• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.14 no.5
• /
• pp.30-37
• /
• 2015

The study for enhancing the data transmission rate of the next generation wireless communication system using MIMO system operating in the frequency selective fading environment is currently actively conducted. Mixed signal from each transmitted antennas are received at antennas. The training signal with orthogonal property is needed to separate the mixed signal and enable to estimate channel and time synchronization. In this paper we introduce several training sequences used in MIMO communication system and proposed the modified WeCAN sequence with good auto-correlation property in interested area. We compared auto-correlation property of each sequence via simulation and compared the performance of sequences in doppler shift and multipath fading channel.

• Journal of Korean Neurosurgical Society
• /
• v.61 no.1
• /
• pp.97-104
• /
• 2018

Objective : Diffusion-weighted magnetic resonance imaging (DW-MRI) has proven useful in the study of the natural history of ischemic stroke. However, the potential of DW-MRI for the evaluation of chronic subdural hematoma (CSDH) has not been established. In this study, we investigated DW-MRI findings of CSDH and evaluated the impact of the image findings on postoperative outcomes of CSDH. Methods : We studied 131 CSDH patients who had undergone single burr hole drainage surgery. The images of the subdural hematomas on preoperative DW-MRI and computed tomography (CT) were divided into three groups based on their signal intensity and density : 1) homogeneous (iso or low) density on CT and homogeneous low signal intensity on DW-MRI; 2) homogeneous (iso or low) density on CT and mixed signal intensity on DW-MRI; and 3) heterogeneous density on CT and mixed signal intensity on DW-MRI. On the basis of postoperative CT, we also divided the patients into 3 groups of surgical outcomes according to residual hematoma and mass effect. Results : Analysis showed statistically significant differences in surgical (A to B : p<0.001, A to C : p<0.001, B to C : p=0.129) and functional (A to B : p=0.039, A to C : p<0.001, B to C : p=0.108) outcomes and treatment failure rates (A to B : p=0.037, A to C : p=0.03, B to C : p=1) between the study groups. In particular, group B and group C showed worse outcomes and higher treatment failure rates than group A. Conclusion : CSDH with homogeneous density on CT was characterized by signal intensity on DW-MRI. In CSDH patients, performing DW-MRI as well as CT helps to predict postoperative treatment failure or complications.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.9
• /
• pp.1158-1165
• /
• 2021

This paper proposes a CNN accelerator which is optimized Pooling layer operation incorporated in Multiplication And Accumulation(MAC) to reduce the memory size. For optimizing memory and data path circuit, the quantized 8bit integer weights are used instead of 32bit floating-point weights for pre-training of MNIST data set. To reduce chip area, the proposed CNN model is reduced by a convolutional layer, a 4*4 Max Pooling, and two fully connected layers. And all the operations use specific MAC with approximation adders and multipliers. 94% of internal memory size reduction is achieved by simultaneously performing the convolution and the pooling operation in the proposed architecture. The proposed accelerator chip is designed by using TSMC65nmGP CMOS process. That has about half size of our previous paper, 0.8*0.9 = 0.72mm². The presented CNN accelerator chip achieves 94% accuracy and 77us inference time per an MNIST image.

• Journal of Korean Society of Transportation
• /
• v.20 no.5
• /
• pp.193-206
• /
• 2002

A signal optimization model is proposed by applying the Cell-Transmission Model(CTM) as an embedded traffic flow model to estimate a system-optimal signal timing plan in a transportation network composed of signalized intersections. Beyond the existing signal-optimization models, the CTM provides appropriate theoretical and practical backgrounds to simulate oversaturation phenomena such as shockwave, queue length, and spillback. The model is formulated on the Mixed-Integer Programming(MIP) theory. The proposed model implies a system-optimal in a sense that traffic demand and signal system cooperate to minimize the traffic network cost: the demand departing from origins through route choice behavior until arriving at destinations and the signal system by calculating optimal signal timings considering the movement of these demand. The potential of model's practical application is demonstrated through a comparison study of two signal control strategies: optimal and fixed signal controls.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1627-1634
• /
• 2021

This paper introduces a low-power compact ADC circuit for analog Convolutional filter for low-power neural network accelerator SOC. While convolutional neural network accelerators can speed up the learning and inference process, they have drawback of consuming excessive power and occupying large chip area due to large number of multiply-and-accumulate operators when implemented in complex digital circuits. To overcome these drawbacks, we implemented an analog convolutional filter that consists of an analog multiply-and-accumulate arithmetic circuit along with an ADC. This paper is focused on the design optimization of a low-power 8bit SAR ADC for the analog convolutional filter accelerator We demonstrate how to minimize the capacitor-array DAC, an important component of SAR ADC, which is three times smaller than the conventional circuit. The proposed ADC has been fabricated in CMOS 65nm process. It achieves an overall size of 1355.7㎛², power consumption of 2.6㎼ at a frequency of 100MHz, SNDR of 44.19 dB, and ENOB of 7.04bit.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2003.06a
• /
• pp.296-299
• /
• 2003

This paper shows a novel and simple IIR/FIR QMF(quadrature mirror filter) filter banks, mixed IIR and FIR structure. Here, FIR filters used for phase compensation. In this paper, we introduced analysis and synthesis filter banks, which used FIR linear phase filters and all pass filters. In result, phase response of analysis and synthesis filter banks become approximately linear characteristic. Simultaneously, a liasing distortion can be completely canceled.

• Journal of Korean Society of Transportation
• /
• v.18 no.4
• /
• pp.7-18
• /
• 2000

KS-SIGNAL, a traffic signal optimization model for coordinated arterials, is an optimization model using the mixed integer linear Programming that minimizes total delay on arterials by optimizing left-turn Phase sequences. However, the Previous version of KS-SIGNAL had a difficulty in reducing computation speed because the related variables and constraints multiply rapidly in accordance with the increase of intersections. This study is designed to propose a new model, improving optimizing computation speed in KS-SIGMAl, and evaluate it. This Paper Puts forth three kinds of methodological approaches as to achieve the above goals. At the first step to reduce run-time in the proposed model objective function and a few constraints are Partially modified, which replaces variable in related to queue clearance time with constant, by using thru-movements at upstream intersection and the length of red time at downstream intersection. The result shows that the run-time can be reduced up to 70% at this step. The second step to load the library in LINDO for Windows, in order to solve mixed integer linear programming. The result suggests that run-time can be reduced obviously up to 99% of the first step result. The third step is to add constraints in related to left-turn Phase sequences. The proposed methodological approach, not optimizing all kinds of left-turn sequences, is more reasonable than that of previous model , only in the view of reducing run-tim. In conclusion, run-time could be reduced up to 30% compared with the second results. This Proposed model was tested by several optimization scenarios. The results in this study reveals that signal timing plan in KS-SIGNAL is closer to PASSER-II (bandwidth maximizing model) rather than to TRANSYT-7F(delay minimizing model).

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1957-1962
• /
• 2003

In order to design the phantom of heart, we have developed the multi-dipole current source system. Such a one be clue to the various motion of heart. The magnetocardiograph (MCG) system for diagnosing the disease of the heart due to an analysis of the heart signal. The multidipole current source system be built by microprocessor. We use the shield room to obtain a good experimental result. Then the signal acquired is mixed with a background noise, through a filtering extracts a pure signal. The pure signal such a heart phantom is analyzed by an electromagnetic map.

• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.39-42
• /
• 1998

In this paper is design of signal discriminatior system for the third generation mobile telecommunication system (IMT-2000 : International mobile telecommunication service-2000) This system is basically using multiplexing method of CDMA, TDMA and FDMA that this system can be solved third generation systems problems-Frequency bandwidth, international roaming and development new architecture of mobile telecommunication systems. Designed signal discriminator system is perfectly separated three styles multiplexing method signals that more easily, nearly and economically implementation of third generation telecommunication systems. As we develop signal discriminator we will contribute to technical evolution of mobile telecommunication for third generation and stay ahead in the interior technical accumulation in addition to international intercourse as to mobile telecommunication.

• Journal of Korean Society of Transportation
• /
• v.15 no.3
• /
• pp.53-74
• /
• 1997

신호연동화모형에서 통과폭모형은 지체를 최소화하지 못하고 지체도모형은 좌회전현시순서를 최적화하지 못한다. 이러한 모형의 단점을 극복하고자 개발된 KS_SIGNAL은 지체를 최소화하는 신호연동화모형이지만 지체를 산정하는데 있어 많은 한계를 갖고 있다. 본 연구에서는 기존에 개발된 KS_SIGNAL의 지체모형을 개선하여 보다 우수한 연동신호시간과 현시순서를 산정해주는 신호최적화모형(KS_SIGNAL II)을 개발하여 FORTRAN 언어로 전산화하였다. 개발된 모형은 다양한 평가를 통하여 기존 모형들보다 전반적으로 우수한 것으로 입증되었다. 본 모형을 통하여 차량 출발 및 도착형태에 따른 대기차량 소거시간을 고려하는 옵셋 산정이 가능해져 지체를 최소화하는 간선도로의 신호최적화모형으로 활용될 경우 도로기능제고 및 지체도 감소에 의한 편익을 얻을 수 있다.