• The Journal of the Acoustical Society of Korea
• /
• v.20 no.2
• /
• pp.27-34
• /
• 2001

Utterance verification is used in variable vocabulary word recognition to reject the word that does not belong to in-vocabulary word or does not belong to correctly recognized word. Utterance verification is an important technology to design a user-friendly speech recognition system. We propose a new utterance verification algorithm for no-training utterance verification system based on the minimum verification error. First, using PBW (Phonetically Balanced Words) DB (445 words), we create no-training anti-phoneme models which include many PLUs(Phoneme Like Units), so anti-phoneme models have the minimum verification error. Then, for OOV (Out-Of-Vocabulary) rejection, the phoneme-based confidence measure which uses the likelihood between phoneme model (null hypothesis) and anti-phoneme model (alternative hypothesis) is normalized by null hypothesis, so the phoneme-based confidence measure tends to be more robust to OOV rejection. And, the word-based confidence measure which uses the phoneme-based confidence measure has been shown to provide improved detection of near-misses in speech recognition as well as better discrimination between in-vocabularys and OOVs. Using our proposed anti-model and confidence measure, we achieve significant performance improvement; CA (Correctly Accept for In-Vocabulary) is about 89％, and CR (Correctly Reject for OOV) is about 90％, improving about 15-21％ in ERR (Error Reduction Rate).

• The korean journal of orthodontics
• /
• v.46 no.1
• /
• pp.3-12
• /
• 2016

Objective: The purpose of this study was to compare the precision of three-dimensional (3D) images acquired using iTero$^{(R)}$(Align Technology Inc., San Jose, CA, USA) and Trios$^{(R)}$(3Shape Dental Systems, Copenhagen, Denmark) digital intraoral scanners, and to evaluate the effects of the severity of tooth irregularities and scanning sequence on precision. Methods: Dental arch models were fabricated with differing degrees of tooth irregularity and divided into 2 groups based on scanning sequence. To assess their precision, images were superimposed and an optimized superimposition algorithm was employed to measure any 3D deviation. The t-test, paired t-test, and one-way ANOVA were performed (p < 0.05) for statistical analysis. Results: The iTero$^{(R)}$ and Trios$^{(R)}$ systems showed no statistically significant difference in precision among models with differing degrees of tooth irregularity. However, there were statistically significant differences in the precision of the 2 scanners when the starting points of scanning were different. The iTero$^{(R)}$ scanner (mean deviation, $29.84{\pm}12.08{\mu}m$) proved to be less precise than the Trios$^{(R)}$ scanner ($22.17{\pm}4.47{\mu}m$). Conclusions: The precision of 3D images differed according to the degree of tooth irregularity, scanning sequence, and scanner type. However, from a clinical standpoint, both scanners were highly accurate regardless of the degree of tooth irregularity.

• Journal of IKEEE
• /
• v.10 no.1 s.18
• /
• pp.75-86
• /
• 2006

The MANET that can make autonomous distributed Network with Routing function has many differences than past wireless communication. For upcoming ALL-IP environment, MANET device should be connected with wired Internet Network and MANET is required to have a gateway to bridge two different networks to share information from any place. In this paper, Using the GMAHN Algorithm proposed Proactive, Reactive, Hybrid method that provides Inteface between Wired Internet network and MANET, we learned each method's the advantage and disadvantage through the various network environments. And also, we presented the optimization method of Hybrid combined Proactive with Reactive.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.29 no.10
• /
• pp.791-798
• /
• 2018

The CFAR algorithm is mainly used for target detection in radar systems. In particular, OS CFAR is used in a non-uniform noise environment. However, it requires a large amount of computation, because it should sort reference cells in ascending order. This makes it difficult to apply the radar system in real time. In this paper, we describe how to reduce the computational burden of OS CFAR. We compared the power of the test cell and reference cell to determine only the presence or absence of target detection. The common reference cells overlapping in the reference cells of the three test cells are obtained. We first compare the test cell with the highest power value among the three test cells to the common reference cells. Next, we compare each test cell to general reference cells, excluding the common reference cells. The computation time is shortened by reducing the power comparison computation amounts.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.7
• /
• pp.2972-2991
• /
• 2016

WLAN is the best choice in the place where complex network is hard to set up. Intelligent terminals are more and more assembled in some areas now. However, according to IEEE 802.11n/802.11ac, the access-point (AP) can only serve one user at a single frequency channel. The spectrum efficiency urgently needs to be improved. In theory, AP with multi-antenna can serve multiple users if these users do not interfere with each other. In this paper, we propose a user clustering scheme that could achieve multi-user selection through the mutual cooperation among users. We focus on two points, one is to achieve multi-user communication with multiple antennas technique at a single frequency channel, and the other one is to use a way of distributed users' collaboration to determine the multi-user selection for user clustering. Firstly, we use the CSMA/CA protocol to select the first user, and then we set this user as a source node using users' cooperation to search other proper users. With the help of the users' broadcast cooperation, we can search and select other appropriate user (while the number of access users is limited by the number of antennas in AP) to access AP with the first user simultaneously. In the network node searching, we propose a maximum degree energy routing searching algorithm, which uses the shortest time and traverses as many users as possible. We carried out the necessary analysis and simulation to prove the feasibility of the scheme. We hope this work may provide a new idea for the solution of the multiple access problem.

• Computers and Concrete
• /
• v.13 no.6
• /
• pp.739-748
• /
• 2014

In this article, by using experimental studies and artificial neural network has been tried to investigate the use of nano-silica as concrete admixture to reduce alkali-silica reaction. If there are reactive aggregates and alkali of cement with enough moisture in concrete, a gel will be formed. Then with high reactivity between alkali of cement and existence of silica in aggregates, this gel will expand by absorption of water, and causes expansive pressure and cracks be formed. At the time passes, this gel will reduce both durability and strength of the concrete. By reducing the size of silicate to nano, specific surface area of particles and number of atoms on the surface will be increased, which causes more pozzolanic activity of them. Nano-silica can react with calcium hydroxide ($Ca(OH)_2$) and produces C-S-H gel. In this study, accelerated mortar bar specimens according to ASTM C 1260 and ASTM C 1567, with different mix proportions were prepared using aggregates of Kerman, such as: none admixture and plasticizer, different proportions of nano-silica separately. By opening the moulds after 24 hour and curing in water at $80^{\circ}C$ for 24 hour, then curing in (1N NaOH) at $80^{\circ}C$ for 14 days, length expansion of mortar bars were measured and compared. It was noted that, the lowest length expansion of a specimens shows the best proportion of admixture based on alkali-silica reactivity. Then, prediction of alkali-silica reaction of concrete has been investigated by using artificial neural network. In this study the backpropagation network has been used and compared with different algorithms to train network. Finally, the best amount of nano silica for adding to mix proportion, also the best algorithm and number of neurons in hidden layer of artificial neural network have been offered.

• Nuclear Engineering and Technology
• /
• v.53 no.10
• /
• pp.3275-3285
• /
• 2021

A Nuclear Power Plant (NPP) is a complex dynamic system-of-systems with highly nonlinear behaviors. In order to control the plant operation under both normal and abnormal conditions, the different systems in NPPs (e.g., the reactor core components, primary and secondary coolant systems) are usually monitored continuously, resulting in very large amounts of data. This situation makes it possible to integrate relevant qualitative and quantitative knowledge with artificial intelligence techniques to provide faster and more accurate behavior predictions, leading to more rapid decisions, based on actual NPP operation data. Data-driven models (DDM) rely on artificial intelligence to learn autonomously based on patterns in data, and they represent alternatives to physics-based models that typically require significant computational resources and might not fully represent the actual operation conditions of an NPP. In this study, a feed-forward backpropagation artificial neural network (ANN) model was trained to simulate the interaction between the reactor core and the primary and secondary coolant systems in a pressurized water reactor. The transients used for model training included perturbations in reactivity, steam valve coefficient, reactor core inlet temperature, and steam generator inlet temperature. Uncertainties of the plant physical parameters and operating conditions were also incorporated in these transients. Eight training functions were adopted during the training stage to develop the most efficient network. The developed ANN model predictions were subsequently tested successfully considering different new transients. Overall, through prompt prediction of NPP behavior under different transients, the study aims at demonstrating the potential of artificial intelligence to empower rapid emergency response planning and risk mitigation strategies.

• Progress in Medical Physics
• /
• v.31 no.4
• /
• pp.163-171
• /
• 2020

Purpose: In this study, the accuracies of electron Monte Carlo (eMC) calculation algorithms were evaluated to determine whether electron beams were modeled by optional air profiles (APs) designed for each applicator size. Methods: Electron beams with the energies of 6, 9, 12, and 16 MeV for VitalBeam (Varian Medical System, Palo Alto, CA, USA) and 6, 9, 12, 16, and 20 MeV for Clinac iX (Varian Medical System) were used. Optional APs were measured at the source-to-detector distance of 95 cm with jaw openings appropriate for each machine, electron beam energy, and applicator size. The measured optional APs were postprocessed and converted into the w2CAD format. Then, the electron beams were modeled and calculated with and without optional APs. Measured profiles, percentage depth doses, penumbras with respect to each machine, and energy were compared to calculated dose distributions. Results: For VitalBeam, the profile differences between the measurement and calculation were reduced by 0.35%, 0.15%, 0.14%, and 0.38% at 6, 9, 12, and 16 MeV, respectively, when the beams were modeled with APs. For Clinac iX, the differences were decreased by 0.16%, -0.31%, 0.94%, 0.42%, and 0.74%, at 6, 9, 12, 16, and 20 MeV, respectively, with the insertion of APs. Of note, no significant improvements in penumbra and percentage depth dose were observed, although the beam models were configured with APs. Conclusions: The accuracy of the eMC calculation can be improved in profiles when electron beams are modeled with optional APs.

• The KIPS Transactions:PartC
• /
• v.14C no.4
• /
• pp.331-340
• /
• 2007

Mobile network environments are the environments where mobile devices are distributed invisible in our daily lives so that we can conventionally use mobile services at my time and place. The fact that we can work with mobile devices regardless of time and place, however, means that we are also in security threat of leaking or forging the information. In particular, without solving privacy concern, the mobile network environments which serve convenience to use, harmonized without daily lives, on the contrary, will cause a serious malfunction of establishing mobile network surveillance infrastructure. On the other hand, as the mobile devices with various sizes and figures, public key cryptography techniques requiring heavy computation are difficult to be applied to the computational constrained mobile devices. In this paper, we propose efficient PKI-based user authentication and java-based cryptography module for the privacy-preserving in mobile network environments. Proposed system is support a authentication and digital signature to minimize encrypting and decrypting operation by compounding session key and public key based on Korean standard cryptography algorithm(SEED, KCDSA, HAS160) and certificate in mobile network environment. Also, it has been found that session key distribution and user authentication is safety done on PDA.

• The Journal of Korean Society for Radiation Therapy
• /
• v.26 no.2
• /
• pp.171-176
• /
• 2014

Purpose : Changing the calculation grid of AAA in Lung SABR plan and to analyze the changes in target dose, and investigated the effects associated with it, and considered a suitable method of application. Materials and Methods : 4D CT image that was used to plan all been taken with Brilliance Big Bore CT (Philips, Netherlands) and in Lung SABR plan($Eclipse^{TM}$ ver10.0.42, Varian, the USA), use anisotropic analytic algorithm(AAA, ver.10, Varian Medical Systems, Palo Alto, CA, USA) and, was calculated by the calculation grid 1.0, 3.0, 5.0 mm in each Lung SABR plan. Results : Lung SABR plan of 10 cases are using each of 1.0 mm, 3.0 mm, 5.0 mm calculation grid, and in case of use a 1.0 mm calculation grid $V_{98}$. of the prescribed dose is about $99.5%{\pm}1.5%$, $D_{min}$ of the prescribed dose is about $92.5{\pm}1.5%$ and Homogeneity Index(HI) is $1.0489{\pm}0.0025$. In the case of use a 3.0 mm calculation grid $V_{98}$ dose of the prescribed dose is about $90{\pm}4.5%$, $D_{min}$ of the prescribed dose is about $87.5{\pm}3%$ and HI is about $1.07{\pm}1$. In the case of use a 5.0 mm calculation grid $V_{98}$ dose of the prescribed dose is about $63{\pm}15%$, $D_{min}$ of the prescribed dose is about $83{\pm}4%$ and HI is about $1.13{\pm}0.2$, respectively. Conclusion : The calculation grid of 1.0 mm is better improves the accuracy of dose calculation than using 3.0 mm and 5.0 mm, although calculation times increase in the case of smaller PTV relatively. As lung, spread relatively large and low density and small PTV, it is considered and good to use a calculation grid of 1.0 mm.