• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2011.04a
• /
• pp.816-820
• /
• 2011

A method for increasing the difference of side-lobe level in spherical microphone array is presented. In array signal processing, it is known that narrow interval between sensors can increase the difference between main lobe and side-lobe of array response which eventually increase the source recognition capability. Recent commercial array being used, however, have shown certain limitation in using the number of sensors due to its costs and geometrical size of array. To overcome this problem, we have adapted MEMS sensors into spherical microphone array. To check out the improvement, two different types of spherical microphone array were designed. One array is composed with 32 regular instrument microphones and the other one is 85 MEMS sensors. Simulation and experiments were conducted on a sinusoidal noise source with two arrays. The time history data were analyzed with spherical harmonic decomposition and beamforming technique. 85 MEMS sensors array showed the improved side-lobe level suppression by more than 4 dB above the frequency content of 2 kHz compared to 32-sensor array.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.15 no.1
• /
• pp.120-128
• /
• 2011

Recently, safety diagnosis of the existing structures has been emerged as important issue. In particular, systematical and precise safety diagnostics for steel structures for power substation, have been required. Steel structures for power substation are under the periodical impact loads from operations of gas insulated switchgear. These loading condition accelerates damage and aging of structure. The objective of this research is to evaluate damage of structure under periodical impact loads. To evaluate the integrity of structures as organizing mathematical models including the dynamic characteristics of structures, Frequency Domain Decomposition method was choiced and an algorism was proposed. For verifying this methods and algorism, a mathematical model is composed of the development of a variety of reverse analysis and a signal processing technology reflecting physical damage of structures. A series of analysis and test results indicatge that proposed method has a confidence for applying a filed test. Therefore, it is expected to be able to take advantage of system identification to detect damage for the maintenance and management of steel structures under periodical impact loads such as power substation.

• Journal of the Korean Magnetics Society
• /
• v.24 no.1
• /
• pp.18-21
• /
• 2014

The microwave absorption ($P_{tot}$), which is the double integration value of ferromagnetic resonance signal, propositional to the saturation magnetization, and the increase of the $P_{tot}$ measured during the thermal reaction time expect the growth process of the nanoparticles. Therefore, in this work, we measured the $P_{tot}$ in order to obtain the growth time of iron oxide nanoparticles after thermal decomposition of $Fe(acac)_3$ precursor at aging temperature $T_a=273$, 300 and $324^{\circ}C$, respectively. The best condition for monodisperse nanoparticles was obtained at $T_a=300^{\circ}C$, which condition showed the most rapid increase of $P_{tot}$ with thermal reaction time. Finally, the rapid growth rate was necessary condition for the synthesis of iron-oxide monodisperse nanoparticles.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.4A
• /
• pp.403-410
• /
• 2008

In this paper, we propose a variant of the QRM-MLD signal detection method that is used for spatially multiplexed multiple antenna system. The original QRM-MLD signal detection method combines the QR decomposition with the M-algorithm, thereby significantly reduces the prohibitive hardware complexity of the ML signal detection method, still achieving a near ML performance. When the number of transmitter antennas and/or constellation size are increased to achieve higher bit rate, however, its increased complexity makes the hardware implementation challenging. In an effort to overcome this drawback of the original QRM-MLD, a number of variants were proposed. A most strong variant among them, in our opinion, is the ranking method, in which the constellation points are ranked and computation is performed for only highly ranked constellation points, thereby reducing the required complexity. However, the variant using the ranking method experiences a significant performance degradation, when compared with the original QRM-MLD. In this paper, we point out the reasons of the performance degradation, and we propose a novel variant that overcomes the drawbacks. We perform a set of computer simulations to show that the proposed method achieves a near performance of the original QRM-MLD, while its computational complexity is near to that of the QRM-MLD with ranking method.

• Journal of KIISE:Software and Applications
• /
• v.36 no.3
• /
• pp.179-186
• /
• 2009

This paper proposes a target speech extraction which restores speech signal of a target speaker form noisy convolutive mixture of speech and an interference source. We assume that the target speaker is known and his/her utterances are available in the training time. Incorporating the additional information extracted from the training utterances into the separation, we combine convolutive blind source separation(CBSS) and non-negative decomposition techniques, e.g., probabilistic latent variable model. The nonnegative decomposition is used to learn a set of bases from the spectrogram of the training utterances, where the bases represent the spectral information corresponding to the target speaker. Based on the learned spectral bases, our method provides two postprocessing steps for CBSS. Channel selection step finds a desirable output channel from CBSS, which dominantly contains the target speech. Reconstruct step recovers the original spectrogram of the target speech from the selected output channel so that the remained interference source and background noise are suppressed. Experimental results show that our method substantially improves the separation results of CBSS and, as a result, successfully recovers the target speech.

• The Journal of the Acoustical Society of Korea
• /
• v.32 no.6
• /
• pp.548-555
• /
• 2013

The localization of sources has a numerous number of applications. To estimate the position of sources, the relative delay between two or more received signals for the direct signal must be determined. Although the generalized cross-correlation method is the most popular technique, an approach based on eigenvalue decomposition (EVD) is also popular one, which utilizes an eigenvector of the minimum eigenvalue. The performance of the eigenvalue decomposition (EVD) based method degrades in the low SNR and the correlated environments, because it is difficult to select a single eigenvector for the minimum eigenvalue. In this paper, we propose a new adaptive algorithm based on Canonical Correlation Analysis (CCA) in order to extend the operation range to the lower SNR and the correlation environments. The proposed algorithm uses the eigenvector corresponding to the maximum eigenvalue in the generalized eigenvalue decomposition (GEVD). The estimated eigenvector contains all the information that we need for time delay estimation. We have performed simulations with uncorrelated and correlated noise for several SNRs, showing that the CCA based algorithm can estimate the time delays more accurately than the adaptive EVD algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.7A
• /
• pp.683-694
• /
• 2002

In this paper, a digital image compression codec using 2D DWT(Discrete Wavelet Transform) is designed using the FPGA technology for real time operation The implemented image compression codec using wavelet decomposition consists of a wavelet kernel part for wavelet filtering process, a quantizer/huffman coder for quantization and huffman encoding of wavelet coefficients, a memory controller for interface with external memories, a input interface to process image pixels from A/D converter, a output interface for reconstructing huffman codes, which has irregular bit size, into 32-bit data having regular size data, a memory-kernel buffer to arrage data for real time process, a PCI interface part, and some modules for setting timing between each modules. Since the memory mapping method which converts read process of column-direction into read process of the row-direction is used, the read process in the vertical-direction wavelet decomposition is very efficiently processed. Global operation of wavelet codec is synchronized with the field signal of A/D converter. The global hardware process pipeline operation as the unit of field and each field and each field operation is classified as decomposition levels of wavelet transform. The implemented hardware used FPGA hardware resource of 11119(45%) LAB and 28352(9％) ESB in FPGA device of APEX20KC EP20k600CB652-7 and mapped into one FPGA without additional external logic. Also it can process 33 frames(66 fields) per second, so real-time image compression is possible.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.539-542
• /
• 2003

A motor is the workhorse of our industry. The issues of preventive and condition-based maintenance, online monitoring, system fault detection, diagnosis, and prognosis are of increasing importance. Different internal motor faults (e.g., inter-turn short circuits, broken bearings, broken rotor bars) along with external motor faults (e.g., phase failure, mechanical overload, blocked rotor) are expected to happen sooner or later. This paper introduces the fault detection technique of induction motors based upon the stator current. The fault motors have rotor bar broken or rotor unbalance defect, respectively. The stator currents are measured by the current meters and stored by the time domain. The time domain is not suitable to represent the current signals, so the frequency domain is applied to display the signals. The Fourier Transformer is used for the conversion of the signal. After the conversion of the signals, the features of the signals have to be extracted by the signal processing methods like a wavelet analysis, a spectrum analysis, etc. The discovered features are entered to the pattern classification model such as a neural network model, a polynomial neural network, a fuzzy inference model, etc. This paper describes the fault detection results that use wavelet decomposition. The wavelet analysis is very useful method for the time and frequency domain each. Also it is powerful method to detect the features in the signals.

• Smart Structures and Systems
• /
• v.11 no.6
• /
• pp.589-604
• /
• 2013

Identification of damage has become an evolving area of research over the last few decades with increasing the need of online health monitoring of the large structures. The visual damage detection can be impractical, expensive and ineffective in case of large structures, e.g., offshore platforms, offshore pipelines, multi-storied buildings and bridges. Damage in a system causes a change in the dynamic properties of the system. The structural damage is typically a local phenomenon, which tends to be captured by higher frequency signals. Most of vibration-based damage detection methods require modal properties that are obtained from measured signals through the system identification techniques. However, the modal properties such as natural frequencies and mode shapes are not such good sensitive indication of structural damage. Identification of damaged jacket type offshore platform members, based on wavelet packet transform is presented in this paper. The jacket platform is excited by simple wave load. Response of actual jacket needs to be measured. Dynamic signals are measured by finite element analysis result. It is assumed that this is actual response of the platform measured in the field. The dynamic signals first decomposed into wavelet packet components. Then eliminating some of the component signals (eliminate approximation component of wavelet packet decomposition), component energies of remained signal (detail components) are calculated and used for damage assessment. This method is called Detail Signal Energy Rate Index (DSERI). The results show that reduced wavelet packet component energies are good candidate indices which are sensitive to structural damage. These component energies can be used for damage assessment including identifying damage occurrence and are applicable for finding damages' location.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.2C
• /
• pp.155-160
• /
• 2010

In this paper, we proposed a forensic mark algorithm which can embed the distributor's information at each distribution step to trace the illegal distribution path. For this purpose, the algorithm has to have the high capacity payload for embedding the copyright and user information at each step, and the embedded information at a step should not interfere with the information at other step. The proposed algorithm can trace the multilevel distribution because the forensic mark is generated by digital hologram and embedded in the DWT-SVD domain. For the high capacity embedding, the off-axis hologram is generated from the forensic mark and the hologram is embedded in the HL, LH, HH bands of the DWT to reduce the signal interference. The SVD which is applied the holographic signal enhanced the detection performance and the safety of the forensic mark algorithm. As the test results, this algorithm was able to embed 128bits information for the copyright and user information at each step. In this paper, we can embed total 384bits information for 3 steps and the algorithm is also robust to the JPEG compression.