• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.6
• /
• pp.44-51
• /
• 1999

The HMM-Net is a neural network architecture that implements the computation of output probabilities of a hidden Markov model (HMM). The architecture is developed for the purpose of combining the discriminant power of neural networks with the time-domain modeling capability of HMMs. Criteria of maximum likehood(ML) and minimization of mean squared error(MMSE) are used for learning HMM-Net classifiers. The criterion MMSE is better than ML when initial learning condition is well established. However Ml is more useful one when the condition is incomplete[3]. Therefore we propose an efficient learning method of HMM-Net classifiers using a hybrid criterion(ML/MMSE). In the method, we begin a learning with ML in order to get a stable start-point. After then, we continue the learning with MMSE to search an optimal or near-optimal solution. Experimental results for the isolated numeric digits from /0/ to /9/, a training and testing time-series pattern set, show that the performance of the proposed method is better than the others in the respects of learning and recognition rates.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.6
• /
• pp.108-114
• /
• 1989

Introducing and optimizing bit-significance to the Hopfield model, ten highly correlated binary images, i.e., numbers "0" to "9", are successfully stored and retrieved in a $6{}8$ node system. Unlike many other neural network models, this model has stronger error correction capability for correlated images such as "6","8","3", and "9". The bit significance optimization is regarded as an adaptive learning process based on least-mean-square error algorithm, and may be implemented with Widrow-Hoff neural nets optimizer. A design for electro-optic implementation including the adaptive optimization networks is also introduced.

• ETRI Journal
• /
• v.21 no.4
• /
• pp.52-64
• /
• 1999

QoS monitoring is a kind of real-time systems which allows each level of the system to track the ongoing QoS levels achieved by the lower network layers. For these systems, real-time communications between corresponding transport protocol objects is essential for their correct behavior. When two or more entities are employed to perform a certain task as in the case of communication protocols, the capability to do so is called interoperability and considered as the essential aspect of correctness of communication systems. This paper describes a formal approach on modeling and interoperability test case generation of a real-time QoS monitoring protocol. For this, we specify the behavior of flow monitoring of transport layer QoS protocol, i.e., METS protocol, which is proposed to address QoS from an end-to-end's point of view, based on QoS architecture model which includes ATM net work in lower layers. We use a real-time Input/Output finite State Machine to model the behavior of real-time flow monitoring over time. From the modeled real-time I/OFSM, we generate interoperability test cases to check the correctness of METS protocol's flow monitoring behaviors for two end systems. A new approach to efficient interoperability testing is described and the method of interoperability test cases generation is shown with the example of METS protocol's flow monitoring. The current TTCN is not appropriate for testing real-time and multimedia systems. Because test events in TTCN are for message-based system and not for stream-based systems, the real-time in TTCN can only be approximated. This paper also proposes the notation of real-time Abstract Test Suite by means of real-time extension of TTCN. This approach gives the advantages that only a few syntactical changes are necessary, and TTCN and real-time TTCN are compatible. This formal approach on interoperability testing can be applied to the real-time protocols related to IMT-2000, B-ISDN and real-time systems.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.11
• /
• pp.1-9
• /
• 1999

As our society turns into the information age from the industrial one, the ministry of information and communication has set up functional software standards for electronic approval systems Several software houses have developed such systems in the client/server environment and subsequently for the intranet. Although electronic approval systems for the intranet have the advantages of less costly implementation and ease of use, they create heavy network traffic, and have a poor document processing functionality resulting from the lack of document processor in web environments. This paper describes a system design that web browsers utilize the resources of clients by adopting the ActiveX technique in order to improve such mallets mentions above. In other words, to use the Hangul word processor as a document processor, the ActiveX control and the Hangul DDE API have been implemented in the form of the DDE server/client, which is capable of mutual communication, and the flow of electronic approve system has been controled by connecting. As a result of running the implemented system lot three months through a real company in multi-server environment, it shows the high usage of electronic approval system as the tate roaches 75%-93% for some departments.

• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.36D no.6
• /
• pp.1-12
• /
• 1999

This paper illustrates and analyzes system structure, link budget and cell coverage to accommodate Korea environment about LMDS systems that are expected to play an important role as the future wireless infrastructure of the information superhighway. In the mean time, many parameters used for the analysis apply to the allocated frequency bandwidth (Up-stream : 24.25GHz-24.75GHz, Down-stream : 25.5GHz-27.5GHz) of Ministry of Information and Communication. The modem and channel coding must be developed to accommodate specifications of DAVIC protocol. The materials about atmospheric phenomena refer to apply specifications of ITU-R. This paper will be need to disign the network structure and system configuration of LMDS system.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.8
• /
• pp.68-76
• /
• 1998

In this paper, a new morphological image segmentation algorithm of hierarchical structure by binary split is proposed. It splits a region with the lowest quality into two regions using only two markers having the highest contrast. Therefore, it improves the quality of image with limited regions and reduces contour information which is not sensitive to human visual system, when compared with the conventional algorithm. It is appropriate to PSTN, LAN, and mobile networks, of which the available transmission bandwidth is very limited, because the number of regions can be controlled. And the proposed algorithm shows very simple structure because it doesn't need post processing to eliminate small regions and reduces much computation by using only structuring element of small size at simplification step of each hierarchical structure when compared with the conventional algorithm.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.6
• /
• pp.81-87
• /
• 1998

This paper proposes an efficient face recognition system using phase difference between the face images. We use a Karhunen-Loeve transform for image compression and reconstruction, and obtain the phase difference by using normalized inner product of the two compressed images. The proposed system is rotation and light-invariant due to using the normalized phase difference, and somewhat shift-invariant due to applying the cosine function. The faster recognition than the conventional system and incremental training is possible in the proposed system. Simulations are conducted on the ORL images of 40 persons, in which each person has 10 facial images, and the result shows that the faster recognition than conventional recognizer using convolution network under the same recognition error rate of 8% does.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.11
• /
• pp.76-88
• /
• 1998

While most of the research on reinforcement learning assumed a discrete control space, many of the real world control problems need to have continuous output. This can be achieved by using continuous mapping functions for the value and action functions of the reinforcement learning architecture. Two questions arise here however. One is what sort of function representation to use and the other is how to determine the amount of noise for search in action space. The ubiquitous neural network is used here to learn the value and policy functions. Next, the reinforcement predictor that is intended to predict the next reinforcement is introduced that also determines the amount of noise to add to the controller output. The proposed reinforcement learning architecture is found to have a sound on-line learning control performance especially at high-speed road following of high curvature road. Both computer simulation and actual experiments on a test vehicle have been performed and their efficiency and effectiveness has been verified.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.12
• /
• pp.51-57
• /
• 1999

In this paper, an adaptive thresholding of the nonuniformly contrasted images is proposed through using the contrast pre-enhancement of the local regions and the bilinear interpolation between the local threshold values. The nonuniformly contrasted image is decomposed into 9${\times}$9 sized local regions, and the contrast is enhanced by intensifying the gray level difference of each low contrasted or blurred region. Optimal threshold values are obtained by iterative method from the gray level distribution of each contrast-enhanced local region. Discontinuities are reduced at the region of interest or at the characters by using bilinear interpolation between the neighboring threshold surfaces. Character recognition experiments are conducted using backpropagation neural network on the characters extracted from the nonuniformly contrasted document, PCB, and wafer images binarized through using the proposed thresholding and the conventional thresholding methods, and the results prove the relative effectiveness of the proposed scheme.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.4
• /
• pp.31-43
• /
• 1999

The main goal of this paper is to design a high performance SVP(Stack based Video Processor) for network applications. The SVP is a comprehensive scheme; 'better' in the sense that it is an optimal selection of previously proposed enhancements of a stack machine and a video processor. This can process effectively object-based video data using a S-RISC(Stack-based Reduced Instruction Set Computer) with a semi -general-purpose architecture having a stack buffer for OOP(Object-Oriented Programming) with many small procedures at running programs. And it includes a vector processor that can improve the MPEG coding speed. The vector processor in the SVP can execute advanced mode motion compensation, motion prediction by half pixel and SA-DCT(Shape Adaptive-Discrete Cosine Transform) of MPEG-4. Absolutors and halfers in the vector processor make this architecture extensive to a encoder. We also designed a VLSI stack-oriented video processor using the proposed architecture of stack-oriented video decoding. It was designed with O.5$\mu\textrm{m}$ 3LM standard-cell technology, and has 110K logic gates and 12 Kbits SRAM internal buffer. The operating frequency is 50MHz. This executes algorithms of video decoding for QCIF 15fps(frame per second), maximum rate of VLBV(Very Low Bitrate Video) in MPEG-4.