• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.67-70
• /
• 2000

최근 들어, 인터넷의 폭넓은 보급과 급속한 대중화에 따라 네트워크를 통하여 음성을 전송하거나 저장하려는 시도가 많이 이루어지고 있다. 본 논문에서는 네트워크를 통한 멀티미디어 전송에서 음성부호화 표준으로 널리 상용되는 ITU-T G.723.1 dual-rate speech coder의 효율적인 다채널 구현을 위한 고속 알고리듬을 제안한다. 고속 알고리듬은 부호화 과정에서 많은 계산량을 차지하는 적응 코드북 검색과 고정 코드북 검색 과정에 적용된다. 적응 코드북 검색 과정에서는 지연과 이득을 동시에 찾는 기존의 방법 대신, 지연과 이득을 순차적으로 검색함으로써 계산량을 개선하였다. 전송률에 따라 다른 알고리듬을 사용하는 고정 코드북 검색 과정에서는 다음과 같은 고속 알고리듬을 제안한다. MP-MLQ(Multi-Pulse Maximum Likely Quantization) 방법을 사용하는 높은 전송률(6.3 kbit/s)인 경우, 펄스를 등 간격으로 검색함으로써 계산량을 줄였다. ACELP(Algebraic CELP) 방법을 사용하는 낮은 전송률(5.3 kbit/s)인 경우는 기존의 nested-loop 검색방법 대신, 펄스를 쌍으로 나누어 순차적으로 찾는 depth-first tree 검색 방법을 적용하여 계산량을 감소시켰다. 제안된 고속 알고리듬에 대해 주관적 음질 평가 방법을 수행한 결과, 제안된 방법이 기존의 방법에 비해 음질의 저하가 없음을 확인하였다. 고정 소수점 DSP인 TMS320C6201을 사용하여 고속 알고리듬을 구현한 결과, 높은 전송률의 경우에는 10.29 MIPS, 낮은 전송률의 경우에는 8.70 MIPS의 연산량으로 구현 가능함을 확인하였다.

• Advances in aircraft and spacecraft science
• /
• v.5 no.4
• /
• pp.431-444
• /
• 2018

Numerical simulations have been conducted to study the unstart/restart characteristics of an over-under turbine-based combined-cycle propulsion system (TBCC) inlet during the inlet transition phase. A dual-solution area exists according to the Kantrowitz theory, in which the inlet states may be different even with the same input parameters. The entire transition process was divided into five stages and the unstart/restart hysteresis loop for each stage was also obtained. These loops construct a hysteresis surface which separates the operating space of the engine into three parts: in which a) inlet can maintain a started state; b) inlet keeps an unstarted state; c) inlet state depends on its initial state. During the transition, the operation of the engine follows a certain order with different backpressures and splitter angles, namely control route, which may result in disparate inlet states. Nine control routes with different backpressures and transition stages were designed to illuminate the route-dependent behavior of the inlet. The control routes operating towards the unstart boundary can make the inlet transit from a started state into an unstarted one. But operating backward the same route cannot make the inlet restart, additional effort should be made.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.4
• /
• pp.1023-1032
• /
• 1995

The multiple cooperating robot system plays an important role in the research of modern manufacturing system as the emphasis of production automation is more on the side of flexibility than before. While the kinematic and dynamic analysis of a single robot is performed as an open-loop chain, the dynamic formulation of robot in a multiple cooperating robot system differs from that of a single robot when the multiple cooperating robots form a closed kinematic chain holding an object simultaneously. The object may be any type from a rigid body to a multi-joint linkage. The mobility of the system depends on the kinematic configuration of the closed kinematic chain formed by robots and object, which also decides the number of independent input parameters. Since the mobility is not the same as the number of robot joints, proper constraint condition is sought. The constraints may be such that : the number of active robot joints is kept the same as mobility, all robot joints are active and have interrelations between each joint forces/torques, two robots have master-slave relation, or so on. The dynamic formulation of system is obtained. The formulation is based on recursive dual-number screw-calculus Newton-Eulerian approach which has been used for single robot analysis. This new scheme is recursive and compact symbolically and may facilitate the consideration of the object in real time.

• Journal of the Korean Mathematical Society
• /
• v.45 no.4
• /
• pp.1089-1100
• /
• 2008

For a pointed space X, the subgroups of self-homotopy equivalences $Aut_{\sharp}_N(X)$, $Aut_{\Omega}(X)$, $Aut_*(X)$ and $Aut_{\Sigma}(X)$ are considered, where $Aut_{\sharp}_N(X)$ is the group of all self-homotopy classes f of X such that $f_{\sharp}=id\;:\;{\pi_i}(X){\rightarrow}{\pi_i}(X)$ for all $i{\leq}N{\leq}{\infty}$, $Aut_{\Omega}(X)$ is the group of all the above f such that ${\Omega}f=id;\;Aut_*(X)$ is the group of all self-homotopy classes g of X such that $g_*=id\;:\;H_i(X){\rightarrow}H_i(X)$ for all $i{\leq}{\infty}$, $Aut_{\Sigma}(X)$ is the group of all the above g such that ${\Sigma}g=id$. We will prove that $Aut_{\Omega}(X_1{\times}\cdots{\times}X_n)$ has two factorizations similar to those of $Aut_{\sharp}_N(X_1{\times}\cdots{\times}\;X_n)$ in reference [10], and that $Aut_{\Sigma}(X_1{\vee}\cdots{\vee}X_n)$, $Aut_*(X_1{\vee}\cdots{\vee}X_n)$ also have factorizations being dual to the former two cases respectively.

• Journal of Power Electronics
• /
• v.11 no.2
• /
• pp.120-131
• /
• 2011

This paper proposes two control strategies for the bidirectional Z-source inverters (BZSI) supplied by batteries for electric vehicle applications. The first control strategy utilizes the indirect field-oriented control (IFOC) method to control the induction motor speed. The proposed speed control strategy is able to control the motor speed from zero to the rated speed with the rated load torque in both motoring and regenerative braking modes. The IFOC is based on PWM voltage modulation with voltage decoupling compensation to insert the shoot-through state into the switching signals using the simple boost shoot-through control method. The parameters of the four PI controllers in the IFOC technique are designed based on the required dynamic specifications. The second control strategy uses a proportional plus resonance (PR) controller in the synchronous reference frame to control the AC current for connecting the BZSI to the grid during the battery charging/discharging mode. In both control strategies, a dual loop controller is proposed to control the capacitor voltage of the BZSI. This controller is designed based on a small signal model of the BZSI using a bode diagram. MATLAB simulations and experimental results verify the validity of the proposed control strategies during motoring, regenerative braking and grid connection operations.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.18 no.2
• /
• pp.181-188
• /
• 2015

This paper present, the result of the guidance and control law for a course correction munitions(CCM) with 2sets of canards positioned in the rotating nose section. The nonlinear simulation model of the CCM was developed based on 7DOF equation of motion. The ability of correcting position was verified by open-loop control input with nonlinear model. The guidance and control command was constructed by reference trajectory which can be obtained with no control. Finally, the performance of the guidance and control law was evaluated through Monte-carlo simulation. The CEP(Circular Error Probability) was obtained by considering the errors in muzzle velocity, aerodynamic coefficient, wind, elevation and azimuth angle and density.

• Journal of Power Electronics
• /
• v.10 no.5
• /
• pp.451-460
• /
• 2010

This paper presents a new kind of digital control metal halide lamp electronic ballast. A zero-voltage-switch quasi-square-wave (ZVS-QSW) dual Buck converter is adopted here. In this paper, a digital control method is proposed to achieve ZVS for the converter. This ZVS can be realized during the whole working condition. Single-cycle-peak-current control is proposed to solve the problem of excessive inductor current during a low-frequency reversal transient. Power loop control is also realized and its consistency for different lamps is good. An AVR special microcontroller for a HID ballast is used to raise the control performance, and the low-frequency square-wave control method is adopted to avoid acoustic resonance. A 70W prototype was built in the laboratory. Experimental results show that the electronic ballast works reliably. Furthermore, the efficiency of the ballast can be higher than 92%.

• Journal of Power Electronics
• /
• v.12 no.4
• /
• pp.519-527
• /
• 2012

A high step-down multiple-output LED driver is proposed in this paper. Firstly, the derivation of the driver with dual-output is presented and its operation principle and steady state performance are analyzed in detail. Secondly, a high step-down N-channel LED driver is proposed and its current auto-balance characteristic and step-down ratio are analyzed. Finally, an experimental prototype is built and the experimental results are given. The theoretical analysis and experimental results show that the proposed driver has the following virtues: First, if load balancing is achieved, the voltage gain is 1/N that of a Buck driver, where N is the number of channels. Second, each output automatically has an equal output current, without requiring more current close-loop control circuits than a Buck driver. Last, the voltage stresses of the switches and diodes are lower than those of a Buck driver, meaning that lower voltage switches and diodes can be used, and a higher efficiency can be expected.

• International Journal of Control, Automation, and Systems
• /
• v.5 no.1
• /
• pp.104-109
• /
• 2007

This paper is about a redundancy management system design for the Smart UAV(unmanned aerial vehicle) which utilizes the tilt..rotor mechanism. In order to meet the safety requirement on the PLOC(probability of loss of control) of $1.7{\times}10^{-5}$ per flight hour for FCS (flight control system) failures, a digital FCS is mechanized with a dual redundant structure. A fault detection system which is composed of a CCM(cross channel monitor) and analytic redundancy using the Kalman filtering is designed, and its effectiveness is evaluated through experiments. A threshold level and persistence count for managing redundant sensors are designed based on the statistical analysis of the FCS sensors. To increase the survivability of the UAV after the loss of critical sensors in the SAS(stability augmentation system) and to provide reference information for a tie-breaking condition at which an ILM(in-line monitor) cannot distinguish the faulty channel between two operating ones, the Kalman filter approach is investigated.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.10 no.3
• /
• pp.119-125
• /
• 2014

Echo cancellation is required for a dual-duplex high speed communication such as digital subscriber line(DSL), in order to allow each individual loop to operate in a full duplex fashion. Echo cancellation was one of the most difficult aspects of DSL design, requiring high linearity and total echo return loss in excess of 70 dB. For a long and rapidly changing echo response, if the echo is cancelled by an adaptive echo canceller, the echo canceller needs more taps and its performance is decreased. But if the response is divided into several responses, which response is estimated by a adaptive digital filter and combined, the computation complexities are decreased and the performance is increased. Therefore, the adaptive multi-stage echo canceller is proposed to decrease the computation complexity and increase the performance of echo return loss, in which the echo canceller is composed of several stage echo canceller estimating each divided echo response. Through computer simulations, this multi-stage echo canceller is verified to have merits for high speed communications such as DSL application.