• Journal of the Semiconductor & Display Technology
• /
• v.17 no.4
• /
• pp.42-45
• /
• 2018

As The semiconductor decrease from 10 nanometer to 7 nanometer, It is suggested that "More than Moore" is needed to follow Moore's Law, which has been a guide for the semiconductor industry. Fan-Out Wafer Level Package(FOWLP) is considered as the key to "More than Moore" to lead the next generation in semiconductors, and the reasons are as follows. the fan-out WLP does not require a substrate, unlike conventional wire bonding and flip-chip bonding packages. As a result, the thickness of the package reduces, and the interconnection becomes shorter. It is easy to increase the number of I / Os and apply it to the multi-layered 3D package. However, FOWLP has many issues that need to be resolved in order for mass production to become feasible. One of the most critical problem is the warpage problem in a process. Due to the nature of the FOWLP structure, the RDL is wired to multiple layers. The warpage problem arises when a new RDL layer is created. It occurs because the solder ball reflow process is exposed to high temperatures for long periods of time, which may cause cracks inside the package. For this reason, we have studied warpage in the FOWLP structure using commercial simulation software through the implementation of the reflow process. Simulation was performed to reproduce the experiment of products of molding compound company. Young's modulus and poisson's ratio were found to be influenced by the order of influence of the factors affecting the distortion. We confirmed that the lower young's modulus and poisson's ratio, the lower warpage.

• Korean Journal of Materials Research
• /
• v.29 no.6
• /
• pp.392-397
• /
• 2019

A cold roll-bonding process is applied to fabricate an AA6061/AA5052/AA6061 three-layer clad sheet. Two AA6061 and one AA5052 sheets of 2 mm thickness, 40 mm width, and 300 mm length are stacked, with the AA5052 sheet located in the center. After surface treatment such as degreasing and wire brushing, sample is reduced to a thickness of 1.5 mm by multi-pass cold rolling. The rolling is performed at ambient temperature without lubricant using a 2-high mill with a roll diameter of 400 mm at rolling speed of 6.0 m/sec. The roll bonded AA6061/AA5052/AA6061 complex sheet is then hardened by natural aging(T4) and artificial aging(T6) treatments. The microstructures of the as-roll bonded and age-hardened Al complex sheets are revealed by optical microscopy; the mechanical properties are investigated by tensile testing and hardness testing. After rolling, the roll-bonded AA6061/AA5052/AA6061 sheets show a typical deformation structure in which grains are elongated in the rolling direction. However, after T4 and T6 aging treatment, there is a recrystallization structure consisting of coarse equiaxed grains in both AA5052 and AA6061 sheets. The as roll-bonded specimen shows a sandwich structure in which an AA5052 sheet is inserted into two AA6061 sheets with higher hardness. However, after T4 and T6 aging treatment, there is a different sandwich structure in which the hardness of the upper and lower layers of the AA6061 sheets is higher than that of the center of the AA5052 sheet. The strength values of the T4 and T6 age-treated specimens are found to increase by 1.3 and 1.4 times, respectively, compared to that value of the starting material.

• Membrane Journal
• /
• v.29 no.3
• /
• pp.130-139
• /
• 2019

Various two-dimensional nano materials such as graphene, zeolite, and metal-organic framework have been utilized to develop an ultra-thin high-performance membrane for water purification, gas separation, and so on. Particularly, in the case of graphene oxide, synthesis methods and thin film coating techniques have been accumulated and established since early 2000s, therefore graphene oxide has been rapidly applied to membrane field. The multi-layered graphene oxide thin film can filter molecules separately by the molecular sieving of interlayer spacing between adjacent layers, and it is also possible to separate various materials depending on the surface functional groups or the degree of interaction to intercalated materials. This review mainly focuses on the nanofiltration application of graphene oxide. The major factors affecting the separation performance of graphene oxide membrane in solvent are summarized and other technical elements required for the commercialization of graphene oxide membranes will be discussed including stability issue and fabrication method.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.18 no.1
• /
• pp.59-64
• /
• 2019

In recent years, the HP-CRTM method, which has the ability to produce carbon fiber-reinforce plastic composites at high speeds, has come into the spotlight in the automotive parts industry, which demands high productivity. Multi-axial carbon fabric, an intermediate material used in this HP-CRTM molding process, consists of layered fibers without crimp, which makes it better in terms of tensile and shear strength than the original woven fabrics. The NCF (non-crimp fabric) can form the layers of the carbon fiber, which have different longitudinal and lateral directions, and ${\pm}{\theta}$ degrees, depending on the product's properties. In this research, preforms were made with carbon fibers of ${\pm}45^{\circ}$ and $0/90^{\circ}$, which were lamination structures under seven different conditions, in order to create the optimal laminated structure for automobile reinforcement center floor tunnels. Carbon fiber composites were created using each of the seven differently laminated preforms, and polyurethane was used as the base material. The specimens were manufactured in accordance with the ASTM D3039 standards, and the effect of the NCF lamination structure on the mechanical properties was confirmed by a tensile test.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.32 no.2
• /
• pp.157-160
• /
• 2019

P(VDF-TrFE-CFE) (Poly (vinylidene fluoride-trifluoroethylene-chlorofluoroethylene)), which exhibits a high electrostriction of about 7%, can transmit tactile output as vibration or displacement. In this study, we investigated the applicability of P(VDF-TrFE-CFE) to wearable piezoelectric actuators. The P(VDF-TrFE-CFE) layers were deposited through spin-coating, and interspaced with patterned Ag electrodes to fabricate a two-layer $3.5mm{\times}3.5mm$ device. This layered structure was designed and fabricated to increase the output and displacement of the actuator at low driving voltages. In addition, a laser vibrometer and piezoelectric force microscope were used to analyze the device's vibration characteristics over the range of ~200~4,200 Hz. The on-off characteristics were confirmed at a frequency of 40 Hz.

• Journal of Digital Convergence
• /
• v.17 no.5
• /
• pp.217-223
• /
• 2019

A study in predicting defects of spot welding in real time in automotive field is essential for cost reduction and high quality production. Welding quality is determined by shear strength and the size of the nugget, and results depend on different independent variables. In order to develop the real-time prediction system, multiple regression analyses were conducted and the two dependent variables were obtained with sufficient statistical results with three independent variables, however, the quality prediction by the regression formula could not ensure accuracy. In this study, a multi-layer neural network circuit was constructed. The neural network by 10 dynamic resistance variables was constructed with three hidden layers to obtain execution functions and weighting matrix. In this case, the neural network was established with three independent variables based on regression analysis, as there could be difficulties in real-time control due to too many input variables. As a result, all test data were divided into poor, partial, and modalities. Therefore, a real-time welding quality determination system by three independent variables obtained by multiple regression analysis was completed.

• Korean Journal of Materials Research
• /
• v.29 no.11
• /
• pp.697-702
• /
• 2019

In this study, using a wet chemical process, we evaluate the effectiveness of different solution concentrations in removing layers from a solar cell, which is necessary for recovery of high-purity silicon. A 4-step wet etching process is applied to a 6-inch back surface field(BSF) solar cell. The metal electrode is removed in the first and second steps of the process, and the anti-reflection coating(ARC) is removed in the third step. In the fourth step, high purity silicon is recovered by simultaneously removing the emitter and the BSF layer from the solar cell. It is confirmed by inductively coupled plasma mass spectroscopy(ICP-MS) and secondary ion mass spectroscopy(SIMS) analyses that the effectiveness of layer removal increases with increasing chemical concentrations. The purity of silicon recovered through the process, using the optimal concentration for each process, is analyzed using inductively coupled plasma atomic emission spectroscopy(ICP-AES). In addition, the silicon wafer is recovered through optimum etching conditions for silicon recovery, and the solar cell is remanufactured using this recovered silicon wafer. The efficiency of the remanufactured solar cell is very similar to that of a commercial wafer-based solar cell, and sufficient for use in the PV industry.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.5
• /
• pp.13-20
• /
• 2019

Recently, object detection technology based on CNN has been actively studied. Object detection technology is used as an important technology in autonomous vehicles, intelligent image analysis, and so on. In this paper, we propose a rotation change robust object detection system by applying MI-FL (Moment Invariant-Feature Layer) to SSD (Single Shot Multibox Detector) which is one of CNN-based object detectors. First, the features of the input image are extracted based on the VGG network. Then, a total of six feature layers are applied to generate bounding boxes by predicting the location and type of object. We then use the NMS algorithm to get the bounding box that is the most likely object. Once an object bounding box has been determined, the invariant moment feature of the corresponding region is extracted using MI-FL, and stored and learned in advance. In the detection process, it is possible to detect the rotated image more robust than the conventional method by using the previously stored moment invariant feature information. The performance improvement of about 4 ~ 5% was confirmed by comparing SSD with existing SSD and MI-FL.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.4
• /
• pp.1317-1341
• /
• 2021

Nowadays, the Internet of Things (IoT) is adopted to enable effective and smooth communication among different networks. In some specific application, the Wireless Sensor Networks (WSN) are used in IoT to gather peculiar data without the interaction of human. The WSNs are self-organizing in nature, so it mostly prefer multi-hop data forwarding. Thus to achieve better communication, a cross-layer routing strategy is preferred. In the cross-layer routing strategy, the routing processed through three layers such as transport, data link, and physical layer. Even though effective communication achieved via a cross-layer routing strategy, energy is another constraint in WSN assisted IoT. Cluster-based communication is one of the most used strategies for effectively preserving energy in WSN routing. This paper proposes a Bio-inspired cross-layer routing (BiHCLR) protocol to achieve effective and energy preserving routing in WSN assisted IoT. Initially, the deployed sensor nodes are arranged in the form of a grid as per the grid-based routing strategy. Then to enable energy preservation in BiHCLR, the fuzzy logic approach is executed to select the Cluster Head (CH) for every cell of the grid. Then a hybrid bio-inspired algorithm is used to select the routing path. The hybrid algorithm combines moth search and Salp Swarm optimization techniques. The performance of the proposed BiHCLR is evaluated based on the Quality of Service (QoS) analysis in terms of Packet loss, error bit rate, transmission delay, lifetime of network, buffer occupancy and throughput. Then these performances are validated based on comparison with conventional routing strategies like Fuzzy-rule-based Energy Efficient Clustering and Immune-Inspired Routing (FEEC-IIR), Neuro-Fuzzy- Emperor Penguin Optimization (NF-EPO), Fuzzy Reinforcement Learning-based Data Gathering (FRLDG) and Hierarchical Energy Efficient Data gathering (HEED). Ultimately the performance of the proposed BiHCLR outperforms all other conventional techniques.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.6
• /
• pp.714-721
• /
• 2020

DNN is expanding its use in embedded systems such as robots and autonomous vehicles. For high recognition accuracy, computational complexity is greatly increased, and multiple DNNs are running aperiodically. Therefore, the ability processing multiple DNNs in embedded environments is a crucial issue. Accordingly, multicore based platforms are being released. However, most DNN models are operated in a batch process, and when multiple DNNs are operated in multicore together, the execution time deviation between each DNN may be large and the end-to-end execution time of the whole DNNs could be long depending on how they are allocated to the cores. In this paper, we solve these problems by providing a framework that decompose each DNN into individual layers and then distribute to multicores through a global queue. As a result of the experiment, the total DNN execution time was reduced by 31%, and when operating multiple identical DNNs, the deviation in execution time was reduced by up to 95.1%.