• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.11a
• /
• pp.202-203
• /
• 2017

비음수 행렬 인수분해(Non-negative Matrix Factorization, NMF)의 신호분리 성능을 개선하기 위해 희소조건을 인가한 방법이 희소 비음수 행렬 인수분해 알고리즘(Sparse NMF, SNMF)이다. 기존의 SNMF 알고리즘은 개별 음원의 희소성을 고려하지 않고 임의로 결정한 희소 조건을 사용한다. 본 논문에서는 음원의 특성에 따른 희소성을 추정하고 이를 SNMF 학습알고리즘에 적용하는 새로운 신호분리 기법을 제안한다. 혼합 신호에서의 잡음제거 실험을 통해, 제안한 방법이 기존의 NMF와 SNMF에 비해 성능이 더 우수함을 보였다.

• Wind and Structures
• /
• v.11 no.3
• /
• pp.241-256
• /
• 2008

This paper adopts autoregressive (AR) model to simulate the wind velocity of spatial three-dimensional fields in accordance with the time and space dependent characteristics of the 3-D fields. Based on the built MATLAB programming, this paper discusses in detail the issues of the AR model deduced by matrix form in the simulation and proposes the corresponding solving methods: the over-relaxation iteration to solve the large sparse matrix equations produced by large number of degrees of freedom of structures; the improved Gauss formula to calculate the numerical integral equations which integral functions contain oscillating functions; the mixed congruence and central limit theorem of Lindberg-Levy to generate random numbers. This paper also develops a method of ascertaining the rank of the AR model. The numerical examples show that all those methods are stable and reliable, which can be used to simulate the wind velocity of all large span structures in civil engineering.

• Water for future
• /
• v.29 no.1
• /
• pp.141-150
• /
• 1996

Steady state analysis of pressure and flow in water supply piping systems is a problem of great importance in hydraulic engineering. The basic equations consist of continuity equation and energy equation. The network equations are solved iteratively by using linear solution method. The resulting linear simultaneous equations are solved by frontal method. Frontal method, which is suitable to sparse matrix, gathers only non-zero entries in coefficient matrix. The suggested methodology can analyze faster than the existing routines by using smaller computer memory. The model presented in this study shows accurate and efficient results for various piping systems.

• Journal of Information Processing Systems
• /
• v.19 no.2
• /
• pp.232-239
• /
• 2023

Aiming at the poor compression quality of traditional image compression coding (ICC) algorithm, a multi-description ICC algorithm based on depth learning is put forward in this study. In this study, first an image compression algorithm was designed based on multi-description coding theory. Image compression samples were collected, and the measurement matrix was calculated. Then, it processed the multi-description ICC sample set by using the convolutional self-coding neural system in depth learning. Compressing the wavelet coefficients after coding and synthesizing the multi-description image band sparse matrix obtained the multi-description ICC sequence. Averaging the multi-description image coding data in accordance with the effective single point's position could finally realize the compression coding of multi-description images. According to experimental results, the designed algorithm consumes less time for image compression, and exhibits better image compression quality and better image reconstruction effect.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.24-26
• /
• 2022

시스톨릭 어레이는 DNN training 등 인공지능 연산의 대부분을 차지하는 행렬 곱셈을 수행하기 위한 하드웨어 구조로 많이 사용되지만, sparsity 가 높은 행렬을 연산할 때 불필요한 동작으로 인해 효율성이 크게 떨어진다. 본 논문에서 제안된 유동적 시스톨릭 어레이는 matrix condensing, weight switching, 그리고 direct output path 의 방법과 구조를 통해 sparsity 가 높은 행렬 곱셈의 수행 사이클을 줄일 수 있다. 시뮬레이션을 통해 기존 시스톨릭 어레이와 유동적 시스톨릭 어레이의 성능을 비교하였으며 8×8, 16×16, 32×32 의 크기를 가진 행렬을 동일 크기의 시스톨릭 어레이로 연산하였을 때 필요 사이클 수를 최대 12 사이클 절감할 수 있는 것을 확인하였다.

• Journal of Information Processing Systems
• /
• v.17 no.3
• /
• pp.462-472
• /
• 2021

With an increase in the scale of recommender systems, users' rating data tend to be extremely sparse. Some methods have been utilized to alleviate this problem; nevertheless, it has not been satisfactorily solved yet. Therefore, we propose an effective pre-rating method based on users' dichotomous preferences and average ratings fusion. First, based on a user-item ratings matrix, a new user-item preference matrix was constructed to analyze and model user preferences. The items were then divided into two categories based on a parameterized dynamic threshold. The missing ratings for items that the user was not interested in were directly filled with the lowest user rating; otherwise, fusion ratings were utilized to fill the missing ratings. Further, an optimized parameter λ was introduced to adjust their weights. Finally, we verified our method on a standard dataset. The experimental results show that our method can effectively reduce the prediction error and improve the recommendation quality. As for its application, our method is effective, but not complicated.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.8
• /
• pp.917-920
• /
• 2016

SpSF algorithm is direction-of-arrival estimation algorithm based on sparse representation of incident signlas. Cost function to be optimized for DOA estimation is multi-dimensional nonlinear function, which is hard to handle for optimization. After some manipulation, the problem can be cast into convex optimiztion problem. Convex optimization problem tuns out to be constrained optimization problem, where the parameter in the constraint has to be determined. The solution of the convex optimization problem is dependent on the specific parameter value in the constraint. In this paper, we propose a rule-of-thumb for determining the parameter value in the constraint. Based on the fact that the noise in the array elements is complex Gaussian distributed with zero mean, the average of the Frobenius norm of the matrix in the constraint can be rigorously derived. The parameter in the constrint is set to be two times the average of the Frobenius norm of the matrix in the constraint. It is shown that the SpSF algorithm actually works with the parameter value set by the method proposed in this paper.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.5
• /
• pp.2082-2102
• /
• 2018

In recent years, social network related applications such as WeChat, Facebook, Twitter and so on, have attracted hundreds of millions of people to share their experience, plan or organize, and attend social events with friends. In these operations, plenty of valuable information is accumulated, which makes an innovative approach to explore users' preference and overcome challenges in traditional recommender systems. Based on the study of the existing social network recommendation methods, we find there is an abundant information that can be incorporated into probability matrix factorization (PMF) model to handle challenges such as data sparsity in many recommender systems. Therefore, the research put forward a unified social network recommendation framework that combine tags, trust between users, ratings with PMF. The uniformed method is based on three existing recommendation models (SoRecUser, SoRecItem and SoRec), and the complexity analysis indicates that our approach has good effectiveness and can be applied to large-scale datasets. Furthermore, experimental results on publicly available Last.fm dataset show that our method outperforms the existing state-of-art social network recommendation approaches, measured by MAE and MRSE in different data sparse conditions.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.29 no.9 s.240
• /
• pp.1049-1056
• /
• 2005

A conservative pressure-based finite-volume numerical method has been developed for computing flow and heat transfer by using an unstructured grid system. The method admits arbitrary convex polyhedra. Care is taken in the discretization and solution procedures to avoid formulations that are cell-shape-specific. A collocated variable arrangement formulation is developed, i.e. all dependent variables such as pressure and velocity are stored at cell centers. Gradients required for the evaluation of diffusion fluxes and for second-order-accurate convective operators are found by a novel second-order accurate spatial discretization. Momentum interpolation is used to prevent pressure checkerboarding and the SIMPLE algorithm is used for pressure-velocity coupling. The resulting set of coupled nonlinear algebraic equations is solved by employing a segregated approach, leading to a decoupled set of linear algebraic equations fer each dependent variable, with a sparse diagonally dominant coefficient matrix. These equations are solved by an iterative preconditioned conjugate gradient solver which retains the sparsity of the coefficient matrix, thus achieving a very efficient use of computer resources.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.40 no.9
• /
• pp.597-604
• /
• 2016

A parallel algorithm of bi-conjugate gradient method was developed based on CUDA for parallel computation of the incompressible Navier-Stokes equations. The governing equations were discretized using splitting P2P1 finite element method. Asymmetric stenotic flow problem was solved to validate the proposed algorithm, and then the parallel performance of the GPU was examined by measuring the elapsed times. Further, the GPU performance for sparse matrix-vector multiplication was also investigated with a matrix of fluid-structure interaction problem. A kernel was generated to simultaneously compute the inner product of each row of sparse matrix and a vector. In addition, the kernel was optimized to improve the performance by using both parallel reduction and memory coalescing. In the kernel construction, the effect of warp on the parallel performance of the present CUDA was also examined. The present GPU computation was more than 7 times faster than the single CPU by double precision.