• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.1394-1397
• /
• 1993

A Fuzzy Microprocessor(FMP) is presented, which is suitable for real-time control applications. The features include high speed inference of maximum 114K FLIPS at 20MHz system clocks, capability of up to 128-rule construction, and handing of 8 input variables with 8-bit resolution. In order to realize these features, the fuzzifier circuit and the processing element(PE) are well optimized for LSI implementation. The chip fabricated in 1.2$\mu\textrm{m}$ CMOS technology contains 71K transistors in 82.8 $\textrm{mm}^2$ die size and is packaged in 100-pin plastic QFP.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.2 no.2
• /
• pp.79-85
• /
• 2009

This research study and design, implement the digital multi-media description technique of Mashup Web which will be evolved into visible visual structure in form of cooperation studying. This technique is study user connect location information and service using pattern, then this information description technique of Mashup pre-operate inference data by tracking user service data.

• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.6
• /
• pp.37-44
• /
• 1995

Grinding is used for machining high precision parts with high additional value. However, the grinding operation needs high skill and long experience of an operator because of a lack of the scientific knowledge and engineering principles. Also, the wheel and grinding conditions affect grinding results. For these reasons, it is difficult to construct computer integrated manufacturing system(CIMA). Therefore, it is necessary for Expert System to be informed of qualitative knowledge of grinding expert's skills and experiences. In this research, the Grinding Expert System is constructed by Fuzzy Decision Making Algorithm. Using this system, unskilled workers will be able to use the knowledge and experience of an expert.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.3
• /
• pp.50-56
• /
• 2001

As the fuzzy control is applied to servo system, the hardware implementation of the fuzzy information systems requires the high speed operations, short real time control and the small size systems. The aims of this study is to develop hardware of the fuzzy information systems to be apply to servo system. In this paper, we propose a calculation method of approximate reasoning for fuzzy control based on $\alpha$ -level set decomposition of fuzzy sets by quantize $\alpha$ -cuts. This method can be easily implemented with analog hardware. The influence of quantization Bevels of $\alpha$-cuts on output from fuzzy inference engine is investigated. It is concluded that 4 quantization levels give sufficient result for fuzzy control performance of dc servo system. The hardware implementation of proposed operation method and of the defuzzification by gravity center method which is directly converted to PWM actuating signal is also presented. It is verified useful with experiment for dc servo system.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2000.05a
• /
• pp.125-134
• /
• 2000

In this paper, we proposed a new circuit construction method that reduced the number of CMOS devices of singleton fuzzy controller(SFC) through the proposing a new membership function circuit(MFC) which uses the language variable selecting and the coefficient selecting circuit. According to the range of input values, we can choose the language variables beforehand which will be used in the inference. So we proposed the new MFC which generates the only necessary language variables. Also, we removed all rules of which adapting degree of their antecedents is zero through proposing the coefficient selecting circuit which beforehand selects the coefficients which will influence the inference result. Though this method, we simplified the structure of SFC and reduced the size of hardware. And to solve the problem in the current mode with respect to the restriction of the fan-out number, voltage-input and current-out membership function circuits are constituted of operational transconductance amplifiers. A membership function circuit which includes the language variable selecting circuit, a minimum operation circuit we implemented by current mode CMOS devices. As a result of applying proposed method, total numbers of blocks and devices wave decreased. If the number of variables and antecedents are getting larger, this method is more efficient.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2000.11a
• /
• pp.357-365
• /
• 2000

In this paper, we proposed a new circuit construction method that reduced the number of CMOS devices of singleton fuzzy controller(SFC) through the proposing a new membership function circuit(MFC) which uses the language variable selecting and the coefficient selecting circuit. According to the range of input values, we can choose the language variables beforehand which will be used in the inference. So we proposed the new MFC which generates the only necessary language variables. Also, we removed all rules of which adapting degree of their antecedents is zero through proposing the coefficient selecting circuit which beforehand selects the coefficients which will influence the inference result. Though this method, we simplified the structure of SFC and reduced the size of hardware. And to solve the problem in the current mode with respect to the restriction of the fan-out number, voltage-input and current-out membership function circuits are constituted of operational transconductance amplifiers. A membership function circuit which includes the language variable selecting circuit, a minimum operation circuit we implemented by current mode CMOS devices. As a result of applying proposed method, total numbers of blocks and devices wave decreased. If the number of variables and antecedents are getting larger, this method is more efficient.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.3
• /
• pp.169-177
• /
• 2008

Grid resources are composed of various communication networks and operation systems. When a grid system searches and selects grid resources, which meet requirements of a grid user, existing grid resource selection systems are limited due to their storage methods for resource information. In order to select grid resources suitable for requirements of a grid user and characteristics of data, this paper constructs an ontology for grid resources and proposes an ontology-based grid resource selection system. This system provides an inference engine based on rules defined by SWRL to create a resource list. Experimental results comparing the proposed system with existing grid resource selection systems, such as the Condor-G and the Nimrod-G, verify the effectiveness of the ontology-based grid resource selection system with improved job throughput and resource utilization and reduced job loss and job processing time.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.10
• /
• pp.1403-1408
• /
• 2021

Recently, FPGA-based AI processors are being studied actively. Deep convolutional neural networks (CNN) are basic computational structures performed by AI processors and require a very large amount of multiplication. Considering that the multiplication coefficients used in CNN inference operation are all constants and that an FPGA is easy to design a multiplier tailored to a specific coefficient, this paper proposes a methodology to optimize the multiplier. The method utilizes 2's complement and distributive law to minimize the number of bits with a value of 1 in a multiplication coefficient, and thereby reduces the number of required stacked adders. As a result of applying this method to the actual example of implementing CNN in FPGA, the logic usage is reduced by up to 30.2% and the propagation delay is also reduced by up to 22%. Even when implemented with an ASIC chip, the hardware area is reduced by up to 35% and the delay is reduced by up to 19.2%.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.17 no.3
• /
• pp.69-80
• /
• 2007

To protect personal information when releasing data, a general privacy-protecting technique is the removal of all the explicit identifiers, such as names and social security numbers. De-identifying data, however, provides no guarantee of anonymity because released information can be linked to publicly available information to identify them and to infer information that was not intended for release. In recent years, two emerging concepts in personal information protection are k-anonymity and $\ell$-diversity, which guarantees privacy against homogeneity and background knowledge attacks. While these solutions are signigicant in static data environment, they are insufficient in dynamic environments because of vulnerability to inference. Specially, the problem appeared in record deletion is to deconstruct the k-anonymity and $\ell$-diversity. In this paper, we present an approach to securely anonymizing a continuously changeable dataset in an efficient manner while assuring high data quality.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.3
• /
• pp.419-432
• /
• 2019

Most of deep learning model training was proceeded by supervised learning, which is to train labeling data composed by inputs and corresponding outputs. Labeling data was directly generated manually, so labeling accuracy of data is relatively high. However, it requires heavy efforts in securing data because of cost and time. Additionally, the main goal of supervised learning is to improve detection performance for 'True Positive' data but not to reduce occurrence of 'False Positive' data. In this paper, the occurrence of unpredictable 'False Positive' appears by trained modes with labeling data and 'True Positive' data in monitoring of deep learning-based CCTV accident detection system, which is under operation at a tunnel monitoring center. Those types of 'False Positive' to 'fire' or 'person' objects were frequently taking place for lights of working vehicle, reflecting sunlight at tunnel entrance, long black feature which occurs to the part of lane or car, etc. To solve this problem, a deep learning model was developed by simultaneously training the 'False Positive' data generated in the field and the labeling data. As a result, in comparison with the model that was trained only by the existing labeling data, the re-inference performance with respect to the labeling data was improved. In addition, re-inference of the 'False Positive' data shows that the number of 'False Positive' for the persons were more reduced in case of training model including many 'False Positive' data. By training of the 'False Positive' data, the capability of field application of the deep learning model was improved automatically.