Various methods for blind source separation (BSS) are based on independent component analysis (ICA) which can be viewed as a nonlinear extension of principal component analysis (PCA). Most existing ICA methods require certain nonlinear functions (which leads to higher-order statistics) depending on the probability distributions of sources, whereas PCA is a linear learning method based on second-order statistics. In this paper we show that the PCA can be applied to the task of BBS, provided that source are spatially uncorrelated but temporally correlated. Since the resulting method is based on only second-order statistics, it avoids the nonlinear function and is able to separate mixtures of several colored Gaussian sources, in contrast to the conventional ICA methods.

In this brief, the design method of robust non-fragile decentralized controllers for uncertain large-scale interconnected systems is proposed. Based on Lyapunov second method, a sufficient condition for asymtotic stability is derived in terms of a linear matrix inequality (LMI), and the measure of non-fragility in controller is presented. The solutions of the LMI can be easily obtained using efficient convex optimization techniques. A numerical example is given to illustrate the proposed method.

To design a controller based on the pole placement method, it is necessary to obtain either a target transfer function or a desired characteristic equation which results in the closed-loop response. Specially, a step response in which no overshoot occurs in highly desirable in many applications. In this paper, we present two new present two new prototypes of Type I target transfer functions whose step responses have an overshoot of less than 0.1%. One prototype is obtained by Taylor's approximation of a Gaussian function. It is, however, observed that the response delays increase with increasing order, while the rise times are nearly constant. The other prototype is a modification of the first prototype, so that their transfer function coefficients have particular values in terms of specific parameters ${\gamma}$i and $\tau$ (see section 2). The second prototype gives very useful properties in which step responses are almost the same shape, irrespective of the order. It, also, has no overshoot. Some other properties of the prototypes and an application example are given.

This paper considers the problem of robust H$\infty$ filter is derived by using the equivalence relationship between the FIR filter and the recursive filter, that would be guarantee a prescribed H$\infty$ performance in the continuous-time context, irrespective of the parameter uncertainty and unknown initial states.

In this study, the luminance, luminous efficiency and address time of an PDP as a parameter of the phosphor thickness and the size of discharge space are investigated. The maximum luminance and luminous efficiency are obtained at the phosphor thickness of 30 ${\mu}{\textrm}{m}$ for the rib height of 110 ${\mu}{\textrm}{m}$. For the rib height of 120~160 ${\mu}{\textrm}{m}$, the maximum values are obtained at the thickness of 50 ${\mu}{\textrm}{m}$ regardless of the kind of R, G and B phosphor. These results do not affected by the variation of discharge space. The average decrease rate of the total charge to the phosphor thickness is about 3.35pc/cell/10 ${\mu}{\textrm}{m}$ and the average increase rate of the addressing time is about 0.027 $mutextrm{s}$/10 ${\mu}{\textrm}{m}$.

This paper presents the detailed statistics on radiation field signatures associated with multiple-intracloud lightning discharges. A transient signal recording system was used to measure the electric and magnetic fields produced by lightning flashes. The measurements were made in th summers of 1995 through 1999, and the location of the observation station was in Inchon on the coast of the Yellow Sea in Korea(37$^{\circ}$25'N, 126$^{\circ}$ 39'E). Most of lightning flashes typically contains between two and five strokes. The individual intracloud stork radiation fields were the bipolar pulse. On the average, the ratio of the peak of the second stroke to the first stroke peak was 75.1$\pm$40.1% for the negative, and a fraction of the subsequent stroke peaks were higher than the first stroke peak. The greater the number of the subsequent stroke order, less time separations between strokes were produced. The mean of the depth of the dip was 81.2$\pm$27.9% for the positive polarity and 75.9$\pm$24.4% for the negative. The depth of the dip increased for the positive bipolar pulses and decreased for the negative as the number of the stroke order increased.

The addressing time should be reduced by modifying cell structure and/or driving method in order to replace the dual scan system by single scan and increase the luminance in large ac plasma display panel(PDP). In this paper, the effects of the dielectric layer thickness and the barrier rib height on the addressing time of ac PDP are investigated. It is found out that the addressing time was decreased with decreasing thickness of dielectric layer on the front glass and thickness of white dielectric layer on the rear glass. The decreasing rate were 160ns/10${\mu}{\textrm}{m}$ and 270nsd/10${\mu}{\textrm}{m}$, respectively. Also in case of decreasing the height of barrier rib, addressing time was decreased at the rate of 550ns/10${\mu}m$.

In order to improve the luminous efficiency in ac PDP(Plasma Display Panel), we have suggested the new structures of sustain electrodes with waffle-type barrier ribs and the discharge current for a given pulse voltage has been decreased by eliminating the electrode area which has nothing to do with visible light emission area compared with conventional structure. As a result, the luminous efficiency was improved about 30% compared with the conventional structure.

In this paper, a MW-Mvar investment technique focused on minimizing the system loss is presented. An optimization technique, in which the system loss is defined as the objective function and the power flow equations as the constraints, is introduced to obtain the Lagrangian multipliers λP and λQ. The Lagrangian multipliers imply the variation of the system loss with respect to incremental bus power and are used as MW-Mvar investment indices for minimizing the system loss. ΔP MW and ΔQ Mvar are invested, step by step, by the priority of λP and λQ index given for each bus. Derivation of the index uses the information from normal power flow calculation.

SIME(Single Machine Equivalent) method has been recognized as a useful tool to determine transient stability of power system. In this paper, SIME method is used to develop the KEPCO transient stability assessment (TSA) tool. A new screening algorithm that can be generators are identified by a new index in the early stage of the time domain simulation. Thus, computational time require to find OMIB(One Machine Infinite Bus) can be reduced significantly. Second, clustering critical machines can be performed even in very stable cases. It enables to be avoid extra calculation of time trajectory that is needed in SIME for classifying the stable cases. This algorithm is applied to the fast TSA of the KEPCO system in the year of 2010.

In this paper it is proposed an advanced modified sliding mode control of a rotor field oriented control of induction motor. The application of this unconventional control has very good results, such as disturbance rejection and nice dynamic properties. Stability can be guaranteed even in the worst situation. A conventional "sliding mode" controller is characterised by fast switching control signal, which causes the chattering of the drive system. To overcome this problem, a modified law is used, by introducing a hysteresis band and a continuous control, which modifies the conventional law. The control is accomplished with dual TMS320C44 floating-point digital signal processor. The validity of the proposed method was verified by experiment on the propulsion system simulator, used for the development of Korean High-Speed Railway Train(KHSRT).in(KHSRT).

This paper describes an image processing algorithm capable of recognizing road lanes by using a CDF(cumulative distribution function). The CDF is designed for the model function of road lanes. Based on the assumptions that there are no abrupt changes in the direction and location of road lanes and that the intensity of lane boundaries differs from that of the background, we formulated the CDF, which accumulates the edge magnitude for edge directions. The CDF has distinctive peak points at the vicinity of lane directions due to the directional and the positional continuities of a lane. To obtain lane-related information a scatter diagram was constructed by collecting edge pixels, of which the direction corresponds to the peak point of the CDF, then the principal axis-based line fitting was performed for the scatter diagram. Noises can cause many similar features to appear and to disappear in an image. Therefore, to reduce the noise effect a recursive estimator of the CDF was introduced, and also to prevent false alarms or miss detection a scene understanding index (DUI) was formulated by the statistical parameters of the CDF. The proposed algorithm has been implemented in real time on video data obtained from a test vehicle driven on a typical highway.