• The Transactions of the Korean Institute of Power Electronics
• /
• v.3 no.3
• /
• pp.263-272
• /
• 1998

This paper presents the new control strategy that can minimizes the torque ripple by considering the magnetic nonlinearity and phase torque averlapping intervals, and describes the whole SRM drive system using proposed control method implemented by DSP(Digital Signal Processor). To do this, inductance and torque are, at first, measured according to the variation of rotor position angle while current is kept constant at predetermined several values. From these measured values, the entire inductance and torque for any current and rotor position are inferred by using neural network. And the waveform of the reference phase torque is determined for the torque ripple to be minimized considering the torque overlap between phases. The controller is designed for the actual torque obtained by the inferred torque look-up table using measured current and rotor position angle to track the predetermined reference phase torque by delta modulation technique. To perform a real time processing and ensure the reliability of the controller, DSP is implemented.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.12
• /
• pp.635-640
• /
• 2016

It is important factors for a production system to get a profitable result in quality and reliability process. For this reason, there's are various type of research papers in a certain type of data acquisition and application to reliability and quality of the level of M2M devices. In general, a classification problem of slightly different signal such as sensing data is difficult to do with classical statistical methods. There's required real-time and instantaneous calculation properties in machine process. Especially a type of injection molding machine which has a property to be decided in accordance with short-term cycle process needs a solution that can be done a certain type of decision like as good or bad quality immediately. This paper presents a simple application of Kalman Filtering by single sensing data to injection molding process in order to get a correct answer from the real time sensing data.

• The Journal of the Acoustical Society of Korea
• /
• v.11 no.2
• /
• pp.73-84
• /
• 1992

This paper studies noise subtraction methods and distance measures for speech recognition in a noisy environment, and investigates noise robustness of the distance measures applied to the problem of isolated word recognition in white Gaussian and colored noise (vehicle noise) environments. Noise subtraction methods which can be used as a pre-processor for the speech recognition system, such as the spectral subtraction method, autocorrelation subtraction method, adaptive noise cancellation and acoustic beamforming are studied, and distance measures such and Log Likelihood Ratio ($d_{LLR}$), cepstral distance measure ($d_{CEP}$), weighted cepstral distance measure ($d_{WCEP}$), spectral slope distance measure ($d_{RPS}$) and cepstral projection distance measure ($d_{CP},\;d_{BCP},\;d_{WCP},\;d_{BWCP}$) are also investigated. Testing of the distance measures for speaker-dependent isolated word recognition in a noisy environment indicate that $d_{RPS}\;and\;d_{WCEP}$ which weigh higher order cepstral coefficients more heavily give considerable performance improvement over $d_{CEP}and\;d_{LLR}$. In addition, when no pre-emphasis is performed, the recognizer can maintain higher performance under high noise conditions.

• Journal of IKEEE
• /
• v.11 no.1 s.20
• /
• pp.30-37
• /
• 2007

This paper proposes a detection method of position of OFF-switch. Each switch has the parallel path with a serial combination of passive element, its parallel path has each different frequency characteristics. Frequency characteristic of ON-switch reveals a flat spectrum irrelevant to frequency characteristic of passive element connected in parallel to its each terminal and frequency characteristic of OFF-switch reveals the same characteristic as one of passive element connected in parallel. Detection of position of OFF-switch is done by measuring the similarity of each spectrum corresponding to frequency characteristic of passive element connected in parallel to OFF-switch. The measure of their similarity is to calculate Euclidean distance between their test spectrum and reference spectrum. The spectrum with the smallest distance among reference spectrum is recognized as the spectrum of OFF-switch. The real time digital signal processing system is implemented to detect the position of OFF-switch by using spectrum matching.

• Journal of the Korean earth science society
• /
• v.32 no.6
• /
• pp.586-594
• /
• 2011

Gravity Recovery and Climate Experiment (GRACE), launched in April, 2002, makes it possible to monitor Earth's mass redistribution with its time-varying gravity observation. GRACE provides monthly gravity solutions as coefficients of spherical harmonics, and thus ones need to convert the gravity spectrum to gravity grids (or mass grids) via the spherical harmonics. GRACE gravity solutions, however, include spatial alias error as well as noise, which requires to suppress in order to enhance signal to noise ratio. In this study, we present the GRACE data processing procedures and introduce some applications of time-varying gravity, which are studies of terrestrial water storage changes, Antarctic and Greenland ice melting, and sea level rise. Satellite missions such as GRACE will continue up to early 2020, and they are expected to be an essential resource to understand the global climate changes.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.9 no.4
• /
• pp.71-80
• /
• 2006

Strapdown Inertial navigation systems consist of an inertial sensor assembly(ISA), electronic modules to process sensor data, and a navigation computer to calculate attitude, velocity and position. In the ISA, most gryoscopes such as RLGs and FOGs, have digital output, but typical accelerometers use current as an analog output. For a high precision inertial navigation system, sufficient stability and resolution of the accelerometer board converting the analog accelerometer output into digital data needs to be guaranteed. To achieve this precision, the asymmetric error and A/D reset scale error of the accelerometer board must be properly compensated. If the relation between the acceleration error and the errors of boards are exactly known, the compensation and estimation techniques for the errors may be well developed. However, the A/D Reset scale error consists of a pulse-train type term with a period inversely proportional to an input acceleration additional to a proportional term, which makes it difficult to estimate. In this paper, the effects on the acceleration output for auto-pilot situations and the effects of A/D reset scale errors during horizontal alignment are qualitatively analyzed. The result can be applied to the development of the real-time compensation technique for A/D reset scale error and the derivation of the design parameters for accelerometer board.

• The KIPS Transactions:PartB
• /
• v.8B no.5
• /
• pp.579-586
• /
• 2001

Digital watermarking is to embed imperceptible mark into image, video, audio, and text data to prevent the illegal copy of multimedia data. arbitrary modification, and also illegal sales of the copies without agreement of copyright ownership. In this paper, we study for the embedding and extraction of watermark key using wavelet in the luminance signal in order to implement the system to protect the copyright for image MPEG. First, the original image is analyzed into frequency domain by discrete wavelet transform. The RSA(Rivest, Shamir, Aldeman) public key of the coded target is RUN parameter of VLD(variable length coding). Because the high relationship among the adjacent RUN parameters effect the whole image, it prevents non-authorizer not to possess private key from behaving illegally. The Results show that the proposed method provides better moving picture and the distortion more key of insert than direct coded method on low-frequency domain based DCT.

• Information and Communications Magazine
• /
• v.11 no.2
• /
• pp.77-89
• /
• 1994

High speed optical fiber transmission technology has been remarkably improved during the past 20 years. This paper presents recent research status and future technological issues for the future information society, that is, the Tb/s transmission by frequency division multiplexing and the ultra long-distance by optical soliton transmission. Erbium-doped fiber amplifier and recent optical technology have brought optical transmission system of up to 10 Gb/s to the point of commercialization. Taking into account the future super information highway, that is, B-ISDN network, ultra wide-band picture-based information can be provided for many subscribers via existing optical fiber cables. However, to achieve the high speed transmission, the technologies must be developed not only for transmission lines but also for transmission nodes. Since the conventional signal transmission/processing technique using electronics has the limit in its speed, novel photonic technology is being developed for this purpose. On the other hand, optical solitons propagate stably through optical fibers, without pulse broadening effect of the fiber dispersion. Since the pulse broadening effect becomes serious as the transmission speed increases, optical solitons is the important technologies to realize the high speed, long distance transmission.

• Journal of IKEEE
• /
• v.2 no.2 s.3
• /
• pp.309-315
• /
• 1998

This paper describes an 8b 52 MHz CMOS subranging analog-to-digital converter (ADC) for Integrated Services Digital Network (ISDN) applications. The proposed ADC based on the improved time-interleaved architecture removes the holding time which is typically observed in the conventional double-channel subranging ADCs to increase throughput rate. Moreover, the ADC employs the interpolation technique in the back-end subranging ADCs far residue signal processing to minimize die area and power consumption. The fabricated and measured prototype ADC in a 0.8 um n-well double-poly double-metal CMOS process typically shows a 52 MHz sampling rate at a 5 V supply voltage with 230 mW, and a 40 MHz sampling rate at a 3 V power supply with 60 mW power consumption.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.39 no.4
• /
• pp.25-35
• /
• 2002

This work describes an 8b 200MHz time-interleaved subranging analog-to-digital converter (ADC) based on a single-poly digital CMOS process. Two fine ADCs for lower digital bits of the proposed ADC employ a time-sharing double-channel architecture to increase system speed and a new reference voltage switching scheme to reduce settling time of the reference voltages and chip area. The proposed intermeshed resistor string, which generates reference voltages for fine ADCs, improves linearity and settling time of the reference voltages simultaneously. The proposed sample- and-hold amplifier(SHA) is based on a highly linear common-drain amplifier and passive differential circuits to minimize power consumption and chip area with 8b accuracy and employs input dynamic common mode feedback circuits for high dynamic performance at a 200MHz sampling rate. A new encoding circuit in a coarse ADC simplifies the signal processing between the coarse ADC and two successive fine ADCs.