• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.10B
• /
• pp.1070-1075
• /
• 2009

In this paper, we propose new depth estimation and intermediate view synthesis algorithms for three-dimensional video generation. In order to improve temporal consistency of the depth map sequence, we add a temporal weighting function to the conventional matching function when we compute the matching cost for estimating the depth information. In addition, we propose a boundary noise removal method in the view synthesis operation. after finding boundary noise areas using the depth map, we replace them with corresponding texture information from the other reference image. Experimental results showed that the proposed algorithm improved temporal consistency of the depth sequence and reduced flickering artifacts in the virtual view. It also improved visual quality of the synthesized virtual views by removing the boundary noise.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.12B
• /
• pp.1211-1218
• /
• 2010

OSPF is the optimized routing protocol in wired network and considered as a tactical routing protocol in wireless tactical communication network. However because it is designed basically based on wired environment, it runs inadequately in wireless tactical environment: noise and jamming signal. So, we proposed new OSPF cost function to develop active routing protocol in wireless tactical communication network. In redefined cost function, there are four parameters that are relative transmission speed, link weight, router utilization, link average BER(Bit Error Rate). These parameters reflect wireless tactical characters. Also, we remodel the option field in Hello packet. It can help user to periodically check the link state. From the simulation result, it is shown that proposed OSPF is better than OSPF in jamming situation and has accumulative delay gain with dispersion of traffic load in entire network.

• Journal of KIISE
• /
• v.42 no.2
• /
• pp.190-202
• /
• 2015

Much research has been conducted on location-based intelligent personal assistants that can understand a user's intention by learning the user's route model and then inferring the user's destinations and routes using data of GPS and other sensors in a smartphone. The intelligence of the location-based personal assistant is contingent on the accuracy and efficiency of the real-time predictions of the user's intended destinations and routes by processing movement information based on uncertain sensor data. We propose a robust particle filter based on Dynamic Bayesian Network model to infer the user's routes. The proposed robust particle filter includes a particle generator to supplement the incorrect and incomplete sensor information, an efficient switching function and an weight function to reduce the computation complexity as well as a resampler to enhance the accuracy of the particles. The proposed method improves the accuracy and efficiency of determining a user's routes and destinations.

• The Journal of the Korea Contents Association
• /
• v.10 no.12
• /
• pp.34-42
• /
• 2010

The DOA(direction of arrival), which is based on parametric and nonparametric estimation algorithm, and adaptive beamforming algorithm for mobile communication environments are researched and analyzed. In parametric estimation algorithm, eigenvalues of the signal component and the noise component are obtained from correlation matrix of received signal by array antenna and power spectrum of the received signal is discriminated from them. Otherwise, in nonparametric estimation algorithm, we minimize a regularized objective function for finding a estimate of the signal energy as a function of angle, using nonquadratic norm which leads to supper resolution and noise suppression. And then, DOA is estimated by the signal and noise spatial steering vector, and adaptive beam-forming pattern is improved by weight vectors obtained from the spatial vector. Therefore, the improved directional estimation algorithm with regularizing sparsity constraints offers super-resolution and noise suppression compared to other algorithms.

• Journal of KIISE:Software and Applications
• /
• v.28 no.9
• /
• pp.603-611
• /
• 2001

This paper proposes a novel fast layer-by-layer algorithm that has better generalization capability. In the proposed algorithm, the weights of the hidden layer are updated by the target vector of the hidden layer obtained by least squares method. The proposed algorithm improves the learning speed that can occur due to the small magnitude of the gradient vector in the hidden layer. This algorithm was tested in a handwritten digits recognition problem. The learning speed of the proposed algorithm was faster than those of error back propagation algorithm and modified error function algorithm, and similar to those of Ooyen's method and layer-by-layer algorithm. Moreover, the simulation results showed that the proposed algorithm had the best generalization capability among them regardless of the number of hidden nodes. The proposed algorithm has the advantages of the learning speed of layer-by-layer algorithm and the generalization capability of error back propagation algorithm and modified error function algorithm.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.2
• /
• pp.113-118
• /
• 2017

This paper propose the NM-MMA (Novel Mixed-MMA) that is possible to improving the convergence speed of current MMA algorithm and reducing the high MSE of SE-MMA algorithm, and its equalization performance were analyzed. The cost function of the NM-MMA configured as the sum of appropriate weights of gradient vector of current MMA and SE-MMA, and then it used for the updating the tap coefficient of equalizer. The computer simulation was performed applying the same environment in the channel, step size and signal to noise ratio, and the same performance index in equalizer output signal constellation, residual isi, MSE, SER was used. As a result of computer simulation, the proposed NM-MMA has fast convergence time than MMA, and less in MSE and SER performance compared to SE-MMA.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.10
• /
• pp.123-131
• /
• 1999

In this paper, we proposed stereo matching algorithm by dynamic programming with edges emphasized. Existing algorithms show blur generally at depth discontinuities owing to smoothness constraint and non-existence of matching pixel in occlusion regions. Also it accompanies matching error by lackness of matching information in the untextured regions. This paper defines new cost function to make up for the problems occurred to existing algorithms. It is possible through deriving matching of edges in left and right images to be carried out between edge regions anf deriving that in the other regions to be peformed between the other regions. In case of the possibility that edges can be Produced in a large amount, matching between edge information adds weight to cost function in proportion to Path distance. Proposed algorithm was applied to various images obtained by convergent camera model as well as parallel camera model. As the result, proposed algorithm showed improved performance in the aspect of matching error and processing in the occlusion regions compared to existing algorithms. Also it could improve blur especially in discontinuity regions.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.13 no.2
• /
• pp.208-216
• /
• 2002

In this paper, the design method of a waveguide broad-wall longitudinal slot array monopulse antenna of X-type monopulse axes is presented, and the method is verified through manufacture and measurement. In the antenna design of this paper, the antenna size is small and the monopulse axes are X-type. Thus, the common continuous aperture distribution fuction is not suitable and the power balance among antenna quadrants should be considered. Also, since the waveguide height is reduced into 0.1 wavelength, the modelling of the slot characteristics is not simple. Thus, in this paper, the aperture distribution is optimized by using random number, and the balance among the quadrants is achieved by applying the quadrant weighting factor during the aperture optimization process. Also, the moment method procedure is accelerated by applying the interpolation technique to some part of the moment matrix, and the moment method procedure is added to the array synthesis program as a subroutine so that the slot characteristics can be calculated directly when it is required. Based on this method, a antenna of 28dBi is designed and manufactured. It is found that the antenna characteristics is similar to design data.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.5
• /
• pp.777-786
• /
• 1995

This paper describes several algorithms for technology mapping of logic functions into interesting and popular FPGAs that use look-up table memories. In order to improve the technology mapping for FPGA, some existing multi-level logic synthesis, decomposition reduction and packing techniques are analyzed and compared. And then new algorithms such as node-pair decomposition, merging fanin, unified reduction and multiple output decomposition which are used for combinational logic design, are proposed. The cost function is used to minimize the number of CLBs and edges of the network. The cost is a linear combination of each weight that is given by user. Finally we compare our new algorithm with previous logic design technique［8］. In an experimental comparison our algorithm requires 10% fewer CLB and nets than SIS-pga.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.6
• /
• pp.503-511
• /
• 2013

This paper presents a speech enhancement method using non-negative matrix factorization. In training phase, we can obtain each basis matrix from speech and specific noise database. After training phase, the noisy signal is separated from the speech and noise estimate using basis matrix in enhancement phase. In order to improve the performance, we model the change of encoding matrix from training phase to enhancement phase using independent Gaussian distribution models, and then use the constraint of the objective function almost same as that of the above Gaussian models. Also, we perform a smoothing operation to the encoding matrix by taking into account previous value. Last, we apply the Log-Spectral Amplitude type algorithm as gain function.