• Title/Summary/Keyword: model reduction technique

Search Result 491, Processing Time 0.026 seconds

Target Speech Detection Using Gaussian Mixture Model of Frequency Bandwise Power Ratio for GSC-Based Beamforming (GSC 기반 빔포밍을 위한 주파수 밴드별 전력비 분포의 혼합 가우시안 모델을 이용한 목표 음성신호의 검출)

  • Chang, Hyungwook;Kim, Youngil;Jeong, Sangbae
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.19 no.1
    • /
    • pp.61-68
    • /
    • 2015
  • Noise reduction is necessary to compensate for the degradation of recognition performance by various types of noises. Among many noise reduction techniques using microphone array, generalized sidelobe canceller (GSC) has been widely applied to reduce nonstationary noises. The performance of GSC is directly affected by its adaptation mode controller (AMC). That is, accurate target speech detection is essential to guarantee the sufficient noise reduction in pure noise intervals and the less distortion in target speech intervals. Thus, this paper proposes an improved AMC design technique in which the power ratio of the output of fixed beamforming to that of blocking matrix is calculated frequency bandwise and probabilistically modeled by mixture Gaussians for each class. Experimental results show that the proposed algorithm outperforms conventional AMCs in receiver operating curves (ROC) and output SNRs.

Plate prebending using a three-dimensional-printed model affords effective anatomical reduction in clavicular shaft fractures

  • Hyungsuk Kim;Younsung Jung;Hyun Seok Song
    • Clinics in Shoulder and Elbow
    • /
    • v.26 no.4
    • /
    • pp.397-405
    • /
    • 2023
  • Background: A precontoured plate rarely fits properly within the patient's clavicle and must be bent intraoperatively. This study aimed to determine whether anatomical reduction could be achieved using a plate bent before surgery. Methods: This study included 87 consecutive patients with displaced mid-shaft clavicle fractures who underwent plate fixation and were followed-up for a minimum of 1 year. After exclusions, 39 consecutive patients underwent fixation with a precontoured plate bent intraoperatively (intraoperative bending group), and 28 underwent fixation with the plate bent preoperatively (preoperative bending group). Using free software and a three-dimensional (3D) printer, ipsilateral clavicle 3D-printed models were constructed. Using plain radiographs, the distance between the edge of the lateral inferior cortex and the medial inferior cortex was measured. The angle between the line connecting the inferior cortex edge and the line passing through the flat portion of the superior cortex of the distal clavicle was measured. Results: Mean length differences between the ipsilateral and contralateral clavicle were smaller on both anteroposterior (AP; P=0.032) and axial images (P=0.029) in the preoperative bending group. The mean angular differences on both AP (P=0.045) and axial images (P=0.008) were smaller in the preoperative bending group. No significant differences were observed between the two groups in functional scores at the last follow-up. Conclusions: Smaller differences in length and angle between the ipsilateral and contralateral clavicle, indicative of reduction, were observed in the preoperative bending group. Using the precontoured technique with low expense, the operation was performed more effectively as reflected by a shorter operation time. Level of evidence: III.

Dementia Prediction Model based on Gradient Boosting (이기종 머신러닝 모델 기반 치매예측 모델)

  • Lee, Taein;Oh, Hayoung
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.25 no.12
    • /
    • pp.1729-1738
    • /
    • 2021
  • Machine learning has a close relationship with cognitive psychology and brain science and is developing together. This paper analyzes the OASIS-3 dataset using machine learning techniques and proposes a model for predicting dementia. Dimensional reduction through PCA (Principal Component Analysis) is performed on the data quantifying the volume of each area among OASIS-3 data, and only important elements (features) are extracted and then various machine learning including gradient boosting and stacking Apply the models and compare the performance of each. Unlike previous studies, the proposed technique has a great differentiation because it uses not only the brain biometric data, but also basic information data such as the participant's gender and medical information data of the participant. In addition, it was shown that the proposed technique through various performance evaluations is a model that can better predict dementia by finding features that are more related to dementia among various numerical data.

Advanced discretization of rock slope using block theory within the framework of discontinuous deformation analysis

  • Wang, Shuhong;Huang, Runqiu;Ni, Pengpeng;Jeon, Seokwon
    • Geomechanics and Engineering
    • /
    • v.12 no.4
    • /
    • pp.723-738
    • /
    • 2017
  • Rock is a heterogeneous material, which introduces complexity in the analysis of rock slopes, since both the existing discontinuities within the rock mass and the intact rock contribute to the degradation of strength. Rock failure is often catastrophic due to the brittle nature of the material, involving the sliding along structural planes and the fracturing of rock bridge. This paper proposes an advanced discretization method of rock mass based on block theory. An in-house software, GeoSMA-3D, has been developed to generate the discrete fracture network (DFN) model, considering both measured and artificial joints. Measured joints are obtained from the photogrammetry analysis on the excavation face. Statistical tools then facilitate to derive artificial joints within the rock mass. Key blocks are searched to provide guidance on potential reinforcement measures. The discretized blocky system is subsequently implemented into a discontinuous deformation analysis (DDA) code. Strength reduction technique is employed to analyze the stability of the slope, where the factor of safety can be obtained once excessive deformation of slope profile is observed. The combined analysis approach also provides the failure mode, which can be used to guide the choice of strengthening strategy if needed. Finally, an illustrated example is presented for the analysis of a rock slope of 20 m height inclined at $60^{\circ}$ using combined GeoSMA-3D and DDA calculation.

The Study on the Effects of Mixer Configurations on Fluid Mixing Characteristics in SCR Systems (SCR 시스템의 믹서 구조 특성에 따른 유동 혼합 특성에 관한 연구)

  • Seo, Jin-Won;Lee, Kyu-Ik;Oh, Jeong-Taek;Choi, Yun-Ho;Lee, Jong-Hwa;Park, Jin-Il
    • Transactions of the Korean Society of Automotive Engineers
    • /
    • v.16 no.6
    • /
    • pp.192-199
    • /
    • 2008
  • The key issues for the reduction technologies of the exhaust gas from diesel engine being developed are to reduce particulate matters and NOx. The SCR system is known to be one of the most efficient and stable technologies to remove NOx through the mixing of NOx and urea solution. In the present research, the effects of mixer configurations of SCR system have been investigated to enhance the SCR performance. First, a Schlieren technique is employed to visualize the mixing characteristics of urea solution and exhaust gas. The results show that a mixer is essential to obtain proper fluid mixing. In addition, numerical studies have been made to understand the mixing characteristics through the comparison of the mal-distribution index of concentration at the several locations of the diffuser. In particular, the effects of number of blade and mixer angles on mixing characteristics were studied. The results show that the blade angle has a larger effect on the mixing characteristics than the number of blades.

A branch-switching procedure for analysing instability of steel structures subjected to fire

  • Morbioli, Andrea;Tondini, Nicola;Battini, Jean-Marc
    • Structural Engineering and Mechanics
    • /
    • v.67 no.6
    • /
    • pp.629-641
    • /
    • 2018
  • The paper describes the development of a two-dimensional (2D) co-rotational nonlinear beam finite element that includes advanced path-following capabilities for detecting bifurcation instability in elasto-plasticity of steel elements subjected to fire without introducing imperfections. The advantage is twofold: i) no need to assume the magnitude of the imperfections and consequent reduction of the model complexity; ii) the presence of possible critical points is checked at each converged time step based on the actual load and stiffness distribution in the structure that is affected by the temperature field in the elements. In this way, the buckling modes at elevated temperature, that may be different from the ones at ambient temperature, can be properly taken into account. Moreover, an improved displacement predictor for estimating the displacement field allowed significant reduction of the computational cost. A co-rotational framework was exploited for describing the beam kinematic. In order to highlight the potential practical implications of the developed finite element, a parametric analysis was performed to investigate how the beam element compares both with the EN1993-1-2 buckling curve and with experimental tests on axially compressed steel members. Validation against experimental data and numerical outcomes obtained with commercial software is thoroughly described.

An Orthogonal Phase-Superimposed Peak-to-Average Power Ratio Reduction Technique

  • Han, Tae-Young;Kim, Nam;Choi, Jung-Hun;Lee, Jae-Hwan
    • Journal of electromagnetic engineering and science
    • /
    • v.7 no.4
    • /
    • pp.169-174
    • /
    • 2007
  • This paper presents a method of superimposing the rotation phases over the pilot and data symbols in order to reduce the peak-to-average power ratio(PAPR) in orthogonal frequency division multiplexing(OFDM). The phases of the rotation vector are added to those of the pilot symbols and those of the data symbols by interlaying them between any two pilot symbols. The receiver restores the data symbols by utilizing the channel estimation of the pilot symbols. Therefore, the bandwidth efficiency is improved by not using the subcarriers that are assigned for the reduction of the PAPR. Also, the enormous increase of the bit error rate which would be caused by incorrectly receiving the side information, i.e. the phases of the rotation vector, is prevented. The simulation results of the bit error rate performance for the BPSK are given using the COST-207 channel model.

Surgical Management of a Mandible Subcondylar Fracture

  • Kang, Dong Hee
    • Archives of Plastic Surgery
    • /
    • v.39 no.4
    • /
    • pp.284-290
    • /
    • 2012
  • Open reduction and anatomic reduction can create better function for the temporomandibular joint, compared with closed treatment in mandible fracture surgery. Therefore, the double miniplate fixation technique via mini-retromandibular incision was used in order to make the most stable fixation when performing subcondylar fracture surgery. Those approaches provide good visualization of the subcondyle from the posterior edge of the ramus, allow the surgeon to work perpendicularly to the fracture, and enable direct fracture management. Understanding the biomechanical load in the fixation of subcondylar fractures is also necessary in order to optimize fixation methods. Therefore, we measured the biomechanical loads of four different plate fixation techniques in the experimental model regarding mandibular subcondylar fractures. It was found that the loads measured in the two-plate fixation group with one dynamic compression plate (DCP) and one adaption plate showed the highest deformation and failure loads among the four fixation groups. The loads measured in the one DCP plate fixation group showed higher deformation and failure loads than the loads measured in the two adaption plate fixation group. Therefore, we conclude that the selection of the high profile plate (DCP) is also important in order to create a stable load in the subcondylar fracture.

Support vector machine and multifactor dimensionality reduction for detecting major gene interactions of continuous data (서포트 벡터 머신 알고리즘을 활용한 연속형 데이터의 다중인자 차원축소방법 적용)

  • Lee, Jea-Young;Lee, Jong-Hyeong
    • Journal of the Korean Data and Information Science Society
    • /
    • v.21 no.6
    • /
    • pp.1271-1280
    • /
    • 2010
  • We have used multifactor dimensionality reduction (MDR) method to study genegene interaction effect of statistical model in general. But, MDR method could not be applied in the continuous data. In this paper, continuous-type data by the support vector machine (SVM) algorithm are proposed to the MDR method which provides an introduction to the technique. Also we apply the method on the identify major interaction effects of single nucleotide polymorphisms (SNPs) responsible for economic traits in a Korean cattle population.

An Optimal Design of Notch Shape of IPM BLDC Motor Using the Differential Evolution Strategy Algorithm (차분진화 알고리즘을 이용한 IPM형 BLDC전동기의 Notch 형상 최적화 설계 연구)

  • Shin, Pan Seok;Kim, Hong Uk
    • The Transactions of The Korean Institute of Electrical Engineers
    • /
    • v.65 no.2
    • /
    • pp.279-285
    • /
    • 2016
  • In this paper, a cogging torque of IPM(Interior Permanent Magnet)-type BLDC motor is analyzed by FE program and the optimized notch on the rotor surface is designed to minimize the torque ripple. A differential evolution strategy algorithm and a response surface method are employed to optimize the rotor notch. In order to verify the proposed algorithm, an IPM BLDC motor is used, which is 50 kW, 8 poles, 48 slots and 1200 rpm at the rated speed. Its characteristics of the motor is calculated by FE program and 4 design variables are set on the rotor notch. The initial shape of the notch is like a non-symmetric half-elliptic and it is optimized by the developed algorithm. The cogging torque of the final model is reduced to $1.5[N{\cdot}m]$ from $5.2[N{\cdot}m]$ of the initial, which is about 71 % reduction. Consequently, the proposed algorithm for the cogging torque reduction of IPM-type BLDC motor using the rotor notch design seems to be very useful to a mechanical design for reducing noise and vibration.