• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.40 no.11
• /
• pp.37-44
• /
• 2003

A distortion suppression technology for employing multiple inputs in 3n＋1 type current mode Max circuit is proposed using the adjustment of transconductance. If the number of input blocks of the current mode Max circuit increases, the high frequency distortion in the output signal grows. In this paper, it has been disclosed that the distortion in the multiple input Max circuit is proportional to such accumulated parasitic capacitance, to the derivative of the output signal and also to tile inverse of transconductance of the common diode-connected transistor. The proposed idea is by employing as larger transconductance of the common diode-connected transistor as possible. The effectiveness of the proposed idea has been proved through the HSPICE simulation for the current mode Max circuits with various numbers of input signals.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.2
• /
• pp.325-328
• /
• 2022

A convolutional neural network(CNN) is widely used in the computer vision tasks, but its computing power requirement needs a design of a special circuit. Most of the computations in a CNN can be implemented efficiently in a digital circuit, but the SoftMax layer has operations unsuitable for circuit implementation, which are exponential and logarithmic functions. This paper proposes a new method to integrate the exponential and logarithmic tables of the conventional circuits into a single table. The proposed structure accesses a look-up table (LUT) only with a few maximum values, and the LUT has the result value directly. Our proposed method significantly reduces the space complexity of the SoftMax layer circuit implementation. But our resulting circuit is comparable to the original baseline with small degradation in precision.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.513-513
• /
• 2000

An integrable circuit technique for implementing high-speed analog two-input Max/Min circuit is described. The realization method is suitable for fabrication using CMOS technology. The proposed circuit comprises a current mirror and electronic switch connected with a absolute value circuit. The maximum or minimum operation of the proposed circuit can be selected by an external control voltage. The proposed analog Max/Min circuit has a very sharp transfer characteristic and is suitable for real-time systems. Simulation results verified the circuit performances are agreed with the expected values.

• Proceedings of the IEEK Conference
• /
• 2003.07b
• /
• pp.1053-1056
• /
• 2003

A distortion suppression technology for employing multiple inputs in 3n+1 type current mode Max circuit is proposed using the adjustment of transconductance. If the number of inputs in current mode Max circuit increases, the high frequency distortion in the output signal grows. In this paper, it has been disclosed that the distortion in the multiple input Max circuit is proportional to sum of parasitic capacitance in input terminals, to the derivative of the output signal and also to the inverse of transconductance of the common diode-connected transistor. The proposed idea is by employing as larger transconductance of the common diode-connected transistor as possible. The effectiveness of the proposed idea has been proved through the HSPICE simulation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.2
• /
• pp.330-333
• /
• 2021

In a CNN for image classification, a SoftMax layer is usually placed at the end. The exponentinal and logarithmic operations in the SoftMax layer are not adequate to be implemented in an accelerator circuit. The operations are usually implemented with look-up tables, and the exponential operation can be implemented in an iterative method. This paper proposes a successive approximation method to calculate a logarithm to remove a very large look-up table. By substituing the large table with two very small tables, the circuit can be reduced much. The experimental results show that the 85% area reduction can be reached with a small error degradation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.2
• /
• pp.248-255
• /
• 2009

Error is the major problem in communication system. Absolute circuit is one of the most important building blocks to implement for the error measurement in communication system as well as in analog circuit design. The main goal of this paper is to design a circuit with high accuracy and minimum error performance. In this paper, a new current mode absolute circuit is implemented to calculate the absolute value of two signals. This new design shows enhanced performance and low distortion over the previous implementation. The proposed circuit is simulated using Hspice and implemented in analog viterbi decoder. It is very suitable for implementing in error calculation for the large scale integrated circuit. Hspice simulation results of previous and new one circuit are reported.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.48 no.2
• /
• pp.67-72
• /
• 2011

8-layer printed circuit board is designed and implemented for triple band(2.3/2.5/3.5GHz) m-WiMAX system. In order to maintain excellent RF performance, low dielectric constant material is used for implementation of the printed circuit board. Also, embedded printed circuit board which embed passive devices is manufactured to reduce total system area. As a result, total system area is cut off by 9%. Triple band m-WiMAX system is produced using embedded printed circuit board. Furthermore, internet connecting test is performed and proved successful running of the system. The developed embedded printed circuit board will provide a effective solution for system area reduction and low loss signal RF communication system.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1992.04a
• /
• pp.354-358
• /
• 1992

This paper describes a development of high speed fuzzy inference circuit for the industrialprocesses. The hardware fuzzy inference circuit is developed utilizing a hardware fuzzy inference circuit is developed utilizing a DSP and a multiplier and accumulator chip. To enhance the inference speed, the pipeline disign is adopted at the bottleneck and the general Max-Min inference method is slightly modified as Max-max method. As a results, the inference speed is evaluated to be 100 KFLIPS. Owing to this high speed feature, satisfactory application can be attained for complex high speed motion control as well as the control of multi-input multi-output nonlinear system. As an application, the developed fuzzy inference circuit is embedded to a PLC (Porgrammable Logic Controller) for industrial process control. For the fuzzy PLC system, to fascilitate the design of the fuzzy control knowledge such as membership functions, rules, etc., a MS-Windows based GUI (Graphical User Interface) software is developed.

• Proceedings of the IEEK Conference
• /
• 2000.06d
• /
• pp.147-150
• /
• 2000

Hausdorff distance(HD) commonly used measures for object matching, and calculates the distance between two point set of pixels in two-dimentional binary images without establishing correspondence. And it is realized as the image filter applying the fuzzy. In this paper, the fuzzy hardware realizes in order to construct the image filter applying HD, also, propose as the method for the noise removal using it in the image. MIN-MAX circuit designs the circuit using MAX-PLUS, and the fuzzy HD hardware results are obtained to the simulation. And then, the previous computer simulation is confirmed to the result by using MATLAB.

• Journal of Welding and Joining
• /
• v.18 no.6
• /
• pp.42-47
• /
• 2000

The object of this study is to estimate Sn-Ag-Bi-Ga solder alloy as a substitute for Sn-37Pb alloy. For Sn-Ag-Bi-Ga alloys, Ag, Bi and Ga contents are varied. (Ag : 1~5%, Ga : 3%, Bi : 3~6%) Comparing to Sn-37Pb alloy Sn-Ag-Bi-Ga alloys have wider melting temperature range up to max. $18.7^{\circ}C$. With increasing Ag, Bi contents, the wettability of the alloys increased up to max. 6.6 mN. The vickers hardness of the alloys was max. 46.4 Hv. The ultimate tensile stress of the alloys was max. 60.3 MPa and the elongation was max. 1.2%. The joint strength between circuit board and solder was max. 55.5 N and the joint strength between connector and solder was max. 176.1 N. There were no cracks in this alloys after thermal shock test.