• Journal of Communications and Networks
• /
• 제8권2호
• /
• pp.194-204
• /
• 2006

This contribution focuses on the maximum likelihood (ML) noncoherent synchronization of multi-antenna transceivers working in faded environments and employing ultra-wideband impulse radio (UWB-IR) transmit technology. In particular, the Cramer-Rao bound (CRB) is derived for the general case of multiple input multiple output (MIMO) UWB-IR systems and used to compare the ultimate performance of three basic transmit schemes, thereinafter referred to as single input multiple output (SIMO), MIMO equal signaling (MIMO-ES), and MIMO orthogonal signaling (MIMO-OS) ones. Thus, the noncoherent ML synchronizer is developed for the better performing transmit scheme (i.e., the SIMO one) and its performance is evaluated under both signal acquisition and tracking operating conditions. The performance gain in the synchronization of UWB- IR signals arising by the utilization of the multi-antenna technology is also evaluated.

• Journal of Communications and Networks
• /
• 제10권2호
• /
• pp.194-203
• /
• 2008

This paper considers the multi-input multi-output (MIMO) relay channel where multiple antennas are employed by each terminal. Compared to single-input single-output (SISO) relay channels, MIMO relay channels introduce additional degrees of freedom, making the design and analysis of optimal cooperative strategies more complex. In this paper, a partial cooperation strategy that combines transmit-side message splitting and block-Markov encoding is presented. Lower bounds on capacity that improve on a previously proposed non-cooperative lower bound are derived for Gaussian MIMO relay channels.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 하계종합학술대회 논문집
• /
• pp.1017-1020
• /
• 1999

Recently, a new learning algorithm for multilayer neural networks has been proposed 〔1〕. In the new learning algorithm, each output neuron is considered as a function of weights and the weights are adjusted so that the output neurons produce desired outputs. And the adjustment is accomplished by taking gradients. However, the gradient computation was performed numerically, resulting in a long computation time. In this paper, we derive the all necessary equations so that the gradient computation is performed analytically, resulting in a much faster learning time comparable to the backpropagation. Since the weight adjustments are accomplished by summing the gradients of the output neurons, we will call the new learning algorithm “multi-gradient.” Experiments show that the multi-gradient consistently outperforms the backpropagation.

• 한국통신학회논문지
• /
• 제16권2호
• /
• pp.128-136
• /
• 1991

A neural network of multi layer perception type, learned by error back propagation learning rule, is generally used for the verification or clustering of similar type of patterns. When learning is completed, the network has a constant value of output depending on a pattern. This paper shows that the intensity of neuron's out put can be controlled by a function which intensifies the excitatory interconnection coefficients or the inhibitory one between neurons in output layer and those in hidden layer. In this paper the value of factor in the function to control the output is derived from the know values of the neural network after learning is completed And also this paper show that the amount of an increased neuron's output in output layer by arbitary value of the factor is derived. For the applications increased recognition performance of a pattern than has distortion is introduced and the output of partially damaged neurons are first managed and this paper shows that the reduced recognition performance can be recovered.

• Journal of Power Electronics
• /
• 제9권3호
• /
• pp.396-402
• /
• 2009

A DC/DC converter generally needs to work under high switching frequency when used as an adjustable power supply to reduce the size of magnetic elements such as inductors, transformers and capacitors, but with the rising of the switch frequency, the switch losses will increase and the efficiency will reduce. Recently, to solve these problems, research is actively being done on a soft switching method that can be applied under high frequency and on a PWM converter that can be applied under low frequency such as a multi-level topology. In this paper a novel DC-DC conversion method for reducing the ripple of output voltage is proposed. In the proposed converter, buck converters are connected in series to generate the output voltage. By using this method, the ripple of output voltage can be reduced compared to a conventional buck converter. Particularly when output voltage is low, the number of acting switching elements is less and the result of ripple reduction is more obvious. It is expected that the converter proposed in this paper could be very useful in the case of wide-range output voltage.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1993년도 한국자동제어학술회의논문집(국내학술편); Seoul National University, Seoul; 20-22 Oct. 1993
• /
• pp.1213-1218
• /
• 1993

Variable structure control is applied to the robust output tracking control problem of general nonlinear multi-input multi-output (MIMO) systems. Using the concept of relative degree and minimum phase, input/output(I/O) linearization is undertaken. For I/O the linearized system, a new sliding hyperplanes design method is proposed. In this procedure, we can construct very robust and efficient sliding mode controller for general nonlinear systems of relative degree higher than two. The control results are illustrated by adopting a numerical example.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1986년도 한국자동제어학술회의논문집; 한국과학기술대학, 충남; 17-18 Oct. 1986
• /
• pp.571-575
• /
• 1986

The squirrel-cage induction motor can be characterized as a nonlinear and multi-input, multi-output system, and has the unmeasurable states which are the rotor currents. In this paper, the time-delayed output feedback method is applied to the speed control of induction motor driven by the current-controlled PWM inverter and its performance is investigated by using the computer simulation.

• 대한전기학회논문지
• /
• 제40권7호
• /
• pp.664-669
• /
• 1991

In order to improve the reliability and the stability of static power conversion system, further to optimize the operational characteristics of the system, it is very important to investigate the effects (output and switching component characteristics) of each parameters in the system. Therefore this paper particualrly deals with the characteristics of a novel ZVS multi-resonant conversion system to duty cycle, resonant capacitance, resonant inductance, and output capacitance. The results show that resonant inductance and output capacitance are dominat factors in the system.

• International Journal of Control, Automation, and Systems
• /
• 제4권2호
• /
• pp.255-270
• /
• 2006

In this paper, a design method of nonlinear autopilot for ship-to-ship missiles is proposed. Ship-to-ship missiles have strongly coupled dynamics through roll, yaw, and pitch channel in comparison with general STT type missiles. Thus it becomes difficult to employ previous control design method directly since we should find three different solutions for each control fin deflection and should verify the stability for more complicated dynamics. In this study, we first propose a control loop structure for roll, yaw, and pitch autopilot which can determine the required angles of all three control fins. For yaw and pitch autopilot design, missile model is reduced to a minimum phase model by applying a singular perturbation like technique to the yaw and pitch dynamics. Based on this model, a multi-input multi-output (MIMO) nonlinear autopilot is designed. And the stability is analyzed considering roll influences on dynamic couplings of yaw and pitch channel as well as the aerodynamic couplings. Some additional issues on the autopilot implementation for these coupled missile dynamics are discussed. Lastly, 6-DOF (degree of freedom) numerical simulation results are presented to verify the proposed method.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2003년도 추계학술대회
• /
• pp.53-58
• /
• 2003

In recent years, the need for multi-stage gear drives, which highly reduce output speed, has been increased. However, the design of multi-stage gear drives have been carried out by a limited number of experts. The consideration for the direction of input and output axes also makes their design very difficult. The purpose of this study is to develop an algorithm for automatically generating complex multi-stage gear drives and to implement a design supporting system for multi-stage gear drives. There are 4 stages in the proposed algorithm, and major design parameters,.such as the direction of input and output axes, reduction ratio, etc. are set up in the first stage. In the second stage, all mechanisms are generated, and various rules are applied to select feasible mechanisms. In the third stage, the gear ratio of each stage is divided from total gear ratio. Next, the specifications of gears for feasible mechanisms are calculated and their bending strength and surface durability are estimated. In the forth stage, appraised indexes are calculated and provided to support the estimation of the generated gear drives.