Go to the main menu
Skip to content
Go to bottom
REFERENCE LINKING PLATFORM OF KOREA S&T JOURNALS
> Journal Vol & Issue
The Journal of Korean Institute of Communications and Information Sciences
Journal Basic Information
Journal DOI :
The Korean Institute of Communications and Information Sciences
Editor in Chief :
Volume & Issues
Volume 24, Issue 12B - Dec 1999
Volume 24, Issue 12A - Dec 1999
Volume 24, Issue 11A - Nov 1999
Volume 24, Issue 10B - Oct 1999
Volume 24, Issue 10A - Oct 1999
Volume 24, Issue 9B - Sep 1999
Volume 24, Issue 9A - Sep 1999
Volume 24, Issue 8B - Aug 1999
Volume 24, Issue 8A - Aug 1999
Volume 24, Issue 7B - Jul 1999
Volume 24, Issue 7A - Jul 1999
Volume 24, Issue 6B - Jun 1999
Volume 24, Issue 6A - Jun 1999
Volume 24, Issue 5B - 00 1999
Volume 24, Issue 5A - 00 1999
Volume 24, Issue 4B - 00 1999
Volume 24, Issue 4A - 00 1999
Volume 24, Issue 3B - 00 1999
Volume 24, Issue 3A - 00 1999
Volume 24, Issue 2B - 00 1999
Volume 24, Issue 2A - 00 1999
Volume 24, Issue 1B - 00 1999
Volume 24, Issue 1A - 00 1999
Volume 24, Issue 12 - 00 1999
Volume 24, Issue 11B - 00 1999
Selecting the target year
Strategies for Rule Execution Considering Parallel Processing in Active Database
O, Seong Gyun ; Lee, Jae Ho ; Im, Hae Cheol ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 777~777
Various real-world semantics have not been fully provided due to the simplicity of conventional parallel control in parallel rule execution model of Active Database. This paper extends to represent correctly the complex situations in real world environment and to analyse the necessary constituent and the form of rule execution. Based on these control constructs, we present graph-based rule execution graph considering real world situation. And, we propose an algorithm and the policy which create priority trigger group according to the number of nested rule and other information for the parallel execution of trigger rule. In case that multi-processors are assigned to the large number of trigger rules, the policy is suggested to assign and to release processor in multi-processors environment based on trigger group priority policy. Performance analysis is performed for the proposed rule execution strategy compared with conventional execution strategy in terms of the processing time.
An Implementation of Generic Conference Control Protocol in the Mulipoint Control Unit
Jang, Seung Seok ; Lee, Jong Hyeong ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 788~788
This paper describes an implementation of Generic Conference Control(GCC) protocol in the ATM based Multipoint Control Unit(MCU) which is required for providing real-time, high-quality, multipoint video for multipoint conference creation, control and termination, and supports real-time collaborative works. In this paper, a conference is created locally at the MCU, and clients call into the MCU and join the conference, or called an invited to the conference by the MCU. Clients can disconnect from the conference but only the MCU can terminate an entire conference. We have implemented the GCC protocol as Win32 Dynamic Link Library(DLL), ported it in the MCU host, and proved its functionality by testing with Microsoft NetMeeting program under ATM network environment.
Architecture of processing of flow control instruction capable of prefetching of return address
Park, Ju Hyeon ; Kim, Yeong Min ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 797~797
In this paper, we propose a method to reduce overhead of processing of flow control instruction in the pipeline architecture by performing pre-fetching in the initial stage of pipeline processing. The basis of pipeline processing is a four-stage configuration. In a program routine with subroutine call and return instructions, a pre-return processing module reduces pipeline processing time by 25% as compared to the conventional processing architecture. Each module of the proposed architecture of processing of flow control instruction is described in VHDL structurally and behaviorally and whether it is working well or not is checked on V-system simulator for model Technology Inc. by processing a simple program with a call, a jump and a return instruction.
Individual Cell Loss Analysis in an ATM Multiplexer with Heterogenous Input Sources
Choe, U Yong ; Kim, Ji Su ; Jeon, Chi Hyeok ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 805~805
Implementation of Interworking Function Between N-ISN and B-ISDN at TDX-10A Electronic Switching System
Kim, Gyu Hwan ; Kim, Yeong Ro ; Lee, Bong Yeong ; Go, Seong Je ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 813~813
We propose a network interworking scheme where the N-ISDN switching systems can accept B-ISDN subscribers. In the proposal scheme, network interworking can be accomplished by adding the Interworking Function(IWF) to the transit-call processing unit of the TDX-10A Electronic Switching System. The upper layer of the interworking protocol converts outgoing and incoming call messages into the desired form of ISUP messages, while the lower layer transmits the data. B-ISUP messages are converted into N-ISUP messages by using mutually compatible parameters. The message mapping test has been performed at the message conversion part which converts B-ISDN to N-ISDN messages using parameters. The test results show that N-IAM messages from the N-ISDN switching system are converted to B-IAM messages at the message conversion part of the interworking switching system.
Digital information transmission algorithm in the frequency hopping communication system
Han, Seong U ; Kim, Yong Seon ; Jeon, Yong Eok ; Jeon, Byeong Min ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 821~821
Frequency Hopping(FH) communication could avoid the interference against enemy and transmit the information under worst condition, Because the Wireless communication condition is severer than the wire’s, and informations could be interfered and lost by jammer, we need the method to transmit the digital information for FH communication. This paper includes the methods to transmit voice or data on FH communication. In transmitter, low rate data are coded to ther majority data blocks, and transmitted repeatedly through FH channel, which are structured to 20Kbps hopping frame cells, In the receiver, the received data are decoded block by block, and taken majority decoding, consequently, channel errors are corrected at the block interleaver and the majority decoder. This algorithm can reduce the coding/decoding time and simplify the data processing method for FH communication. Besides, we reduced the jamming influence and improved the performance of data transmission on FH channel, and proved the performance in the real FH system.
Performance Analysis of Dynamic Multicast Routing Algorithm for High-Speed Optical Network Based Multimedia Communications
Gang, Mun Sik ; Lee, Sang Min ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 829~829
In this paper, a dynamic multicast routing algorithm(MRA) is presented for multimedia communication on high-speed optical networks. The proposed algorithm is based on multicast routing tree construction method, but is designed to satisfy such network requirements as efficient dynamic group support, high-quality data distribution, and dynamic adaptability to the changing network events. Design parameters such as acceptable delay limit and maximum group size are determined based on the consideration of network states and requirements on delay and cost. Also, the delay-bounded optimal dynamic multicast tree is analyzed by running simulations on randomly generated test networks. The simulation results show that the proposed algorithm produces the optimal solution which, under the given delay constraint, minimizes the network cost and adapts dynamically to the change in multicast group members.
A Study on the Output Queuing control scheme for QoS Improvement in ATM Switch
Lee, Sang Tae ; Kim, Nam Hui ; No, Gi Yong ; Jeon, Byeong Sil ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 839~839
In this paper, we propose a new cell discarding and scheduling scheme which reduce cell loss rate by measuring, in real time, the number of discarded cells in the queuing system with a different loss priority for each class of service such that each class of service meets its cell loss late requirements and reduce average delay rate for the traffic that is sensitive in average delay rate in output buffer of the ATM switch. In proposed scheme, we use three address buffers, one common buffer, and idle address buffer which is control the address in common buffer for improving the buffer utility and delay requirements can be reduced by the modified existing cell scheduling scheme. Throughout the computer simulation, the existing scheduling scheme and proposed scheme are compared with respect to cell loss rate and average delay time. And, in proposed scheme, we compare cell discarding algorithm I with algorithm II in cell loss rate and average delay time.
Performance Analysis and Design of Database Interworking Architecture for Supporting UPT Service in IMT-2000 Satellite System
Lee, Dong Hui ; Park, Jong Tae ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 850~850
International mobile telecommunication 2000(IMT-2000) satellite system is a next generation mobile telecommunication system, which provides terminal mobility such as roaming service. Universal personal telecommunication(UPT) supports personal mobility, which provides service by personal number through registration procedure in service network. In this paper, we have taken performance analysis and comparison of network architecture for supporting UPT service in IMT-2000 satellite system. In detail, we designed the physical network architecture of Imt-2000 satellite system and database interworking methods. Additionally, we designed signaling procedures for UPT call setup and analyzed the call setup delay performance of the database interworking methods by using time delay modeling.
UML based modeling for the behavior and schedulablity analysis of autonomous soccer robots
Lee, Jae Ik ; Gang, Sun Ju ; Seo, Dae Hwa ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 858~858
As real-time embedded systems become more complex, the reactive behavior of the real-time systems is hard to implement using ad-hoc techniques. State machine based approaches are naturally suited for modeling reactive behavior of real-time systems, thus several object-oriented development methodologies are attractive as a potential development methodology for real-time systems. However these methodologies do not provide any mechanism to specify and enforce timing constraints that are required for schedulability analysis of hard real-time systems. In this paper, we show UML, adapted as standard object-oriented development by OMG, based modeling for autonomous soccer robot as a case study of a hard real-time system, and show how the proposed real time scheduling theory may be applied to the UML based model.
Multiwavelength Optical Orthogonal Code Design for Optical CDMA
Lee, Ssang Su ; Seo, Seung U ; O, Byeong Gwon ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 869~869
In this paper, we proposed an extended construction method for Multiwavelength optical orthogonal codes (MWOOC’s)Which is applied to Optical CDMA. MWOOC’s are two-dimensional codes consisting of (0,1) non-negative pulses with the additional degree of freedom caused by multiwavelengths.Given available wavelengths m, codeword length n, constant Hamming weight ω and correlation value λ, we present a generalized methodology for constructing (mn, ω, λ) MWOOC’s.Especially, we construct (mn, ω, 1) MWOOC’s when ω is a prime number or 4. The proposed construction method can be used without any constraint on the relationship between the number of available wavelengths and codeword length. We investigate the cardinality, the optimality, and the bit error rate of our method by comparing with other optical orthogonal codes with/without multiwavelengths. It is shown that the constructed MWOOC’s enable much more users to transmit their data on a common fiber optic asynchronously at a lower bit error rate than previous optical orthogonal codes without multiwavelengths(OOC’s).
A Study on the Improvement Methods of Propagation Impediment at Tollgate
Kim, Gwang Su ; Kim, Jong Seop ; Yun, Tae Seok ; Sin, Seung Hun ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 878~878
Receiving the radio signal at the tollgate, the mobile unit undergoes signal attenuation due to the tollgate structure. Performing measurements, we have investigated the propagation impediment of radio waves at the tollgate, and proposed the improvement methods. From the measurement results, it seems to us that the propagation impediment at tollgate is due to shielding from the metallic structure of tollgate rather than electromagnetic interference from electronic devices. Introducing a noise isolator, automatic gain controller and delay line designed to eliminate the feedback effect which may occur from receiving antenna to transmitting antenna, signal degradation at the tollgate can be reduced.
Object Based Fiscal Image Coding
Gang, Hyeon Su ; Kim, Seong Dae ; Yun, Byeong Ju ; Jo, Sun Je ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 885~885
This paper presents an object based fractal coding scheme based on PIFS(partitioned iterative function system) like the standard fractal coding schemes for images. The partitioned blocks in our object based fractal coding scheme are classified to boundary blocks and internal blocks according to their shapes. The boundary blocks are located in the boundary region of an object and their shapes are arbitrary one, whereas the internal blocks are located in the internal region of an object and their shapes are square one. As the internal blocks make no difference with the blocks to appear in the process of the ordinary fractal coding expect that domain pool is not an image but an object, encoding for the internal blocks follows the ordinary fractal coding schemes as they are. However, the boundary blocks should be carefully handled, as they have properties to be different from the internal blocks, i.e., they have arbitrary shapes and their domain pool is composed of the internal blocks not the boundary blocks. Considering these properties, we propose 3 methods to encode the boundary blocks by making an alteration to the shapes of range blocks and domain blocks, and a method to do them by using the pre-decoded internal blocks. The methods are mathematically analyzed and their performance is experimentally evaluated.
Optimal Significance Testing for Partially Supervised Classification
Jeon, Byeong U ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 900~900
Significance testing is one of the most widely used techniques in various applications of statistical analysis. We apply significance testing to the partially supervised classification problem in which one is interested in identifying only a particular class of interest, based on the class definition through training samples of that class. One important element in applying significance testing to classification is the significance level, which should be provided by the data analyst such a way that omission or type I error is limited at a pre-specified level. This paper addresses the problem of unsupervised estimation of an optimal significance level using the class-averaged and generalize total classification error criteria, and applies its result to partially supervised classification.
A New Error Concealment Method Using Motion Parameter in Block Based Image Coding Schemes
Park, Su Yeol ; Yu, Sang Jo ; Hong, Seong Hun ; Kim, Seong Dae ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 907~907
In this paper, we propose a new error concealment method using 6 parameter affine transform in the block based image coding schemes such as H.236, H.261, and MPEG-2 without modifying the encoder structure and bit stream syntax. This method can be used for temporal direction error concealment of visual information under the channel environment with high BER and low bandwidth, especially mobile networks. For extracting the exact 6 parameter motion information in the decoder side, we partition believable adjacent macro blocks of lost macro blocks 4x4 blocks, and re-search the 2-D motion vectors, the proposed method provides the effective error concealment in the low frame rate visual communications and when the compensation error using 2-D motion vector is large.
Adaptive Quantization Algorithm for Video Coding
Jo, Nam Ik ; Lee, Sang Yong ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 919~919
This paper proposes adaptive quantization algorithms for video coding using the information obtained from the previously encoded image. Before quantizing the DCT coefficients, the properties of reconstruction error of each macro block(MB) is estimated from the previous frame. For the estimation of the error of the current MB, a block with the size of MB in the previous frame is chosen. Since the original and reconstructed images of the previous frame are available in the encoder, we can evaluate the tendency of reconstruction error of this block in advance. Then, this error is considered as the expected error of the current MB. Comparing the error of the MB with the average of overall MBs, if it is larger than the average, a small step size is given for this MB, and vice versa. As a result, the error distribution of the MB is more concentrated to the average, yielding low variance and improved image quality. For reducing the blocking artifact, the error measure for the blocking artifact is also considered as well as SAD(sum of absolute difference) and PSNR. Especially for low bit application, the proposed algorithm yields much smaller error variance and higher PSNR compared to the conventional TM5. We also propose a modified algorithm for efficient hardware implementation.
Adaptive motion estimation technique for motion compensated interframe interpolation
Seong, Won Rak ; Gang, Eung Gwan ; Kim, Dong Uk ; Choe, Jong Su ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 932~932
Interframe interpolation is the technique that inserts one or more frames between two consecutive frames in video sequence. Conventional motion compensated interframe interpolation algorithm requires additional computation because it does new search procedure at the receiver for region prediction needed for interframe interpolation. Also, the algorithm causes severe block effect because it interpolates the skipped frames by using bloc based motion vectors. Therefore, in this paper, we propose the new algorithms that reduce the computational complexity by the excessive search at the receiver and the block effect in the interpolated frames.
Adaptive Image Interpolation Method Using Wavelet Transform
Kim, Tae Hyeon ; Jeong, Jin Ho ; Go, Seong Je ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 940~940
In order to reconstruct pixels in the lost object areas of a still image, conventional interpolation methods utilize the weighted average of values of the closest neighbor pixels. However, this linear interpolation scheme cannot restore details and sharp edges of the lost image object. In this paper, an Image interpolation method using wavelet transform is proposed to restore the lost object area. The proposed, method uses the characteristics of wavelet transformed subbands and applies individual interpolation schemes to the subbands. In the low-frequency subband, the lost areas are reconstructed by an adaptive interpolation method using neighbor pixels selected according to the estimated edge structures. The high-frequency subbands are restored by horizontal and vertical interpolation methods using edge orientations. Experimental results show that the proposed interpolation method performs better than conventional linear interpolation methods.
An Efficient Compression Technique for Affine Motion Model
Choe, Jae Gak ; Go, Yun Ho ; Kim, Seong Dae ; No, Cheol Gyun ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 949~949
In this paper, we describe an efficient compression technique for affine motion. Because the estimated parameters of affine motion model are real-valued, we have to quantize them before the transmission. Also, there are poor correlation and similarity between motion parameters of neighboring objects, and each parameters has different statistical properties and dynamic range. To overcome these problems, a new method is proposed in this paper. The core of this scheme includes the calculation of real translation motion and the line transformation of scaling and rotation motions. By the calculation of real translation, translation terms which have good statistical properties and strong temporal correlation are generlated. They are differentially encoded in the temporal direction. By the line transformation, the scaling and the rotation motions are converted into 2D vector elements which have a good correlation. Then, they are encoded using vector quantization. Simulation results show that our proposed scheme provides better performance than the direct quantization method.
MPEG-2 Traffic Shaping for Real Time Video Services on ATM Network-“Intra Slice Coding
Jang, Seung Gi ; Seo, Deok Yeong ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 959~959
ATM(asynchronous transfer mode, the transfer mode of B-ISDN(broadband-integrated service digital networks) is so flexible to support various kinds of traffic classes including VBR(variable bit rate) traffic. VBR traffic class is useful for delay-sensitive application such as video phone and teleconferencing. Video bit stream encoded by MPEG-2(moving picture experts group) encoder is variable in bit rate if it is not buffered. Most network other than ATM network require constant bit rate. It caused delay and jitter or channel waste to transport traffic with variation in bit-rate through a CBR(constant bit rate) channel. Even for ATM, it is desirable to reduce such variation. This paper shows that the proposed traffic shaping algorithm using the intra slice coding mode in MPEG-2 can be useful to smooth variation in MPEG-w video traffic without inducing any buffering delay, and jitter. The less is variation in bit rate of a session, the less communication resources are allocated to the session. For quantitative analysis of this effect, GCRA(generic cellrate algorithm) is used, GCA is an algorithm popular for an ATM switch to police users.
Calibration of a Thin Film Micromirror Array Inspection
Lee, Eun Suk ; Lee, Byeong Uk ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 968~968
TMA(Thin-film Micromirror Array) is under development as a next generation display due to its merits of high brightness at large screen size. Calibration for TMA mirror variation will enhance the image quality. We propose a geometric calibration method which has subpixel accuracy. We measured the gamma of the CCD camera to calculate the optical power from the camera output. We measured the camera image noise and size of the bits of the calibration factor was calculated based on the image noise. The calibration method developed in this paper will be applied to the TMA inspection system.
Design of a Robot-Tracking System using Neural Networks
Baek, Seong Hyeon ; Park, Dong Seon ; Jo, Jae Wan ; Lee, Yong Beom ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 977~977
In this paper, a neural network robot tracking system is designed and partly implemented. The neural network approach is employed to recognize the robot endeffector covering the situation of three types of motion: translation, scaling and rotation. Features for the neural network to detect the position o the endeffector are extracted from preprocessed images. A feedforward neural network with the backpropagation learning is used to detect the position of the robot endeffector. Another neural network is used to estimate the motion from a sequence of images and to control movements of the robot endeffector. Inherent characteristics of neural networks include excellent object recognition capability, the robustness to noise and the generalization capability. Combining the two neural networks for recognizing the robot endeffector and estimation the motion with the preprocessing stage, the whole system keeps tracking of the robot endeffector effectively.
Blind Array Beamformer using Cyclostationary Property
Park, Sang Yong ; Kim, Gi Man ;
The Journal of Korean Institute of Communications and Information Sciences, volume 24, issue 5B, 1999, Pages 989~989
In recent years, in order to increase the capacity and to suppress co-channel interference in mobile cellular and satellite communication systems, various array beamforming techniques have been proposed. In this paper, we propose blind array beamformer. The Beamformer’s optimum weight vector is calculated from auto-correlation matrix of the received signals and cross-correlation matrix between the received signal vector which shifted to the cyclic frequency of the desired signal vector. The proposed method utilizes the cyclostationary properties of the communication signal, and has the constraints that maximize output SNR(Signal-to-Noise Ratio).We performed computer simulations to compare the performances between the proposed method and the conventional methods.