• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.21 no.5
• /
• pp.797-804
• /
• 2012

Magnesium has a long tradition of use as a lightweight material in the field of automotive industry. This paper presents the design optimization process of Mg armrest frame to minimize its weight by replacing the steel frame. formerly, the analysis of steel armrest frame was peformed to determine the design specifications for Mg armrest frame. The initial design of Mg armrest frame was carried out by topological optimization technique. After six types of design variables and four types of response variables were defined, DOE(Design of Experiment) and RSM (Response Surface Method) were applied in order to measure sensitivity of design variables and realize optimization through regression model. After design optimization, the weight of the optimized Mg armrest frame was reduced by about 3% compared to the initial design of the Mg frame and was decreased by 41.7% in comparison with that of the steel frame. Some prototypical armrest frames were also made by die casting process and tested. The results were satisfying for its design specifications.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2006.04a
• /
• pp.928-931
• /
• 2006

Due to their light weight, high stiffness-to-weight and strength-to-weight ratios, and potentially high resistance to environmental degradation, resulting in lower life-cycle costs, polymer composites, are increasingly being considered for use in civil infrastructure applications. Recently, an FRP deck has been installed on a state highway, located in New York State. In this study, a thermal stress analysis was conducted using finite element method to study failure mechanisms of the superstructure. This analysis evaluated small and large temperature gradient effects on the FRP deck considering lightweight of FRP deck and ply orientations at the interface between steel girders and FRP deck Finite element model was verified using the load tests of the bridge deck. Finally, the analytical results shows the possible failure mechanism of FRP deck under various temperature changes and its corresponding index is suddenly varied depending on the rapid change of temperature on the deck plate.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.978-1002
• /
• 2019

With the rapid development of the Internet of Things, the problem of privacy protection has been paid great attention. Recently, Nikooghadam et al. pointed out that Kumari et al.'s protocol can neither resist off-line guessing attack nor preserve user anonymity. Moreover, the authors also proposed an authentication supportive session initial protocol, claiming to resist various vulnerability attacks. Unfortunately, this paper proves that the authentication protocols of Kumari et al. and Nikooghadam et al. have neither the ability to preserve perfect forward secrecy nor the ability to resist key-compromise impersonation attack. In order to remedy such flaws in their protocols, we design a lightweight authentication protocol using elliptic curve cryptography. By way of informal security analysis, it is shown that the proposed protocol can both resist a variety of attacks and provide more security. Afterward, it is also proved that the protocol is resistant against active and passive attacks under Dolev-Yao model by means of Burrows-Abadi-Needham logic (BAN-Logic), and fulfills mutual authentication using Automated Validation of Internet Security Protocols and Applications (AVISPA) software. Subsequently, we compare the protocol with the related scheme in terms of computational complexity and security. The comparative analytics witness that the proposed protocol is more suitable for practical application scenarios.

• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.614-630
• /
• 2023

Dead trees significantly impact forest production and the ecological environment and pose constraints to the sustainable development of forests. A lightweight YOLOv4 dead tree detection algorithm based on unmanned aerial vehicle images is proposed to address current limitations in dead tree detection that rely mainly on inefficient, unsafe and easy-to-miss manual inspections. An improved logarithmic transformation method was developed in data pre-processing to display tree features in the shadows. For the model structure, the original CSPDarkNet-53 backbone feature extraction network was replaced by MobileNetV3. Some of the standard convolutional blocks in the original extraction network were replaced by depthwise separable convolution blocks. The new ReLU6 activation function replaced the original LeakyReLU activation function to make the network more robust for low-precision computations. The K-means++ clustering method was also integrated to generate anchor boxes that are more suitable for the dataset. The experimental results show that the improved algorithm achieved an accuracy of 97.33%, higher than other methods. The detection speed of the proposed approach is higher than that of YOLOv4, improving the efficiency and accuracy of the detection process.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.6
• /
• pp.319-325
• /
• 2023

In this study, we proposed a method for hyperparameter optimization in the building and training of a deep learning model designed to process point cloud data collected by a millimeter-wave radar system. The primary aim of this study is to facilitate the deployment of a baseline model in resource-constrained IoT devices. We evaluated a RadHAR baseline deep learning model trained on a public dataset composed of point clouds representing five distinct human activities. Additionally, we introduced a coarse-to-fine hyperparameter optimization procedure, showing substantial potential to enhance model efficiency without compromising predictive performance. Experimental results show the feasibility of significantly reducing model size without adversely impacting performance. Specifically, the optimized model demonstrated a 3.3% improvement in classification accuracy despite a 16.8% reduction in number of parameters compared th the baseline model. In conclusion, this research offers valuable insights for the development of deep learning models for resource-constrained IoT devices, underscoring the potential of hyperparameter optimization and model size reduction strategies. This work contributes to enhancing the practicality and usability of deep learning models in real-world environments, where high levels of accuracy and efficiency in data processing and classification tasks are required.

• Journal of Communications and Networks
• /
• v.18 no.5
• /
• pp.857-869
• /
• 2016

This paper proposes an X-MAC/BEB protocol that runs a binary exponential backoff (BEB) algorithm on top of an X-MAC protocol to save more energy by reducing collision, especially in densely populated wireless sensor networks (WSNs). X-MAC, a lightweight asynchronous duty cycle medium access control (MAC) protocol, was introduced for spending less energy than its predecessor, B-MAC. One of X-MAC 's conspicuous technique is a mechanism to allow senders to promptly send their data when their receivers wake up. X-MAC, however, has no mechanism to deal with sudden traffic fluctuations that often occur whenever closely located nodes simultaneously diffuse their sense data. To precisely evaluate the impact of the BEB algorithm on X-MAC, this paper builds an analytical model of X-MAC/BEB that integrates the BEB model with the X-MAC model. The analytical and simulation results confirmed that X-MAC/BEB outperformed X-MAC in terms of throughput, delay, and energy consumption, especially in congested WSNs.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.3
• /
• pp.1188-1203
• /
• 2020

The human expression contains a lot of information that can be used to detect complex conditions such as pain and fatigue. After deep learning became the mainstream method, the traditional feature extraction method no longer has advantages. However, in order to achieve higher accuracy, researchers continue to stack the number of layers of the neural network, which makes the real-time performance of the model weak. Therefore, this paper proposed an expression recognition framework based on densely concatenated convolutional neural networks to balance accuracy and latency and apply it to humanoid robots. The techniques of feature reuse and parameter compression in the framework improved the learning ability of the model and greatly reduced the parameters. Experiments showed that the proposed model can reduce tens of times the parameters at the expense of little accuracy.

• The Journal of the Korean dental association
• /
• v.56 no.6
• /
• pp.339-347
• /
• 2018

Conventional mouthguard fabrication process which consists of elastomeric impression taking and followed gypsum model making is changing into intraoral scanning and dental model printing with 3D printer. In addition, new 3D printing materials for mouthgurad, 3D Computer-Aided Design(CAD) software for dental appliance, evaluation of a virtual dentoalveolar model for testing virtually 3D designed mouthguard, and lightweight sensor technology will lead dental professionals to the new era of Sports Dentistry, including information technology integrated custom mouthguard fabrication and creating value with analytic data acquired from sensors in mouthguard.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.9 no.6
• /
• pp.133-139
• /
• 2000

Vibrational characteristics of the vehicle structure are mainly influenced by the shape of the pillar cross section. In this paper a vehicle structural optimization technique has been developed to investigate a lightweight vehicle structure subject to constraints on natural frequencies in a simple beam-and-shell model. In this technique, the optimization procedures involve two stages. In the first stage, the section procedures involve tow stages. In the first stage, the section properties of beam elements of the vehicle structure has been optimized to have minimum weight while satisfying the constraints of natural frequencies. And, in the second stage, the shape of the cross section of the elements of the structure has been determined.

• Transactions on Semiconductor Engineering
• /
• v.1 no.1
• /
• pp.23-30
• /
• 2023

With the recent development of Generative AI technology by IT giants, the size of the transformer model is increasing exponentially over trillion won. In order to continuously enable these AI services, it is essential to reduce the weight of the model. In this paper, we find a hardware-friendly structured pruning pattern and propose a lightweight method of the transformer model. Since compression proceeds by utilizing the characteristics of the model algorithm, the size of the model can be reduced and performance can be maintained as much as possible. Experiments show that the structured pruning proposed when pruning GPT-2 and BERT language models shows almost similar performance to fine-grained pruning even in highly sparse regions. This approach reduces model parameters by 80% and allows hardware acceleration in structured form with 0.003% accuracy loss compared to fine-tuned pruning.