• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.1
• /
• pp.57-62
• /
• 2019

This paper evaluates the QE-MMA (Quantized Error-MMA) adaptive equalization algorithm by the number of quantizer in order to compensates the intersymbol interference due to channel in the transmission of high spectral efficient nonconstant modulus signal. In the adaptive equalizer, the error signal is needed for the updating the tap coefficient, the QE-MMA uses the polarity of error signal and correlation multiplier that condered nonlinear finite bit power-of-two quantizing component in order to convinience of H/W implementation. The different adaptive equalization performance were obtained by the number of quantizer, these performance were evaluated by the computer simulation. For this, the equalizer output signal constellation, residual isi, maximum distortion, MSE, SER were applied as a performance index. As a result of computer simulation, it improved equalization performance and reduced equalization noise were obtained in the steady state by using large quantizer bit numbers, but gives slow in convergence speed for reaching steady state.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.1
• /
• pp.55-60
• /
• 2021

This paper relates with the performance evaluation of QE-MMA (Quantized Error-MMA) adaptive equalization algorithm based on the stepsize and quantizer bit number in order to reduce the intersymbol interference due to nonlinear distortion occurred in the time dispersive channel. The QE-MMA was proposed using the power-of-two arithmetic for the H/W implementation easiness substitutes the multiplication and addition into the shift and addition in the tap coefficient updates process that modifies the SE-MMA which use the high-order statistics of transmitted signal and sign of error signal. But it has different adaptive equalization performance by the step size and quantizer bit number for obtain the sign of error in the generation of error signal in QE-MMA, and it was confirmed by computer simulation. As a simulation, it was confirmed that the convergence speed for reaching steady state depend on stepsize and the residual quantities after steady state depend on the quantizer bit number in the QE-MMA adaptive equalization algorithm performance.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.1
• /
• pp.668-673
• /
• 2021

As the fourth industrial revolution begins in earnest, related technologies are becoming a hot topic. Hardware development is accelerating to make the most of technologies such as high-speed wireless communication, and related companies are growing rapidly. Artificial intelligence often uses desktops in general for related research, but it is mainly used for the learning process of deep learning and often transplants the generated models into devices to be used by including them in programs, etc. However, it is difficult to produce results for devices that do not have sufficient power or performance due to excessive learning or lack of power due to the use of models built to the desktop's performance. In this paper, we analyze efficiency using boards with several Neural Process Units on sale before developing the performance of deep learning to match embedded boards, and deep learning accelerators that can increase deep learning performance with USB, and present a simple development direction possible using embedded boards.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.6
• /
• pp.759-767
• /
• 2020

This paper proposes a new Recurrent neural network (RNN) based spectrum sensing technique for cognitive radio communications. The proposed technique determines the existence of primary user's signal without any prior information of the primary users. The method performs high-speed sampling by considering the whole sensing bandwidth and then converts the signal into frequency spectrum via fast Fourier transform (FFT). This spectrum signal is cut in sensing channel bandwidth and entered into the RNN to determine the channel vacancy. The performance of the proposed technique is verified through computer simulations. According to the results, the proposed one is superior to more than 2 [dB] than the existing threshold-based technique and has similar performance to that of the existing Convolutional neural network (CNN) based method. In addition, experiments are carried out in indoor environments and the results show that the proposed technique performs more than 4 [dB] better than both the conventional threshold-based and the CNN based methods.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.24 no.4
• /
• pp.61-75
• /
• 2021

Roadside greenery in the city is not only a means of reducing fine dust, but also an indispensable element of the city in various aspects such as improvement of urban thermal environment, noise reduction, ecosystem connectivity, and aesthetics. However, in studies dealing with the effect of reducing fine dust through trees in existing urban spaces, microscopic aspects such as the adsorption effect of plants were dealt with, structural changes such as the width of urban buildings and streets, and the presence or absence of trees, Impact studies that reflect the actual form of In this study, the effect of greenery composition applicable to urban space on PM2.5 was simulated through the microclimate epidemiologic model ENVI-met, and field measurements were performed in parallel to verify the results. In addition, by analyzing the results of fine dust background concentration, wind speed, and leaf area index, the sensitivity to major influencing variables was tested. As a result of the study, it was confirmed that the fine dust reduction effect was the highest in the case with a high planting amount, and the reduction effect was the greatest at a low background concentration. Based on this, the cost of planting street green areas and the effect of reducing PM2.5 were compared. The results of this study can contribute as a basis for considering the effect of pedestrian space on air quality when planning and designing street green spaces.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.4
• /
• pp.17-22
• /
• 2022

With the spread of web services, data types are becoming more diversified. In addition to the form of storing data such as images, videos, and texts, the number and form of properties and metadata expressing the data are different for each data. In order to efficiently process such unstructured data, a key-value store is widely used for state-of-the-art applications. LSM-Tree (Log Structured Merge Tree) is the core data structure of various commercial key-value stores. LSM-Tree is optimized to provide high performance for small writes by recording all write and delete operations in a log manner. However, there is a problem in that the delay time and processing speed of user requests are lowered as batches of deletion operations for expired data are inserted into the LSM-Tree as special key-value data. This paper presents a Filtered LSM-Tree (FLSM-Tree) that solves the above problem by separating the deleted key from the main tree structure while maintaining all the advantages of the existing LSM-Tree. The proposed method is implemented in LevelDB, a commercial key-value store and it shows that the read performance is improved by up to 47% in performance evaluation.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.21 no.5
• /
• pp.98-104
• /
• 2022

When processing parts, the cutting oil can improve the cooling performance of the workpiece and tool to increase the precision of the workpiece or extend the life of the tool and facilitate chip extraction. Since such cutting oil has a harmful effect on the environment and the human body due to additives such as sulfur, research on a minimum lubrication supply method using an eco-friendly oil is recently underway. The minimum lubrication supply method minimizes the amount of cutting oil used during processing and processes it, which can reduce the amount of cutting oil used, but has a problem in that cooling performance efficiency is poor. Therefore, this study conducted a study on mist cooling of lubricants to reduce the amount of cutting oil used and maximize the cooling effect of processing heat generated during tapping processing. Spray pressure, processing speed, direction, and lubricant spray amount, which are considered to have an effect on cooling performance, were set as process conditions, and the effect on temperature was analyzed by performing an experiment using the box benquin method among experiments were analyzed. Through the experimental analysis results, the optimal conditions for mist and processing that maximize the cooling effect were derived, and the validity of the results derived through additional experiments was verified. In the case of processing by applying the mist lubrication method verified through this study, it is considered that high-precision processing is possible by improving the cooling effect.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.193-198
• /
• 2023

We study a learner-customized lecture recommendation project using deep learning. Recommendation systems can be easily found on the web and apps, and examples using this feature include recommending feature videos by clicking users and advertising items in areas of interest to users on SNS. In this study, the sentence similarity Word2Vec was mainly used to filter twice, and the course was recommended through the Surprise library. With this system, it provides users with the desired classification of course data conveniently and conveniently. Surprise Library is a Python scikit-learn-based library that is conveniently used in recommendation systems. By analyzing the data, the system is implemented at a high speed, and deeper learning is used to implement more precise results through course steps. When a user enters a keyword of interest, similarity between the keyword and the course title is executed, and similarity with the extracted video data and voice text is executed, and the highest ranking video data is recommended through the Surprise Library.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.5
• /
• pp.529-534
• /
• 2023

Although line printing technology is capable of high-speed and large area printing, residual stresses generated during the manufacturing process can deform the feedhole, causing nozzle plate crack or ink leaks. Therefore, in this paper, we propose a new thermal inkjet print head that is robust, reliable and more suitable for line-printing. The amount of deformation of the conventional line printing head measured through the experiment was converted into an equivalent load, and the validity of the load estimation method was verified through FEA analysis. In addition, in order to minimize deformation without increasing the head size, the head structure was designed to increase internal rigidity by reinforcing the unit nozzle with a pillar or support wall or by adding a support beam or dry/wet etched bridge. The FEA analysis results show that the feedhole deformation was reduced by up to 90%, and it is confirmed that the suggested print head with dry etched feedhole bridge operates normally without nozzle plate cracks and ink leakage through fabrication.

• Smart Structures and Systems
• /
• v.32 no.3
• /
• pp.135-151
• /
• 2023

Using 3D asphalt pavement surface data, a deep and multiscale network named CrackNet-M is proposed in this paper for pixel-level crack detection for improvements in both accuracy and robustness. The CrackNet-M consists of four function-specific architectural modules: a central branch net (CBN), a crack map enhancement (CME) module, three pooling feature pyramids (PFP), and an output layer. The CBN maintains crack boundaries using no pooling reductions throughout all convolutional layers. The CME applies a pooling layer to enhance potential thin cracks for better continuity, consuming no data loss and attenuation when working jointly with CBN. The PFP modules implement direct down-sampling and pyramidal up-sampling with multiscale contexts specifically for the detection of thick cracks and exclusion of non-crack patterns. Finally, the output layer is optimized with a skip layer supervision technique proposed to further improve the network performance. Compared with traditional supervisions, the skip layer supervision brings about not only significant performance gains with respect to both accuracy and robustness but a faster convergence rate. CrackNet-M was trained on a total of 2,500 pixel-wise annotated 3D pavement images and finely scaled with another 200 images with full considerations on accuracy and efficiency. CrackNet-M can potentially achieve crack detection in real-time with a processing speed of 40 ms/image. The experimental results on 500 testing images demonstrate that CrackNet-M can effectively detect both thick and thin cracks from various pavement surfaces with a high level of Precision (94.28%), Recall (93.89%), and F-measure (94.04%). In addition, the proposed CrackNet-M compares favorably to other well-developed networks with respect to the detection of thin cracks as well as the removal of shoulder drop-offs.