• Journal of Internet Computing and Services
• /
• v.15 no.3
• /
• pp.71-77
• /
• 2014

The equalization algorithm based on the cross-information potential concept and Dirac-delta functions (CIPD) has outstanding ISI elimination performance even under impulsive noise environments. The main drawback of the CIPD algorithm is a heavy computational burden caused by the use of a block processing method for its weight update process. In this paper, for the purpose of reducing the computational complexity, a new method of the gradient calculation is proposed that can replace the double summation with a single summation for the weight update of the CIPD algorithm. In the simulation results, the proposed method produces the same gradient learning curves as the CIPD algorithm. Even under strong impulsive noise, the proposed method yields the same results while having significantly reduced computational complexity regardless of the number of block data, to which that of the e conventional algorithm is proportional.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.21 no.2
• /
• pp.164-169
• /
• 2010

Since the range and Doppler resolution of the synthetic aperture radar(SAR) image becomes very high, the target detection accuracy can be significantly increased, but the computational burden is also increased. The conventional single-cell based CFAR detector performs the target detection on every single cell basis, thus it causes the serious increment of the computational load. In this paper, the improved two-step MCA-CFAR detector is proposed for the improvement of the target detection as well as the reduction of computational load: the first step is to use the MCA-CFAR, and the second step is to use the single-cell based CFAR detection in the expected target area for final decision. The performance of the proposed algorithm is compared with the conventional single-cell based CFAR and MCA-CFAR on SAR images.

• Composites Research
• /
• v.14 no.1
• /
• pp.39-46
• /
• 2001

Experimental studies on the adaptive multi-mode vibration control of composite beams have been performed using neuro-controller. Neuro-controllers require too much computational burden, which blocks wide real-time applications of neuro-controllers. Therefore, in this paper, an adaptive notch filter is proposed to separate a vibration signal into each modal vibration signal. Two neuro-controllers with fewer weights are connected to the corresponding modal signals to generate proper modal control forces. The vibration controls using the adaptive notch filter and neuro-controllers have been performed for two specimens. A and B, which have different natural frequencies because of different positions of tip masses. Significant vibration reduction has been observed in both cases. The vibration control results show that the present neuro-controller has good adaptiveness under the system parameter variations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.2
• /
• pp.710-726
• /
• 2018

This paper presents an extended particle filter to increase the accuracy and decrease the computation load of vehicle tracking. Particle filter has been the subject of extensive interest in video-based tracking which is capable of solving nonlinear and non-Gaussian problems. However, there still exist problems such as preventing unnecessary particle consumption, reducing the computational burden, and increasing the accuracy. We aim to increase the accuracy without an increase in computation load. In proposed method, we calculate the direction angle of the target vehicle. The angular difference between the direction of the target vehicle and each particle of the particle filter is observed. Particles are filtered and weighted, based on their angular difference. Particles with angular difference greater than a threshold is eliminated and the remaining are stored with greater weights in order to increase their probability for state estimation. Threshold value is very critical for performance. Thus, instead of having a constant threshold value, proposed algorithm updates it online. The first advantage of our algorithm is that it prevents the system from failures caused by insufficient amount of particles. Second advantage is to reduce the risk of using unnecessary number of particles in tracking which causes computation load. Proposed algorithm is compared against camshift, direction-based particle filter and condensation algorithms. Results show that the proposed algorithm outperforms the other methods in terms of accuracy, tracking duration and particle consumption.

• Journal of Korean Society of Water and Wastewater
• /
• v.11 no.1
• /
• pp.21-32
• /
• 1997

Optimum design of water distribution network(WDN) in many times means just reducing redundancy. Given only a few situations are taken into consideration for such design, WDN deprived of inherited redundancy may not work properly in some unconsidered cases. Quantifying redundancy and incorporating it into the optimal design process will be a way of overcoming just reduction of redundancy. Expected shortage is developed as a reliability surrogate in WDN. It is an indicator of the frequency, duration and severity of failure. Using this surrogate, Expected Shortage Optimization Model (ESOM) is developed. ESOM is tested with an example network and results are analyzed and compared with those from other reliability models. The analysis results indicate that expected shortage is a quantitative surrogate measure, especially, good in comparing different designs and obtaining tradeoff between cost and. reliability. In addition, compared other models, ESOM is also proved useful in optimizing WDN with reliability and powerful in controlling reliability directly in the optimization process, even if computational burden is high. Future studies are suggested which focus on how to increase applicability and flexibility of ESOM.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.7
• /
• pp.846-853
• /
• 2018

In this paper, an improved model predictive control (MPC) method is proposed, which reduces the amount of calculations caused by the increased number of candidate voltage vectors with the increased voltage level in multi-level inverters. When the conventional MPC method is used for multi-level inverters, all candidate voltage vectors are considered to predict the next-step current value. However, in the case that the sampling time is short, increased voltage level makes it difficult to consider the all candidate voltage vectors. In this paper, the improved MPC method which can get a fast transient response is proposed with a small amount of the computation by adding new candidate voltage vectors that are set to find the optimal vector. As a result, the proposed method shows faster transient response than the method that considers the adjacent vectors and reduces the computational burden compared to the method that considers the whole voltage vector. the performance of the proposed method is verified through simulations and experiments.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.7
• /
• pp.1860-1868
• /
• 1998

A simple postprocessing algorithm is proposed to reduce the blocky artifacts of Block Discrete Cosine Transform (BDCT) coded images. Since the block noise is mostly antisymmetric relative to the block boundaries, we model the blocky noise as one-dimensional antisymmertric functions made by superposing DCT basis functions. observing the frequency characteristics of the noies model, we approximate its high frequency components as those of step functions. Then the proposed postprocessing algorithm eliminates the carefully selected high frequency components of step functions in the one-dimensional sN-point DCT domain, when the encoding block size is $N\;{\times}\;N$. It is shown that the proposed algorithm can also be performed in the spatial domain without computational burden of transforms. The experimental results show that the proposed algorithm well reduces the blocky artifacts in both subjective and objectie viewpoints.

• Proceedings of the IEEK Conference
• /
• summer
• /
• pp.391-394
• /
• 2004

This paper focuses on the DSP implementation of an HMM-based speech recognizer that can handle several hundred words of vocabulary size as well as speaker independency. First, we develop an HMM-based speech recognition system on the PC that operates on the frame basis with parallel processing of feature extraction and Viterbi decoding to make the processing delay as small as possible. Many techniques such as linear discriminant analysis, state-based Gaussian selection, and phonetic tied mixture model are employed for reduction of computational burden and memory size. The system is then properly optimized and compiled on the TMS320C6711 DSP for real-time operation. The implemented system uses 486kbytes of memory for data and acoustic models, and 24.5kbytes for program code. Maximum required time of 29.2ms for processing a frame of 32ms of speech validates real-time operation of the implemented system.

• Journal of Information Processing Systems
• /
• v.18 no.5
• /
• pp.677-687
• /
• 2022

We propose a detection algorithm based on tree-structured convolutional neural networks (TsCNNs) that finds pornography, propaganda, or other inappropriate content on a social media network. The algorithm sequentially applies the typical convolutional neural network (CNN) algorithm in a tree-like structure to minimize classification errors in similar classes, and thus improves accuracy. We implemented the detection system and conducted experiments on a data set comprised of 6 ordinary classes and 11 inappropriate classes collected from the Korean military social network. Each model of the proposed algorithm was trained, and the performance was then evaluated according to the images and videos identified. Experimental results with 20,005 new images showed that the overall accuracy in image identification achieved a high-performance level of 99.51%, and the effectiveness of the algorithm reduced identification errors by the typical CNN algorithm by 64.87 %. By reducing false alarms in video identification from the domain, the TsCNNs achieved optimal performance of 98.11% when using 10 minutes frame-sampling intervals. This indicates that classification through proper sampling contributes to the reduction of computational burden and false alarms.

• Journal of Broadcast Engineering
• /
• v.19 no.2
• /
• pp.240-249
• /
• 2014

According to the compressive sensing theory, it is possible to perfectly reconstruct a signal only with a fewer number of measurements than the Nyquist sampling rate if the signal is a sparse signal which satisfies a few related conditions. From practical viewpoint for image applications, it is important to reduce its computational complexity and memory burden required in reconstruction. In this regard, a Block-based Compressive Sensing (BCS) scheme with Smooth Projected Landweber (BCS-SPL) has been already introduced. However, it still has the computational complexity problem in reconstruction. In this paper, we propose a method which modifies its stopping criterion, tolerance, and convergence control to make it converge faster. Experimental results show that the proposed method requires less iterations but achieves better quality of reconstructed image than the conventional BCS-SPL.