• Journal of Korea Water Resources Association
• /
• v.34 no.6
• /
• pp.701-711
• /
• 2001

Forecasting of water quality variation is not an easy process due to the complicated nature of various water quality factors and their interrelationships. The objective of this study is to test the applicability of neural network models to the forecasting of the water quality at Gongju station in Geum River. This is done by forecasting monthly water qualities such as DO, BOD, and TN, and comparing with those obtained by ARIMA model. The neural network models of this study use BP(Back Propagation) algorithm for training. In order to improve the performance of the training, the models are tested in three different styles ; MANN model which uses the Moment-Adaptive learning rate method, LMNN model which uses the Levenberg-Marquardt method, and MNN model which separates the hidden layers for judgement factors from the hidden layers for water quality data. the results show that the forecasted water qualities are reasonably close to the observed data. And the MNN model shows the best results among the three models tested

• Journal of Digital Contents Society
• /
• v.18 no.1
• /
• pp.141-147
• /
• 2017

Scalable extension of High Efficiency Video Coding (SHVC) standard uses the up-sampled residual data from the base layer to make a residual data in the enhancement layer. This paper describes an efficient algorithm for improving coding gain by using the filtered residual signal of base layer in the Scalable extension of High Efficiency Video Coding (SHVC). The proposed adaptive filter selection mechanism uses the smoothing and sharpening filters to enhance the quality of inter-layer prediction. Based on two filters and the existing up-sampling filter, a rate-distortion (RD)-cost fuction-based competitive scheme is proposed to get better quality of video. Experimental results showed that average BD-rate gains of 1.5%, 2.1%, and 1.7% for Y, U and V components, respectively, were achieved, compared with SHVC reference software 5.0, which is based on HEVC reference model (HM) 13.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.3B
• /
• pp.255-263
• /
• 2009

In this paper, we propose an adaptive beacon scheduling algorithm to control a reception period of actual data according to variation of amount of traffic in IEEE 802.15.4 LR-WPAN(Low Rate-Wireless Personal Area Network) with the cluster-tree topology. If a beacon interval is shortened, the amount of the traffic a node receives can be increased while consumption of the energy can be also increased. In this sense, we can achieve optimal on orgy consumption by controlling the beacon interval when the amount of data to be received is being decreased. The result of simulation using NS-2 shows that the proposed algorithm improves performances in terms of packet loss rate and end-to-end delay compared with algorithm using a fixed beacon interval. For a design of cluster-tree based LR-WPAN managing delay-sensitive services, the proposed algorithm and the associated results can be applied usefully.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.7
• /
• pp.1618-1626
• /
• 2013

Virtualization service processes all operation including the data creation, storing, and disposal in a server and transmits processed data as the streaming media form. Therefore, client can use the same environment as the traditional desktop environment without considering the type of device. Virtualization service should consider not only the video quality but also the delay bounds and continuity of video playback for improving the user perceived Quality of Service(QoS) of streaming service. In this paper, we propose a network-adaptive transmission architecture that focuses on guaranteeing QoS requirements for virtualization service. In order to provide those, the proposed architecture have the transmission rate adaptation function based on available bandwidth and the content bit-rate control function based on sender buffer state. Through each function, proposed architecture guarantee the delay bounds and continuity of virtualization contents playback. The simulation results show that proposed network-adaptive transmission architecture provides a improve performance of throughput and transmission delay.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.6
• /
• pp.62-69
• /
• 2015

In this paper, a new frame rate up conversion (FRUC) algorithm using adaptive motion estimation (AME-FRUC) is proposed. The proposed algorithm performs extended bilateral motion estimation (EBME) conducts motion estimation (ME) processes on the static region, and extract region of interest with the motion vector (MV). In the region of interest block, the proposed AME-FRUC uses the texture block partitioning scheme and the unilateral motion estimation for improving ME accuracy. Finally, motion compensated frame interpolation (MCFI) are adopted to interpolate the intermediate frame in which MCFI is employed adaptively based on ME scheme. Experimental results show that the proposed algorithm improves the PSNR up to 3dB, the SSIM up to 0.07 and 68% lower SAD calculations compared to the EBME and the conventional FRUC algorithms.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.1
• /
• pp.142-149
• /
• 2011

This paper presents a new state of charge(SOC) estimation of large capacity of Li-ion battery (LIB) based on the multiple model adaptive estimation(MMAE) method. We first introduce an equivalent circuit model of LIB. The relationship between the terminal voltage and the open circuit voltage(OCV) is nonlinear and may vary depending on the changes of temperature and C-rate. In this paper, such behaviors are described as a set of multiple linear time invariant impedance models. Each model is identified at a temperature and a C-rate. These model set must be obtained a priori for a given LIB. It is shown that most of impedances can be modeled by first-order and second-order transfer functions. For the real time estimation, we transform the continuous time models into difference equations. Subsequently, we construct the model banks in the manner that each bank consists of four adjacent models. When an operating point of cell temperature and current is given, the corresponding model bank is directly determined so that it is included in the interval generated by four operating points of the model bank. The MMAE of SOC at an arbitrary operating point (T $^{\circ}C$, $I_{bat}$[A]) is performed by calculating a linear combination of voltage drops, which are obtained by four models of the selected model bank. The demonstration of the proposed method is shown through simulations using DUALFOIL.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.11C
• /
• pp.1498-1509
• /
• 2004

In this paper, we evaluate the performance of a noncoherent OOK (On-Off Keying) UWB (Ultra Wide Band) system based on power detection with noise power calibration and noise power windowing for ubiquitous sensor network applications in typical indoor wireless channels. Utilizing noise power calibration and noise power windowing, the current noise information can be initially or adaptively provided to determine suitable detection threshold value for signal demodulation. Simulation results show that the noncoherent OOK UWB system using noise power calibration achieves good BER (Bit Error Rate) performance which is favorably comparable to that of the system using the ideal adaptive threshold, while maintaining simple receiver structure. However, despite the serious loss of the data transmission rate, the performance improvement by noise power windowing is not so remarkable. furthermore, these performance results are similarly maintained in BEE 802.15 Task Group 3a UWB indoor channel model, and it is also revealed that the BER performance can be significantly improved by increasing the pulse repetition rate.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.1B
• /
• pp.77-88
• /
• 2000

This paper presents an approach for rate control by adaptively selecting macroblock modes in video coding.The problem of rate control has been investigated by many authors where quantizer level is adjustedbased on the buffer fullness. The proposed approach is different fron the previous ones [4] id that it finds the optimal decision curve rather than finding a set of the modes. Proposed algorithm extends the coding decision options for rate control to motion/no-motion compensation as well as inter/intra decisions. Instead of having a fixed motion/no-notion compensation or inter/intra decision curve, one can utilize an adaptive decision curvebased on the characteristics of input frames so that the PSNR at a given bit rate is maximized. Therefore, the proposed approach provides better rate control than simple quantizer feedback approach interns of visual quality. The curve is obtained by utilizing simulated annealing optimization technique. Thealgorithm is implemented and simulations are compared with other approaches within H.261 video codec.

• Smart Media Journal
• /
• v.2 no.4
• /
• pp.51-59
• /
• 2013

With increasing transmission speed of backbone networks, it is getting to provide enough available bandwidth. However, the bandwidth is not effectively utilized in volumetric data transfer. This mainly comes from the transmission protocol, TCP, which is used for most applications. TCP is inherently difficult to adapt the available bandwidth because of it's own characteristic of transfer mechanism. UDT is a prominent application level data transfer protocol which is targeting high speed network. In this paper, we propose UDT parallel transfer technologies which is adaptive to network status and then evaluate their performance in two points of view. Firstly, we measure data transfer rate of UDT with rate congestion control methods, and compare them with basic UDT. Secondly, we apply parallel transfer technologies adapted to network status, and measure their performance. Experimental results showed that UDT rate congestion control method outperforms UDT with 106% improvement in RTT 100ms section set with jitter 30ms. In addition, performance of parallel transfer with rate congestion control method showed 107% improvement than that of parallel transfer in RTT 400ms section set with jitter 20ms.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.57-62
• /
• 2004

This paper deals with a waypoint trajectory following problem for the tilt-rotor UAV under development in Korea (TR-KUAV). In this problem, dynamic model inversion based on the linearized model and Sigma-Phi neural network with adaptive weight update are involved to realize the waypoint following algorithm for the vehicle in the helicopter flight mode (nacelle angle=0 deg). This algorithms consists of two main parts: outer-loop system as a command generator and inner-loop system as stabilizing controller. In this waypoint following problem, the position information in the inertial axis is given to the outer-loop system. From this information, Attitude Command/Attitude Hold logic in the longitudinal channel and Rate Command/Attitude Hold logic in the lateral channel are realized in the inner-loop part of the overall structure of the waypoint following algorithm. The nonlinear simulation based on the TR-KUAV is carried out to evaluate the stability and performance of the algorithm. From the numerical simulation results, the algorithm shows very good tracking performance of passing the waypoints given. Especially, it is observed that ACAH/RCAH logic in the inner-loop has the satisfactory performance due to adaptive neural network in spite of the model error coming from the linear model based inversion.