• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.1B
• /
• pp.8-14
• /
• 2000

This paper analyzes and accesses the gain of optically pumped surface-normal MQW optical amplifiers. The proposed amplifiers have the advantage of polarization independence, high coupling efficiency to and from optical fibers, and flexibility of operating wavelength. We analyzed the gain characteristics of 100 - 200-period MQWs and verified the dependence of a strained lattice and selective doping. Theoretical analysis of such MQWs showsa single-pass gain of 3 dB with broad operation bandwidth. A single-pass gain of 2.6 dB is obtained experimentally in an InGaAs/InGaAlAs MQW amplifier, which is compared with calculations. The use of Fabry-Perot interferometer (FPI) structure in an optical amplifier is a useful way to increase the gain, but causes a problem of narrow operation bandwidth when the single-pass gain is low. Therefore, a single-pass gain above 2to 3 dB is a prerequisite to achieve both a high gain and moderate operation bandwidth in FPI-structured opticalamplifiers. We have designed an FPI-structured surface-normal optical amplifier both with a high gain of broad operation bandwidth of 4.6mm, when a single-pass gain is 3 dB.

• Korean Journal of Optics and Photonics
• /
• v.18 no.2
• /
• pp.93-99
• /
• 2007

We investigated analytically and numerically the occurrence of modulation instability in fibers with periodic changes both in dispersion and gain. Previously, it has been known that the modulation instability is suppressed in dispersion managed solitons where dispersion is managed in such a way that the local dispersion alternates between the normal and the anomalous regimes. In this work, we enhanced the advantage of the dispersion management scheme by additionally introducing proper gain/loss profiles in fibers. The gain/loss profile is given by $\Gamma(z)=0.5/D(z)*(dD/dz)$, where D(z) represents the dispersion profile. The fundamental gain spectra of the modulation instability in the dispersion and gain managed fibers have been derived analytically and confirmed by numerical calculation. Our investigation reveals that in the dispersion and gain fibers the modulation instabilities are always much more suppressed compared to the case with only dispersion managed. In practical dispersion management schemes, dispersion profiles show discontinuity. and thus. the corresponding gain/loss profiles tend to be finite. In these cases, the gain/loss profiles were approximated by lumped gains/losses of finite values. Our numerical calculations confirm that this approximation also works well.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.10
• /
• pp.1888-1892
• /
• 2008

In this paper, a new parallel CMOS self-bias differential amplifier is designed to use in high-speed analog signal processing circuits. The designed parallel CMOS self-bias differential amplifier is developed by using internal biasing circuits and the complement gain stages which are parallel connected. And also, the parallel architecture of the designed parallel CMOS self-bias differential amplifier can improve the gain and gain-bandwidth product of the typical CMOS self-bias differential amplifier. With 1.8V $0.8{\mu}m$ CMOS process parameter, the results of HSPICE show that the designed parallel CMOS self-bias differential amplifier has a dc gain and a gain-bandwidth product of 64 dB and 49 MHz respectively.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.62 no.9
• /
• pp.1234-1239
• /
• 2013

This paper provides a technical review of speed control using variable gain PI algorithm for BLDC(Brushless DC) motor. Usually the PI control is used in many motor applications, but a general PI control has problems of overshooting and disturbance for response. By the change of PI gain in motor control operation, these problems can be solved. To find the optimized PI gains for BLDC motor control, many control methods have been proposed. In this paper, the control algorithm with a variable PI gain is applied to improve overshooting response in transient region and rapid load disturbance rejection. Fixed gain and variable gain PI controls are compared. The validity of the propose method is verified by experiment.

• ETRI Journal
• /
• v.25 no.5
• /
• pp.407-411
• /
• 2003

A dielectric superstrate layer above a microstrip patch antenna has remarkable effects on its gain and resonant characteristics. This paper experimentally investigates the effect of a superstrate layer for high gain on microstrip patch antennas. We measured the gain of antennas with and without a superstrate and found that the gain of a single patch with a superstrate was enhanced by about 4 dBi over the one without a superstrate at 12 GHz. The impedance bandwidths of a single patch with and without a superstrate for VSWR < 2 were above 11%. The designed $2{\times}8$ array antenna using a superstrate had a high gain of over 22.5 dB and a wide impedance bandwidth of over 17%.

• Korean Journal of Optics and Photonics
• /
• v.9 no.3
• /
• pp.181-185
• /
• 1998

Gain characteristic of concatenated erbium-doped fiber amplifiers(EDFA) are studied with a recirculating EDFA loop. For a non-flat gain EDFA, the 3 dB gain bandwidth was reduced to 6 nm after the 20th EDFA. However, for an optimized gain flattened EDFA, in a simple configuration, the 5 dB gain bandwidth was found to be 9nm, even after the 100th EDFA, corresponding to 8000km transmission. This results suggest that the simple optimized flat gain amplifier could be a good candidate for ultra-long distance wavelength division multiplexed transmissions.

• ETRI Journal
• /
• v.20 no.1
• /
• pp.28-36
• /
• 1998

A simple experimental method to design gain-flattened erbium-doped fiber amplifier is proposed and demonstrated based on the two linear relations between the output power and the pump power, and between the gain and the length of the eribium-doped fiber at the gain flattened state. The spectral gain variation of the eribium-doped fiber amplifiber constructed by this method was less than 0.4 dB over 12 nm (1,545~1,557nm) wavelength region. The gain flatness is also controlled within 0.4 dB over the input power range of -30~-15dBm/ch through the feedback control utilizing the amplified spontaneous emission power in the 1,530 nm region.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.3
• /
• pp.724-730
• /
• 1995

Fixed-gain and variable-gain PID control of tension in the winding section of a simple web transport system were evaluated. An open-loop mathematical model for the web transport system was derived and used for the design of the PID controllers. The winding roll radius is a timevarying parameter in the model. The fixed-gain PID controller designed at a particular instant of time could not meet the desired specifications, whereas the variable-gain PID controller could produce accurate tension control in the winding section. An advantage of the variable-gain control is its simplicity. This approach is easy to implement and shows promise for applications where the time-varying parameters are easily measured.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.56 no.12
• /
• pp.2257-2261
• /
• 2007

A new CMOS transresistance amplifier for low-voltage analog integrated circuit design applications is presented. The proposed transresistance amplifier circuit based on common-source and negative feedback topology is compared with other recent reported transresistance amplifier. The proposed transresistance amplifier achieves high transresistance gain, gain-bandwidth with the same input/output impedance and the minimum supply voltage $2V_{DSAT}+V_T$. Hspice simulation using 1.8V TSMC $0.18{\mu}m$ CMOS technology was performed and achieved $59dB{\Omega}$ transresistance gain which is above the maximum about $18dB{\Omega}$ compared to transresistance gain of the reported circuit.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.52 no.10
• /
• pp.583-590
• /
• 2003

This paper deals with speed control of DC servo motor using a high gain obserber. It was designed to estimate rotor speed of DC servo motor and it carries out speed control from the feedback of the estimated speed signal. Also, PI controller was used in speed controller. In order to verify the performance of the high gain observer which is proposed in this paper, it is compared estimate performance of Luenberger Observer and High Gain Observer with the computer simulation. Effectiveness of the proposed high gain observer is proved from the experiment to compare the case with a speed sensor to the case with high gain observer in the speed control of DC servo motor.