• The Transactions of the Korea Information Processing Society
• /
• v.4 no.11
• /
• pp.2822-2831
• /
• 1997

A B-ISDN network based on ATM must support several kinds of transport services with different traffic characteristics and service requirements. There is neither link-by-link flow control nor error control in the ATM layer. For different services, different flow/error controls could be performed at the AAL layer or at a higher Iayer(e.g. transport layer). In traditional data networks, the window now control mechanism combined with error control was used prevalently. But, the window flow control mechanism might be useless in ATM networks because the propagation delay is too large compared with the transmission rate. In this paper, we propose a simple flow control mechanism, called RCT(Rate Control for end-to-end Transport), for end-to-end data transport. The RCT shows acceptable performance when the average overload period is bounded by a certain time.

• Journal of Korea Water Resources Association
• /
• v.32 no.6
• /
• pp.665-674
• /
• 1999

This paper confirmed the applicability of model to Korean rivers through the calibration and sensitivity analysis of LRCS rainfall runoff model for 18 storm events of Songriweon station in Nakdong river system, and achieved that LS and WLS were better than LAD by model fitting results. Diagonal element of "hat" matrix and affluence measures were used by analysis of parameter estimates, and parameter IL was the most important parameter in model output. By the results of error propagation according to parameter error, parameters IL, TP, F1 were affected by error propagation, and this is measure of sensitivity for the model output. This paper confirmed the relationship of calibration and sensitivity analysis of model through analysis of sensitivity coefficient, diagonal element $h_i$ and $D_i$._i$.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.6
• /
• pp.987-994
• /
• 2020

In this paper, we studied the effect of kernel size of CNN layer on performance in acoustic noise attenuators. This system uses a deep learning algorithm using a neural network adaptive prediction filter instead of using the existing adaptive filter. Speech is estimated from a single input speech signal containing noise using a 100-neuron, 16-filter CNN filter and an error back propagation algorithm. This is to use the quasi-periodic property in the voiced sound section of the voice signal. In this study, a simulation program using Tensorflow and Keras libraries was written and a simulation was performed to verify the performance of the noise attenuator for the kernel size. As a result of the simulation, when the kernel size is about 16, the MSE and MAE values are the smallest, and when the size is smaller or larger than 16, the MSE and MAE values increase. It can be seen that in the case of an speech signal, the features can be best captured when the kernel size is about 16.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.24 no.3
• /
• pp.310-315
• /
• 2013

This paper presents a two-port time-domain reflectometry(TDR) measurement technique for extracting the complex propagation constant of a cross-linked polyethylene(XLPE) cable. For the extraction, a short pulse transmitted through the cable is measured in the time domain and analyzed in the frequency domain. The propagation constant of a 22.9 kV XLPE cable with a conductor area of 325 $mm^2$ is extracted up to a frequency of approximately 2.14 GHz. The $S_{21}$ measured using a network analyzer and the two-port TDR technique are compared for verification. As a result compared with previous TDR method, the upper possible frequency limit for extracting the propagation constant increases and the measurement error decreases.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.6
• /
• pp.1511-1519
• /
• 2010

In this paper, we proposed and analyzed a RSSI based cooperative localization algorithm considering wireless propagation characteristics in indoor and outdoor environments for wireless sensor networks, which can estimate the BN position. The conventional RSSI based estimation scheme has low precision ranging due to instability propagation characteristics by time variable. Hence, we implemented ray-launching simulator for analysis of propagation characteristics in 4 case, and experimented proposed localization scheme with 4 RN and 1 to 5 BN. Simulation results show that NLCA has estimation error as 2m-3.5m, however, proposed CLA/ECLA has 1.3m-2.5m/0.5m-1.2m by same environments. Therefore, if we can consider channel characteristics, the proposed algorithm provides higher localization accuracy than RSSI based conventional one.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.8
• /
• pp.1595-1611
• /
• 1994

In this paper, the numerical wave propagation properties of the finite difference-time domain(FD-TD) method is investigated as a discrete model describing electromagnetic(EM) wave propagation phenomena. The leap-frog approximation of Maxwell's curl equations in time-space simulates EM wave propagation in terms of the numerical characteristic and the domain of dependence. A geometrical interpretation of the FD-TD numerical procedure is presented. The numerical dispersion error due to the leap-frog approximation and its dependence on the stability factor are illustrated. The FD-TD method using the leap-frog approximation is inherently a descriptive model. Thus, not only any physical picture about EM wave propagation phenomena can be drawn through this model, but also physical or engineering parameters in the frequency domain can be extracted from descriptive results. E-plane filter characteristics in the WR-28 rectangular waveguide and reflection property of an inductive iris in the WR-90 rectangluar waveguide extracted from simulation of the FD-TD model is included.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.7
• /
• pp.1450-1456
• /
• 2015

The received signal strengths according to the propagation distance between a transmitting shore station with a 53 m antenna hight and a receiving ship station with about 6m antenna hight from 700 m upto about 20 km are measured at ship's moving velocity of $5{\pm}1m/s$ using a single carrier signal at 150.0625MHz. In this paper, the path loss exponents at the propagation distance from 700 m to 20km are estimated via minimum-mean-square-error method using the measurements of the received signal strength, and the mean of the estimated path loss exponent become 3.79. The estimated propagation path loss exponents is similar to that calculated based on the field-strength values from 2 km to 20 km in Annex 2 of ITU-R P.1546-4, which is measuring results at maritime VHF at 100 MHz carrier frequency. Therefore, the estimated propagation path loss exponent shall be used as the expected path loss exponents for summer in costal area of South Sea of Korea.

• Journal of Advanced Navigation Technology
• /
• v.23 no.5
• /
• pp.392-399
• /
• 2019

In this paper, we have studied wireless time division duplex(TDD) time synchronization technique considering the propagation delay between mobile vehicles. The existing IEEE 1588 precision time protocol(IEEE 1588 PTP) algorithm was applied and the time synchronization between the two nodes was achieved through the propagation delay and clock offset time correction calculated between master slave nodes during wireless TDD communication. The time synchronization process and procedure of IEEE 1588 PTP algorithm were optimized, thereby reducing the propagation delay error sensitivity for real-time moving vehicles. The sync flag signal generated through the time correction has a time synchronization accuracy of max +252.5 ns within 1-symbol(1.74 M symbol/sec, ${\pm}287.35ns$) through test and measurement, and it was confirmed that the time synchronization between master slave nodes can be achieved through sync flag signal generated during GPS disturbance.

• Journal of IKEEE
• /
• v.9 no.2 s.17
• /
• pp.170-178
• /
• 2005

The video data corrupted by the transmission error due to packet loss induce error propagation in decoded video data, and cause poor video quality. To remedy these corrupted video data, there have been introduced two types of error concealment techniques: spatial or temporal error concealment algorithm. Computational overhead by using spatial error concealment algorithm is a serious disadvantage in mobile video data streaming environment. In this paper, we propose hybrid type error concealment technique recovering video quality of mobile device using MPEG-4 video streaming on error-prone wireless network. Our algorithm is implemented in MPEG-4 decoder. The algorithm adopts Intel Wireless MMX technology to provide high performance of portable embedded multimedia mobile device. It is proven that the proposed algorithm shows expected performance for a mobile streaming system(PDA) on IP channels. Our approach showed better processing speed and better video quality comparing with traditional error concealment algorithm.

• The Journal of Information Technology
• /
• v.4 no.1
• /
• pp.71-79
• /
• 2001

This paper presents an adaptive back propagation algorithm that update the learning parameter by the generated error, adaptively and configuration of the range for the initial connecting weight according to the different maximum target value from minimum target value. This algorithm is expected to escaping from the local minimum and make the best environment for the convergence. On the simulation tested this algorithm on three learning pattern. The first was 3-parity problem learning, the second was $7{\times}5$ dot alphabetic font learning and the third was handwritten primitive strokes learning. In three examples, the probability of becoming trapped in local minimum was reduce. Furthermore, in the alphabetic font and handwritten primitive strokes learning, the neural network enhanced to loaming efficient about 27%~57.2% for the standard back propagation(SBP).