to obtain the cross section shape of intermediate die in shaped drawing process.

• Structural Monitoring and Maintenance
• /
• v.5 no.3
• /
• pp.417-428
• /
• 2018

In this paper, various model reduction methods were assessed using a shear frame, plane and space truss structures. Each of the structures is one-dimensional, two-dimensional and three-dimensional, respectively. Three scenarios of poor, better, and the best were considered for each of the structures in which 25%, 40%, and 60% of the total degrees of freedom (DOFs) were measured in each of them, respectively. Natural frequencies of the full and reduced order structures were compared in each of the numerical examples to assess the performance of model reduction methods. Generally, it was found that system equivalent reduction expansion process (SEREP) provides full accuracy in the model reduction in all of the numerical examples and scenarios. Iterated improved reduced system (IIRS) was the second-best, providing acceptable results and lower error in higher modes in comparison to the improved reduced system (IRS) method. Although the Guyan's method has very low levels of accuracy. Structures were classified with the excitation frequency. High-frequency structures compared to low-frequency structures have been poor performance in the model reduction methods (Guyan, IRS, and IIRS).

• 한국전산유체공학회:학술대회논문집
• /
• 2009.04a
• /
• pp.245-251
• /
• 2009

Numerical algorithms for solving the incompressible Navier-Stokes equations using P2P1 finite element are compared regarding the eigenvalues of matrices. P2P1 element allocates pressure at vertex nodes and velocity at both vertex and mid nodes. Therefore, compared to the P1P1 element, the number of pressure variables in the P2P1 element decreases to 1/4 in the case of two-dimensional problems and to 1/8 in the three-dimensional problems. Fully-implicit-integrated, semi-implicit- integrated and semi-segregated finite element formulations using P2P1 element are compared in terms of elapsed time, accuracy and eigenvlue distribution (condition number). For the comparison,they have been applied to the well-known benchmark problems. That is, the two-dimensional unsteady flows around a fixed circular cylinder and decaying vortex flow are adopted to check spatial accuracy.