• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.4
• /
• pp.49-58
• /
• 2014

A low-swing differential near-ground signaling (NGS) transceiver for low-power high-speed mobile I/O interface is presented. The proposed transmitter adopts an on-chip regulated programmable-swing voltage-mode driver and a pre-driver with asymmetric rising/falling time. The proposed receiver utilizes a new multiple gain-path differential amplifier with feed-forward capacitors that boost high-frequency gain. Also, the receiver incorporates a new adaptive bias generator to compensate the input common-mode variation due to the variable output swing of the transmitter and to minimize the current mismatch of the receiver's input stage amplifier. The use of the new simple and effective impedance matching techniques applied in the transmitter and receiver results in good signal integrity and high power efficiency. The proposed transceiver designed in a 65-nm CMOS technology achieves a data rate of 13 Gbps/channel and 0.3 pJ/bit (= 0.3 mW/Gbps) high power efficiency over a 10 cm FR4 printed circuit board.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.38 no.3
• /
• pp.1-12
• /
• 2001

In this paper, we investigate and analyze weight distribution of feed forward two-layer neural networks with a hidden layer in order to understand and improve time-consuming training process of neural networks. Generally, when a new problem is presented, neural networks have to be trained again without any benefit from the previous training process. In order to address this problem, training process is viewed as finding a solution point in the weight space and the distribution of solution points is analyzed. Then we propose to initialize neural networks using the information of the distribution of the solution points. Experimental results show that the proposed initialization using the weight distribution provides a better performance than the conventional one.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.31 no.1 s.256
• /
• pp.26-35
• /
• 2007

The vehicle of urban transit is a complex system that consists of various electric, electronic, and mechanical equipments, and the maintenance cost of this complex and large-scale system generally occupies sixty percent of the LCC (Life Cycle Cost). For reasonable establishing of maintenance strategies, safety security and cost limitation must be considered at the same time. The concept of system reliability has been introduced and optimized as the key of reasonable maintenance strategies. For optimization, three preceding studies were accomplished; standardizing a maintenance classification, constructing RBD (Reliability Block Diagram) of VVVF (Variable Voltage Variable Frequency) urban transit, and developing a web based reliability evaluation system. Historical maintenance data in terms of reliability index can be derived from the web based reliability evaluation system. In this paper, we propose applying inverse problem analysis method and hybrid neuro-genetic algorithm to system reliability optimization for using historical maintenance data in database of web based system. Feed-forward multi-layer neural networks trained by back propagation are used to find out the relationship between several component reliability (input) and system reliability (output) of structural system. The inverse problem can be formulated by using neural network. One of the neural network training algorithms, the back propagation algorithm, can attain stable and quick convergence during training process. Genetic algorithm is used to find the minimum square error.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.46 no.3
• /
• pp.51-59
• /
• 2009

This paper presents the design of continuous-time sigma-delta modulator for acoustic sensor. The feedforward structure without summing block is used to reduce power consumption of sigma-delta modulator. A high-linearity active-RC filter is used to improve resolution of sigma-delta modulator. Excess loop delay problem in conventional continuous-time sigma-delta modulators is solved by our proposed architecture. A low power, high resolution fourth-order continuous-time sigma-delta modulator with 1-bit quantization was realized in a 0.13-${\mu}m$ 1-Poly 8-metal CMOS technology, with a core area of $0.58\;mm^2$. Simulation results show that the modulator achieves 91.3-dB SNR over a 25-kHz signal bandwidth with an oversampling ratio of 64, while dissipating $290{\mu}W$ from a 3.3-V supply.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.45 no.2
• /
• pp.28-33
• /
• 2008

This paper presents a continuous-time equalizer adopting a clock loss tracking technique for digital display interface. This technique uses bottom hold circuit to detect the incoming clock loss. The generated loss signal is directly fed to equalizer filters, building adaptive feed-forward loops which contribute the stability of the system. The design was done in $0.18{\mu}m$ CMOS technology. Experimental results summarize that eye-width of minimum 0.7UI is achieved until -33dB channel loss at 1.65Gbps. The average power consumption of the equalizer is a maximum 10mW, a very low value in comparison to those of previous researches, and the effective area is $0.127mm^2$.

• Journal of IKEEE
• /
• v.7 no.1 s.12
• /
• pp.88-96
• /
• 2003

In this paper, We designed and implemented a linear high-power amplifier which can be used for the commercial service in band of $1.9GHz(1.93{\sim}1.99GHz)$ at U.S.A. The output power of the implemented linear high power amplifier is 25W. In order to satisfy IMD characters decided by FCC, the Feedforward linearization techniques has been used. The used feedforward method has improved the IMD characteristics from 10.51dBc to 19.01dBc in each power level from 1W(30dBm) to 25W(44dBm). The IMD level of the final output shows from minimum 64.84dBc to maximum 68.17dBc. Because this good characteristics of IMD, the LPA is expected to use as a commercial product of PCS base station.

• Journal of IKEEE
• /
• v.19 no.1
• /
• pp.87-93
• /
• 2015

In this paper, a $145{\mu}W$, 87dB SNR, Low power 3rd order Sigma-Delta Modulator with Op-amp sharing is proposed. Conventional architecture with analog path and digital path is improved by adding a delayed feed -forward path for disadvantages that coefficient value of the first integrator is small. Proposed architecture has a larger coefficient value of the first integrator to remove the digital path. Power consumption of proposed architecture using op-amp sharing is lower than conventional architecture. Simulation results for the proposed SDM designed in $0.18{\mu}m$ CMOS technology with power supply voltage 1.8V, signal bandwidth 20KHz and sampling frequency 2.8224MHz shows SNR(Signal to Noise Ratio) of 87dB, the power consumption of $145{\mu}W$.

• Journal of IKEEE
• /
• v.14 no.1
• /
• pp.33-39
• /
• 2010

In this paper, a 5Gb/s receiver with an adaptive equalizer for serial link interfaces is proposed. For effective gain control, a least-mean-square (LMS) algorithm was implemented with two internal signals of slicers instead of output node of an equalizing filter. The scheme does not affect on a bandwidth of the equalizing filter. It also can be implemented without passive filter and it saves chip area and power consumption since two internal signals of slicers have a similar DC magnitude. The proposed adaptive equalizer can compensate up to 25dB and operate in various environments, which are 15m shield-twisted pair (STP) cable for DisplayPort and FR-4 traces for backplane. This work is implemented in $0.18-{\mu}m$ 1-poly 4-metal CMOS technology and occupies $200{\times}300{\mu}m^2$. Measurement results show only 6mW small power consumption and 2Gbps operating range with fabricated chip. The equalizer is expected to satisfy up to 5Gbps operating range if stable varactor(RF) is supported by foundry process.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.1 no.1
• /
• pp.9-25
• /
• 1991

This tutorial paper has been written for biologists, physicians or beginners in fuzzy sets theory and applications. This field is introduced in the framework of medical diagnosis problems. The paper describes and illustrates with practical examples, a general methodology of special interest in the processing of borderline cases, that allows a graded assignment of diagnoses to patients. A pattern of medical knowledge consists of a tableau with linguistic entries or of fuzzy propositions. Relationships between symptoms and diagnoses are interpreted as labels of fuzzy sets. It is shown how possibility measures (soft matching) can be used and combined to derive diagnoses after measurements on collected data. The concepts and methods are illustrated in a biomedical application on inflammatory protein variations. In the case of poor diagnostic classifications, it is introduced appropriate ponderations, acting on the characterizations of proteins, in order to decrease their relative influence. As a consequence, when pattern matching is achieved, the final ranking of inflammatory syndromes assigned to a given patient might change to better fit the actual classification. Defuzzification of results (i.e. diagnostic groups assigned to patients) is performed as a non fuzzy sets partition issued from a "separating power", and not as the center of gravity method commonly employed in fuzzy control. It is then introduced a model of fuzzy connectionist expert system, in which an artificial neural network is designed to build the knowledge base of an expert system, from training examples (this model can also be used for specifications of rules in fuzzy logic control). Two types of weights are associated with the connections: primary linguistic weights, interpreted as labels of fuzzy sets, and secondary numerical weights. Cell activation is computed through MIN-MAX fuzzy equations of the weights. Learning consists in finding the (numerical) weights and the network topology. This feed forward network is described and illustrated in the same biomedical domain as in the first part.

• Journal of Korea Society of Industrial Information Systems
• /
• v.23 no.4
• /
• pp.81-92
• /
• 2018

Bitcoin is a cryptographic digital currency and has been given a significant amount of attention in literature since it was first introduced by Satoshi Nakamoto in 2009. It has become an outstanding digital currency with a current market capitalization of approximately $60 billion. By 2019, it is expected to have over 5 million users. Nowadays, investing in Bitcoin is popular, and along with the advantages and disadvantages of Bitcoin, learning how to forecast is important for investors in their decision-making so that they are able to anticipate problems and earn a profit. However, most investors are reluctant to invest in bitcoin because it often fluctuates and is unpredictable, which may cost a lot of money. In this paper, we focus on solving the Bitcoin forecasting prediction problem based on deep learning structures and neural decomposition. First, we propose a deep learning-based framework for the bitcoin forecasting problem with deep feed forward neural network. Forecasting is a time-dependent data type; thus, to extract the information from the data requires decomposition as the feature extraction technique. Based on the results of the experiment, the use of neural decomposition and deep neural networks allows for accurate predictions of around 89%.