• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.22.3-22
• /
• 2001

An attention task, in which context information should be extracted from the first presented pattern, and the recognition answer of the second presented pattern should be generated using the context information, is employed in this paper. An Elman-type recurrent neural network is utilized to extract and keep the context information. A reinforcement signal that indicates whether the answer is correct or not, is only a signal that the system can obtain for the learning. Only by this learning, necessary context information became to be extracted and kept, and the system became to generate the correct answers. Furthermore, the function of an associative memory is observed in the feedback loop in the Elman-type neural network.

• Journal of information and communication convergence engineering
• /
• v.19 no.3
• /
• pp.131-135
• /
• 2021

The interference between closely placed co-coverage satellites was analyzed. In general, a satellite network may use different orthogonal polarizations and frequencies to increase the throughput of a satellite. However, when orthogonal linear polarization (horizontal polarization and vertical polarization) or orthogonal circular polarization (left-handed circular polarization and right-handed circular polarization) is used, the signal from one polarization sense to another may be coupled, resulting in cross-polarization interference. This signal-coupling arises due to the finite value of the cross-polarization discrimination of the earth station. In this study, field equations were used to analyze the interference between adjacent satellites using co-frequency. The level of interference was compared to that when two adjacent satellites used the same polarization. The simulation results show that the interference mainly depends on the off-axis co-polar pattern and the cross-polar pattern of the earth station antenna.

• Current Optics and Photonics
• /
• v.7 no.5
• /
• pp.511-517
• /
• 2023

Pipelines play an important role in urban water supply and drainage, oil and gas transmission, etc. This paper presents a technique for pattern recognition of fiber optic vibration signals collected by a distributed vibration sensing (DVS) system using a deep learning residual network (ResNet). The optical fiber is laid on the pipeline, and the signal is collected by the DVS system and converted into a 64 × 64 single-channel grayscale image. The grayscale image is input into the ResNet to extract features, and finally the K-nearest-neighbors (KNN) algorithm is used to achieve the classification and recognition of pipeline damage.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.20 no.7
• /
• pp.625-630
• /
• 2009

In this paper, we introduce the analysis and counter-measure of signal distortion of search radar equipped on the ship. Using ShipEDF program, the search radar's main antenna and ship structures are modeled in the view of electromagnetism. Ray tracing method is used for analysis of the search radar's radiation patterns in free space and ship condition. From analyzed radiation patterns, we can conclude that the search radar's signal distortion is due to radiation pattern distortion. We also analyze the surface current distribution of the mast to propose the counter-measure to reduce electro-magnetic field reflection of mast.

• Journal of Sensor Science and Technology
• /
• v.26 no.3
• /
• pp.199-203
• /
• 2017

In this paper, we developed an underwater localization system for underwater robot docking using the electromagnetic wave attenuation model. Electromagnetic waves are generally known to be impossible to use in water environment. However, according to the conclusions of the previous studies on the attenuation characteristics in underwater, the attenuation pattern is uniform and its model was accurately proposed and verified in 3-dimensional space via the omnidirectional antenna. In this paper, a docking structure and localization sensor system are developed for a widely used cone type docking mechanism. First, we fabricated electromagnetic wave range sensor transmit modules. And a mobile sensor node is equipped with unmanned underwater vehicle(UUV)s. The mobile node senses the four different signal strength (RSS: Received Signal Strength) from fixed nodes, and the obtained RSS data are transformed to each distance information using the 3-Dimensional EM wave attenuation model. Then, the relative localization between the docking area and underwater robot can be achieved according to optimization algorithm. Finally, experimental results show the feasibility of the proposed localization system for the docking induction by comparing the errors in the actual position of the mobile node and the theoretical position through the model.

• Journal of the Korean Magnetics Society
• /
• v.1 no.2
• /
• pp.74-78
• /
• 1991

The method for analyzing the recording and reading process of magnetic recording system by numerical simulation is introduced. which can simulate the effect of elements of the magnetic recording system. During the whole simulation, hysteresis characteristics must be included. Preisach model and finite element method are used to construct the algorithm. From the wave of recording current, recorded pattern of magnetization in recording media is simulated in recording process, and then waveform of reading signal in head coil is obtained in reading process. To show the validity of this method, this is applied to perpendicular magnetic recording system. Waveform of recording signal and reproduced signal is obtained in several density. Pattern of reprocuced voltage magnitude according to increase of recording density shows roll-off curve.

• International Journal of Fluid Machinery and Systems
• /
• v.2 no.3
• /
• pp.215-222
• /
• 2009

Jets issuing through small holes in a wall into a freestream has proven effective in the control of flow separation. This technique is known as the vortex generator jet (VGJs) method. If a precursor signal of separation is found, the separation control system using VGJs can be operated just before the onset of separation and the flow field with no separation is always attained. In this study, we measured the flow field and the wall static pressure in a two-dimensional diffuser to find a precursor signal of flow separation. The streamwise velocity measurements were carried out in the separated shear layer and spectral analysis was applied to the velocity fluctuations at some angles with respect to the diffuser. The pattern of peaks in the spectral analysis changes as the divergence angle increases over the angle of which the whole separation occurs. This change in the spectral pattern is related to the enhancement of the growth of shear layer vortices and appears just before the onset of separation. Therefore, the growth of shear layer vortices can be regarded as a precursor signal to flow separation.

• Journal of the Ergonomics Society of Korea
• /
• v.29 no.3
• /
• pp.357-364
• /
• 2010

This study proposes a lower-limb exoskeleton system that is controlled by a wearer's muscle activity. This system is designed by following procedure. First, analyze the muscle activation patterns of human leg while walking. Second, select the adequate actuator to support the human walking based on calculation of required force of knee joint for step walking. Third, unit type knee and ankle orthotics are integrated with selected actuator. Finally, using this knee-assistive system (KAS) and developed muscle stiffness sensors (MSS), the muscle activity pattern of the subject is analyzed while he is walking on the stair. This study proposes an operating algorithm of KAS based on command signal of MSS which is generated by motion intent of human. A healthy and normal subject walked while wearing the developed powered-knee exoskeleton on his/her knees, and measured effectively assisted plantar flexor strength of the subject's knees and those neighboring muscles. Finally, capabilities and feasibility of the KAS are evaluated by testing the adapted motor pattern and the EMG signal variance while walking with exoskeleton. These results shows that developed exoskeleton which controlled by muscle activity could help human's walking acceptably.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.23 no.4
• /
• pp.330-336
• /
• 2014

The continuous chip generated in cutting process deteriorates workpiece, tool, and machine tool system. It is necessary to treat this continuous chip in ductile material machining condition for stable cutting. This paper deals with the chip control method using acoustic emission(AE) signal in pure copper turning operation. AE raw signals, root mean square(RMS) signals and wavelet transformed signals measured in turning process are introduced to analysis for chip patterns. With analysis of AE signals, it is obtained that the produced chip patterns are correlated with the specified AE signals which are transformed by fuzzy pattern algorithm. By this experimental investigation, the chip patterns can be classified at significant level in pure copper machining process and controlled from continuous chips to reduced-length stable chips.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.2
• /
• pp.178-183
• /
• 2001

Welding is essential for the manufacture of a range of engineering components which may vary from very large structures such as ships and bridges to very complex structures such as aircraft engines, or miniature components for microelectronic applications. Especially, a domestic situation of the welding automation is still depend on the arc sensing system in comparison to the vision sensing system. Specially, the gap-detecting of workpiece using conventional arc sensor is proposed in this study. As a same principle, a welding current varies with the size of a welding gap. This study introduce to the fuzzy membership filter to cancel a high frequency noise of welding current, and ART2 which has the competitive learning network classifies the signal patterns the filtered welding signal. A welding current possesses a specific pattern according to the existence or the size of a welding gap. These specific patterns result in different classification in comparison with an occasion for no welding gap. The patterns in each case of 1mm, 2mm, 3mm and no welding gap are identified by the artificial neural network.